Division of Research
& Innovation
Report CA08- 0279
December 2008
Intelligent Herbicide Application System
For Reduced Herbicide Vegetation
Control Phase II – Commercialization
Final Report
Intelligent Herbicide Application System for
Reduced Herbicide Vegetation Control
Phase II – Commercialization
Final Report
Report No. CA08- 0279
December 2008
Prepared By:
Biological and Agricultural Engineering
University of California, Davis
Davis, CA 95616
Prepared For:
Advanced Highway Maintenance and Construction Technology ( AHMCT) Center
University of California, Davis
California Department of Transportation
Division of Research and Innovation, MS- 83
1227 O Street
Sacramento, CA 95814
DISCLAIMER STATEMENT
This document is disseminated in the interest of information exchange. The contents of this report
reflect the views of the authors who are responsible for the facts and accuracy of the data presented
herein. The contents do not necessarily reflect the official views or policies of the State of California
or the Federal Highway Administration. This publication does not constitute a standard,
specification or regulation. This report does not constitute an endorsement by the Department of
any product described herein.
California AHMCT Program
University of California, Davis
California Department of Transportation
Intelligent Herbicide Application System for
Reduced Herbicide Vegetation Control
Phase II – Commercialization
D. C. Slaughter, D. K. Giles, D. Downey,
K. Gillis, R. Zarghami, E. Staab, F. Vanucci,
M. Shaffi, C. Gliever, P. Fontes and J. Schlottman
Biological and Agricultural Engineering
University of California, Davis
AHMCT Research Report
UCD- ARR- 06- 06- 30- 10
Final Report for Contract
65A0049/ 65A0139, T. O. 02- 22
June 2006
This work was supported by the Division of Research and Innovation of the California
Department of Transportation ( Caltrans) and the Advanced Highway Maintenance and
Construction Technology ( AHMCT) Center at the University of California, Davis.
i
Technical Report Documentation Page ( Form DOT F 1700.7)
1. Report No.
CA08- 0279
2. Government Accession No. 3. Recipient's Catalog No.
4. Title and Subtitle
INTELLIGENT HERBICIDE APPLICATION SYSTEM FOR
5. Report Date
June 2006
REDUCED HERBICIDE VEGETATION CONTROL
PHASE II – COMMERCIALIZATION
6. Performing Organization Code
7. Author( s)
D. C. Slaughter, D. K. Giles, D. Downey, K. Gillis, R. Zarghami, E.
Staab, F. Vanucci, M. Shaffi, C. Gliever, P. Fontes and J. Schlottman
8. Performing Organization Report No.
UCD- ARR- 06- 06- 30- 10
9. Performing Organization Name and Address
Biological and Ag. Engineering
10. Work Unit No. ( TRAIS)
University of California, Davis
Davis, CA 95616
11. Contract or Grant No.
65A0049/ 65A0139, T. O. 02- 22
12. Sponsoring Agency Name and Address
California Department of Transportation
13. Type of Report and Period Covered
Final Report 5/ 02 to 6/ 06
Sacramento, CA 95819 14. Sponsoring Agency Code
15. Supplementary Notes
16. Abstract
This report describes the development of a commercial prototype intelligent herbicide application system
( IHAS). The improved design incorporates a parallel “ add- on” type fluid handling system to allow existing
variable- rate herbicide injection systems currently used on Caltrans herbicide spray vehicles to be retrofitted
with IHAS technology, and a dual camera system for weed recognition under partially shaded lighting
conditions. The new IHAS is capable of targeting green plant material within a 3.66 m perpendicular distance
to the direction of travel on either side of the herbicide spray vehicle for herbicide application.
The basic principal of an intelligent herbicide application system ( IHAS) is that a real- time machine vision
system can detect live ( green) plant material growing along the roadside, and when coupled to a rapid- response
spray control system, will permit the California Department of Transportation to selectively apply
post- emergence herbicides exclusively to the unwanted plant material. The implementation of the IHAS
technology will allow the California Department of Transportation to reduce the amount of resources required to
maintain an effective weed control program using herbicides while at the same time reducing the amount of
chemicals unnecessarily released into the environment.
17. Key Words
Vegetation Control, Herbicide Injection, Machine
Vision, Weeds.
18. Distribution Statement
No restrictions. This document is available to the
public through the National Technical Information
Service, Springfield, VA 22161
20. Security Classif. ( of this report)
Unclassified
20. Security Classif. ( of this page)
Unclassified
21. No. of Pages
96
22. Price
Form DOT F 1700.7 ( 8- 72) Reproduction of completed page authorized
ii
Table of Contents
Technical Report Documentation Page ( Form DOT F 1700.7) ..................................................... i
List of Figures........................................................................................................................ .. iii
List of Tables......................................................................................................................... ... iv
Disclosure Statement................................................................................................................... v
Disclaimer Statement .................................................................................................................. v
Introduction ............................................................................................................................... 1
Objective ............................................................................................................................... .... 4
Intelligent Herbicide Application System ( IHAS) ....................................................................... 4
Valve Control System............................................................................................................. 5
Operator Controls................................................................................................................ 6
Fluid Handling System............................................................................................................ 8
Plumbing Modifications.................................................................................................... 11
Weed Mapping System ......................................................................................................... 13
Camera resolution tests...................................................................................................... 14
Spray deposition tests........................................................................................................ 15
Data Logging System............................................................................................................ 19
Measurement Verification. ................................................................................................ 20
Conclusions .............................................................................................................................. 27
Appendix A: Adaptive Equipment VCS Manual ....................................................................... 28
Appendix B: VCS Schematics from Adaptive Equipment ......................................................... 36
Appendix C: IHAS User Manual............................................................................................... 53
Appendix D: Data Logger User Manual .................................................................................... 65
Appendix E: Sensor Specifications............................................................................................ 81
Appendix F: IHAS Direct Nozzle Injection Research Study...................................................... 87
iii
List of Figures
Figure 1. Illustration of the precision offset spray concept developed by Slaughter et al. 3............ 3
Figure 2. Passenger side of the IHAS vehicle showing locations of several subsystems for
targeted herbicide applications................................................................................... 5
Figure 3. IHAS main enclosure; general schematic showing component layout ( left) and
enclosure photograph ( right). ..................................................................................... 7
Figure 4. Driver view inside cab of spray vehicle containing IHAS operator interface with
touch- panel ( left), emergency stop switch ( above touch- panel) and global positioning
system ( right) installed adjacent to conventional spray control systems...................... 8
Figure 5. IHAS touchscreen interface as seen by vehicle operator with several options shown:
a) default ( main screen) display when vehicle is stationary, b) system status screen, c)
spray operation screen, automatically displayed when vehicle is in motion and
spraying, d) nozzle control screen. ............................................................................. 9
Figure 6. IHAS spray tower with supply and return manifolds. a) single nozzle tower showing
six spray valves/ nozzles with supply manifold on left and return manifold behind
valves/ nozzles. b) All four nozzle towers on one side of the vehicle. c) three- way
valve with spray nozzle............................................................................................ 10
Figure 7. Schematic of the IHAS fluid bypass system used to minimize fluid pressure and
herbicide concentration variations under variable spray demands ( note IHAS
additions in blue were added to the existing fluid handling system on the vehicle). .. 11
Figure 8. Camera mounting system for IHAS; the left image shows the camera mounting frame
with four camera enclosures and the right image shows one camera enclosure with the
front cover removed revealing the sun camera ( top) and shadow camera ( bottom). .. 14
Figure 9. Grid set- up for spray deposition showing replicate and speed of vehicle for the tests. 16
Figure 10. DLS instrumentation: ( a) DGPS antennae, wind sensor and ultrasonic sensor, ( b)
Legacy interface screen............................................................................................ 21
Figure 11. DGPS wind map from field verification tests............................................................ 22
Figure 12. Temperature map of morning to mid- day.................................................................. 23
Figure 13. Plot of position offset in the direction of travel of GPS map data from actual spray
deposition versus vehicle travel speed...................................................................... 26
iv
List of Tables
Table 1. Camera resolution tests; results are indicative of size of material recognized by cameras
for spray application. .................................................................................................. 15
Table 2. Weather conditions for spray deposition tests; data were obtained from the California
irrigation management information system ( CIMIS) website for the UC Davis campus.
............................................................................................................................... ... 16
Table 3. Broadcast depositions as a percentage of spray tank mix ( 19.8 ppm) for two test speeds
and range of actual spray depositions concentrations ( ppb) for all targets.................... 16
Table 4. Random target deposition results as a percentage of tank concentration ( 19.8 ppm). ... 17
Table 5. Random target deposition averages for two test speeds; deposition for each replicate
target location was averaged and normalized to average broadcast deposition ( all 24
targets) and spray tank mix concentration ( 19.89 ppm). .............................................. 18
Table 6. Absolute wind speed and direction. ............................................................................. 21
Table 7. Ambient temperature on spray vehicle vs. stationary thermometer............................... 22
Table 8. Measured roadside slope versus actual slope................................................................ 24
Table 9. Geo- referenced offset between valve triggers and actual spray deposition. .................. 25
v
Disclosure Statement
The California Department of Transportation and the FHWA reserve a royalty- free,
non- exclusive and irrevocable license to reproduce, publish or otherwise use, and to authorize
others to use, this work for government purposes.
Disclaimer Statement
The research reported herein was performed as part of the Advanced Highway Maintenance and
Construction Technology ( AHMCT) Program, at the University of California, Davis and the
Division of Research and Innovation of the California Department of Transportation.
The contents of this report reflect the views of the author( s) who is ( are) responsible for the facts
and the accuracy of the data presented herein. The contents do not necessarily reflect the official
views or policies of the STATE OF CALIFORNIA or the FEDERAL HIGHWAY
ADMINISTRATION or the UNIVERSITY OF CALIFORNIA. This report does not constitute a
standard, specification, or regulation.
1
Introduction
The California Department of Transportation expends a considerable amount of human and
financial resources in its highway maintenance program for the control of vegetation along the
shoulders of highways. Effective weed control has multiple benefits including reduced fire
hazard, increased visibility and safety for drivers, reduced loss of natural resources ( e. g., water)
to unwanted vegetation and a reduction in alternative hosts for insect pests and diseases.
Application of herbicides is one of the more efficacious and economical methods of weed
control.
A major issue in California is the current reliance on chemical methods of pest control. Helsel1
estimated that in 1984 16 billion dollars were spent worldwide on pesticides. Further, Helsel
reported the United States as the largest pesticide user in the world applying more than three
times the quantity of pesticides as the second largest user ( Japan). A total of 4.5 billion pounds
( 2 billion kg) of chemical pesticides were used in 1995 in the US2. Unfortunately, the continued
reliance on chemically based pest control practices has potentially detrimental effects upon the
environment and human health in the form of contamination of water supplies and soils. In
addition, the effect of chemical residues is often cumulative and their continued use can be
increasingly detrimental to the environment.
There is a need to develop improved means of weed control in a reduced herbicide environment.
Slaughter et al. 3 and others have demonstrated that one possible solution is to reduce herbicide
requirements by targeting post- emergence herbicide sprays only to plant vegetation and not the
surrounding soil. If the California Department of Transportation applied herbicides only to
targeted plant material, the energy and material costs involved in weed control could be reduced
and, subsequently, the amount of herbicides released into the environment greatly reduced.
Additionally, the productivity of the weed control operation could be increased, allowing more
reliable control.
The concept of intermittent spray control for plant sprayers has been previously investigated.
Reichard and Ladd4 discussed work in which plant conductivity and a charged probe were used
to detect the presence of target plants. They also developed intermittent spray control systems
that detected vegetable plants through steel wires and systems based on photo- detectors. The
detection systems were limited to targeted plants that could fit between the sensor system based
on operational conditions of tripping a wire switch or interruption of a light beam. Field tests of
the control systems ( Ladd et al. 5 and Ladd and Reichard6) found reduction of applied spray
1 Helsel, Z. R. 1987. Pesticide use in world agriculture. In: Z. R. Helsel ( Ed.), Energy in plant nutrition and pest control.
Elsevier, New York, p. 194.
2 Aspelin, A. L. 1997. Pesticides Industry Sales and Usage, 1994 and 1995 Market Estimates. Biological and Economic
Analysis Division, Office of Pesticide Programs, Office of Prevention, Pesticides and Toxic Substances, US Environmental
Protection Agency, Washington, DC.
3 Slaughter, D. C., D. K. Giles, and C. Tauzer. 1999. Precision offset spray system for roadway shoulder weed control. ASCE
Journal of Transportation Engineering. 125( 4): 364- 371.
4 Reichard, D. L. and T. L. Ladd. 1981. An automatic intermittent sprayer. Trans. of the ASAE. Vol. 24( 4): 893- 896.
5 Ladd, T. L., D. L. Reichard, D. L. Collins and C. R. Buriff. 1978. An automatic intermittent sprayer: a new approach to the
insecticidal control of horticultural pests. J. Econ. Entomology. Vol. 71: 789- 792.
2
material ranged from 24 to 51% with little or no reduction in pest control efficacy. Giles et al. 7,8
developed a spray application system that used ultrasonic sensors to trigger spray nozzles on or
off based on tree presence or absence. The system was physically limited to large targets
typically found in orchards.
Several researchers ( e. g. Hollaender9) have documented distinct absorption characteristics of
chlorophyll; detectable maximum peaks occur at 675 nm region of the visible spectrum. Other
researchers have attempted to use this information to develop a non- contact sensor for detecting
chlorophyll- containing materials ( e. g., plants) versus non- chlorophyll containing materials ( e. g.,
soils). Generally, in those studies, the difference in reflectance between plants and the soil
background is based on foliage chlorophyll absorbing red radiation while soil reflects red
radiation. Additionally, in other earlier studies the ratio of visible to near- infrared radiation10 and
the ratio of red to near- infrared radiation11 were used to distinguish green vegetation from the
soil background. Several of these studies have led to commercial plant detector- sprayers ( e. g.,
Weed Seeker PhD 1620, Patchen California, Inc., Los Gatos, CA and Detectspray- S45, Concord
Inc., Fargo, ND).
Systems based on discrete reflectance sensors, such as those described above, are limited to
relatively short operating ranges due to averaging of the signals of all of the objects in the field
of view, including the plant and the background levels. As sensor height is increased, the plant
detection resolution decreases. Merritt et al12 reported that plants greater than 20 cm2 were
detectable using a sensor height of 23 cm. While this sensor range is allowable for many boom
spraying applications, it is not useful for detecting smaller plants from boomless systems where
sensor to target distances can exceed 3 m. Alternatively, computer vision systems, having
greater resolution, can provide the plant detection performance required for detecting smaller
plants.
Slaughter et al. 3 developed a machine vision based a precision offset spray system ( figure 1) for
control of vegetation along roadsides. In this system spray material is delivered to the region
adjacent to the vehicle in which the plant lies and not to surrounding soil. The system used a
single camera, mounted approximately 2 m above the road and was capable of detecting weeds
as small as 6.25 cm2. The system substantially reduced the amount of herbicide applied to non-plant
material ( up to 97% reduction in applied pesticide compared to conventional continuous
6 Ladd, T. L., and D. L. Reichard. 1980. Photoelectrically- operated intermittent sprayers for insecticidal control of horticultural
pests. J. Econ. Entomology. Vol. 73: 525- 528.
7 Giles, D. K., M. J. Delwiche and R. B. Dodd. 1988. Electronic measurement of tree canopy volume. Trans. of the ASAE Vol.
31( 1): 264- 272.
8 Giles, D. K., M. J. Delwiche and R. B. Dodd. 1987. Control of orchard spraying based on electronic sensing of target
characteristics. . Trans. of the ASAE Vol. 30( 6): 1624- 1630, 1636.
9 Hollaender, A. 1956. Radiation biology. Vol. III. McGraw- Hill. New York.
10 Hooper, A. W., G. O. Harries, and B. Ambler. 1976. A photoelectric sensor for distinguishing between plant material and
soil. J. Agric. Engng. Res. Vol. 21.
11 Haggar, R. L. C. J. Stent, and S. Isaac. 1983. A prototype hand- held patch sprayer for killing weeds, activated by spectral
differences in crop/ weed canopies. J. Agric. Engng. Res. Vol. 28.
12 Merritt, S. J., G. E. Meyer, K. Von Bargen, D. A. Mortensen. 1994. Reflectance sensor and control system for spot spraying.
ASAE Paper No. 94- 1057. ASAE, 1950 Niles Rd., St. Joseph, MI, USA.
3
Figure 1. Illustration of the precision offset spray concept developed by Slaughter et al. 3
spray applications). This study indicated the value of machine vision technology for reducing
both environmental and economic costs of weed control. Additionally, the concept of “ offset" or
“ boomless” spray applications ( i. e., no spray boom extending beyond the vehicle boundary) was
shown to be technically feasible.
The research prototype system developed by Slaughter et al. 3 did not address issues of uneven
and unlevel terrain, or varying light conditions. Also, the influence of wind on spray accuracy
was corrected by applying additional herbicide prior to and after the intended target to ensure
herbicide hit the target when wind deflected the spray in flight.
The basic principal of an improved precision offset spray system is that a real- time machine
vision system can detect live ( green) plant material growing along the roadside and, when
coupled to a rapid- response spray control system, will permit the California Department of
Transportation to selectively apply post- emergence herbicides exclusively to the unwanted plant
material. The implementation of this technology will allow the California Department of
Transportation to reduce the amount of resources required to maintain an effective weed control
program using herbicides while at the same time reducing the amount of chemicals unnecessarily
released into the environment.
4
Objective
The technical objective of this research was to develop an intelligent herbicide application
system ( IHAS); that is, a spray application vehicle that uses machine vision system to detect
plant material along a roadway shoulder and specifically target plant material with herbicide in
real- time. Herbicide is delivered to the plant region, not adjacent soil, based on automated
commands from the IHAS.
The improved IHAS was capable of:
1) Targeting green plant material within a 3.66 m perpendicular distance to the direction of
travel on either side of the herbicide spray vehicle.
2) Working in parallel to the existing pre- emergence herbicide application system to allow
Caltrans to operate the vehicle in either the non- IHAS mode ( conventional spray
applications) or in the IHAS mode ( targeted herbicide application).
3) Improved detection of green plant material under non- uniform illumination conditions ( e. g.,
shadows caused by roadside signs) typical of naturally illuminated roadsides in California.
4) Having the potential for commercial manufacture; the IHAS project involved the cooperative
effort of researchers at the University of California, Davis and design engineers at Adaptive
Equipment, Inc., with the ultimate objective of producing a commercial prototype IHAS.
5) Recording spray events and ambient conditions of wind speed and direction, air temperature,
average slope of ground, and vehicle location in the form of a DGPS ( Differential Global
Positioning System) “ as applied” spray map. The system has a “ lock- out” capability, where
spray valve actuation is disabled if any of the monitored ambient parameters are outside
specified or defined limits. Additionally, the system can warn the vehicle operator if the
vehicle is located inside a previously identified prohibited spray zone. The IHAS uses
commercially available sensors, controllers and other components, with custom interfaces as
needed, for robust spray control of roadside vegetation.
Intelligent Herbicide Application System ( IHAS)
The IHAS ( figure 2 shows the passenger side of the vehicle) consists of three subsystems:
1) Valve Control System ( VCS)
2) Weed Mapping System ( WMS)
3) DGPS Data Logging System ( DLS)
Briefly, the VCS is responsible for setting the parameters used for spraying. That is, this portion
of the IHAS accepts input and displays information to the vehicle operator. The WMS uses
machine vision to develop a spray application map with concurrent one- to- one mapping to spray
valves for targeted herbicide application. The DLS archives ambient conditions in addition to
location application information for later retrieval and assessment. An in- depth discussion of
these subsystems follows.
5
Figure 2. Passenger side of the IHAS vehicle showing locations of several subsystems for
targeted herbicide applications.
Valve Control System
The VCS is the primary controller component of the IHAS. It is responsible for accepting
operator input, displaying information to the operator, coordinating the image acquisition process
and operating the spray valves based upon the digital video spray map information. The VCS
was originally developed in an earlier phase of the project. Changes to the VCS in this phase of
the project included an expanded spray coverage up to 3.66 m perpendicular to the truck foot
print on both sides of the vehicle ( i. e., PLC processing speed was increased and additional
control hardware added for the additional spray valves). A general overview of the VCS is
provided here; additional details on the VCS system can be found in Appendix A.
A programmable logic controller ( PLC) acts as the central administrator of the VCS. The PLC
monitors the ground speed radar sensor ( which is shared between the existing Raven spray
controller and the IHAS) to determine vehicle speed and displacement and uses this information
to control valve actuation and timing signals for 24 valves specific to the side of the road being
sprayed. Spray application, or weed map, data is transferred from the WMS to the PLC via an
Ethernet connection and is stored until the vehicle has traveled to the appropriate location for
valve actuation to spray the identified targets in the weed map. The PLC uses valve activation
delay timing, spray time- of- flight, physical distance ( in the direction of travel) between the
camera and the nozzle, and vehicle speed to determine valve actuation times.
IHAS spray valves Conventional
spray boom
IHAS cameras
Generator for powering
IHAS control system
GPS antennae and
wind speed sensor
6
The commercial prototype VCS ( Adaptive Equipment, Gainesville, FL) developed for IHAS is
shown in Figure 3 and consists of the following hardware and software components:
1) A NEMA 12 enclosure mounted to the bed of the Caltrans spray vehicle;
2) A gasoline powered generator mounted to the bed of the Caltrans spray vehicle to
generate 120VAC for all IHAS components except IHAS spray valves;
3) A PLC with non- volatile memory;
4) An in- cab touch- screen operator interface to allow for initial pray parameter set- up,
monitoring of system parameters and capable of tuning system parameters;
5) Wiring and connectors for the PLC- valve interface and other external components
( e. g. radar, PC, power source monitor, operator interface, etc.); and
6) Software and hardware required for communication between the VCS, WMS, and
DLS.
All wiring diagrams, component information, and layout diagrams associated with the VCS can
be found in Appendix B.
Most of the electronic components associated with the VCS ( e. g., PLC, DC power supplies,
valve relays, etc.) were housed in the NEMA 12 rain- tight enclosure mounted on vibration
resistant shocks at the rear of the vehicle deck on the driver’s side of the vehicle. Additional
NEMA 12 enclosures were used for various devices and connectivity. Valve relays were housed
in an enclosure mounted at the base of the VCS main cabinet on the deck. An enclosure resides
at the front of the vehicle deck to select the specific side of the vehicle for spray applications. A
monitor and keyboard for communicating with the weed map computer ( machine vision
computer) were enclosed and mounted approximately mid- deck on driver’s side of the vehicle.
An additional enclosure was mounted near the monitor enclosure and contained an electrical
input connection for a function generator used to simulate radar pulses of vehicle motion and
allowed system diagnostic tests while the truck was stationary. Power for the VCS and
additional components was provided by a gasoline- powered generator mounted on the passenger
side of the vehicle.
Operator Controls. Communication with the VCS by the spray vehicle operator is through a
touch panel interface located in the cab of the vehicle ( figure 4). The interface includes an
emergency stop switch above the touch panel display. Both are located in the cab and can be
easily accessed by the operator.
The touch panel ( PanelView Plus 1000) allows the vehicle operator to input specific conditions.
Figure 5 shows several screen shots of the touch panel interface: system start- up menu, system
status menu, system run- time menu and spray nozzle selection options. For example, the vehicle
operator can select specific spray nozzles ( or all spray nozzles) to continuously spray ( manual
mode) or automatically spray based on the WMS communication with the VCS. The vehicle
operator can set time of flights for the spray valves based on their location and spacing
configurations for the nozzle booms. Additionally there is a run- time menu that gives the vehicle
operator immediate feedback on vehicle ground speed and spray flow rate through the system.
Detailed information on instructions for the vehicle operator interfacing with touch panel and
7
Legend of selected components ( see Appendix B for additional details)
1, 2 & 4: DC Power supplies & conditioning
3: Programmable logic controller ( PLC) 5, 9: Relays
6: Video signal bulkhead 7: Machine Vision Computer
16 - 19: Power circuit breakers 11 & 12: Exhaust vent rain shields
Figure 3. IHAS main enclosure; general schematic showing component layout ( left) and enclosure photograph ( right).
8
Figure 4. Driver view inside cab of spray vehicle containing IHAS operator interface with
touch- panel ( left), emergency stop switch ( above touch- panel) and global positioning
system ( right) installed adjacent to conventional spray control systems.
setting up initial operating conditions in addition to a pre- spray check list with start- up and shut
down procedures are given in Appendix C.
Fluid Handling System
One goal of this project was to adapt the fluid handling system of the existing conventional direct
injection spray system to work with the IHAS machine vision- activated spray system. The main
design constraint was that the IHAS was to be an " add- on" system placed in parallel with the
existing Caltrans pre- emergence herbicide application system. Specifically, this parallel design
will allow Caltrans to operate the modified herbicide vehicle in either the normal conventional
( non- IHAS) mode or in the IHAS mode. Note that both modes can not be operated
simultaneously.
Unlike conventional spray systems, a fundamental principal of the IHAS is that a variable
number of spray valves can be opened and closed every 15 cm ( 6 inches) of travel. This distance
is a non- adjustable defined parameter for the IHAS system and is the basic spray dimension unit.
9
( a)
( b)
( c)
( d)
Figure 5. IHAS touch- screen interface as seen by vehicle operator with several options shown:
a) default ( main screen) display when vehicle is stationary, b) system status screen, c)
spray operation screen, automatically displayed when vehicle is in motion and
spraying, d) nozzle control screen.
For example, if the vehicle is traveling at 16 kph ( 10 mph), a variable number of valves ( up to
24) may be opened and closed 15 times per second; and spray can be launched from the vehicle
to targeted weeds within 15 cm square blocks ( 6 in square) up to 3.7 m ( 12 feet) away from the
vehicle foot print.
For accurate targeting of weeds, constant pressure at each nozzle is required independent of the
number of valves firing ( i. e., nozzles spraying). To accomplish this physical requirement, the
IHAS uses three- way valve technology to resolve the variable valve/ pressure issue.
The required fluid handling changes were:
1) Addition of a fixed nozzle tower for each 0.9 m ( 3 foot) spray region consisting of six
IHAS nozzles per tower with each nozzle targeting a 15 cm square area,
perpendicular to the vehicle foot print, and parallel to the vehicle path ( see figure 1).
10
2) Addition of electronically actuated 3- way valves ( one per IHAS nozzle).
3) Implementation of a method of fluid buffering to minimize herbicide concentration
variation.
4) Implementation of a method of communication between the IHAS and the Raven
injection system.
5) Modification of truck plumbing to allow the IHAS to share system components with
the existing spray system.
Figure 6 shows the nozzle towers on one side of the vehicle with spray valves, supply manifolds
and return manifolds to maintain a constant pressure during variable spray applications.
( a)
( b)
( c)
Figure 6. IHAS spray tower with supply and return manifolds. a) single nozzle tower showing
six spray valves/ nozzles with supply manifold on left and return manifold behind
valves/ nozzles. b) All four nozzle towers on one side of the vehicle. c) three- way
valve with spray nozzle.
11
Plumbing Modifications. A schematic of the modified plumbing system is shown in figure 7.
The only components added in- line with the existing system were a shut- off valve and flowmeter
added to the outlet of the water tank. Nozzle towers were connected into the existing pressure
line in place of the rear hand- gun, however they could simply be added to the existing system to
maintain functionality of the rear handgun if desired. The bypass flow from the valve towers
enters just below the chemical injection point and is mixed as it passes through the centrifugal
pump. The motor and pump pulleys were changed to increase pump speed and the 517 kPa ( 75
psi) regulating valve was opened to increase flow pressure at the nozzle. The desired “ at nozzle”
pressure was 276 kPa ( 40 psi). Pressure gauges were installed on the nozzle/ valve supply
manifolds to allow visual monitoring that adequate pressure was available during spray events.
Figure 7. Schematic of the IHAS fluid bypass system used to minimize fluid pressure and
herbicide concentration variations under variable spray demands ( note IHAS
additions in blue were added to the existing fluid handling system on the vehicle).
To maintain targeting accuracy and constant supply pressure at all valves, independent of the
number of nozzles actively spraying, IHAS incorporated the use of a continuously circulating
bypass system. The IHAS 3- way nozzle control valves allow fluid to flow either through the
nozzle or, when the nozzle control valve is not activated, allows fluid flow bypass to the return
line; flow re- enters the fluid system upstream of the main pump. This design maintains a
12
constant flowrate through the valves and minimizes pressure fluctuations associated with a
variable number of valve activations. The volume of fluid re- circulated in the loop,
approximately 3.2 liters ( 0.85 gal), buffers and stabilizes herbicide concentrations at the nozzle.
These concentrations would likely vary due to transport delays of newly injected chemicals due
to rapid changes in flow demands. This artificial “ fluid tank” is momentarily diluted under
sudden heavy loads due to the increased flowrate of fresh water drawn in to replace the recent
spray output. However, as the heavier weed load is sensed, a control signal is sent to the
injection pump to increase chemical concentrate added to the solution in order to bring the
concentration back to the desired level. One of the IHAS to Raven communication designs,
“ boom switch control,” ( described below) anticipates changes in demand by “ looking ahead”
into the spray map and communicating this information to the Raven with sufficient lead time to
eliminate any concentration lags.
Two different control methods can be used by IHAS to control the rate of chemical injection
with the Raven SLC750 controller ( Raven Industries, Inc., Sioux Falls, SD): “ boom switch
control” and “ ratio rate control.” Since individual IHAS nozzles target the same ground area, it
is possible to use the Raven boom switches to electronically adjust the injection rate in
proportion to the number of IHAS valves open at any specific time (“ boom switch control”). For
“ ratio rate control,” usually intended for handgun operation, the chemical injection rate is set
proportional to the flow of water leaving the system.
When IHAS uses “ boom switch” control, the VCS continuously updates the Raven controller
with the number of nozzle/ valve configurations currently spraying. It is possible to represent the
number of activated valves as a 4- bit binary value and set the boom width value of the least
significant bit to the coverage width of one IHAS nozzle. Therefore, in the Raven SLC750 set-up,
the boom width value for boom 1 is set to 15 cm ( 6 in.), boom 2 to 30 cm ( 12 in.), boom 3 to
61 cm ( 24 in.), and boom 4 to 122 cm ( 48 in.). In IHAS mode the VCS is connected to the
Raven controller in place of the Raven boom switch box and if, for example, four valves are on
( 4 x 15.2 cm = 61 cm), the controller sends a signal on the boom 3 line, which corresponds to 61
cm ( 24 in.). In this way the Raven injection controller receives continuously up- to- date valve
activation data from the VCS and adjusts the herbicide injected accordingly.
The other control method used by IHAS is the “ ratio rate” mode. In this scenario, the system
determines the chemical injection rate set point as a user- specified percentage of the flow of
water leaving the system. This simplifies the VCS operation as no electrical communication is
needed between the Raven injection controller and the VCS. However, the flowmeter on the
conventional ( non- IHAS configured) spray vehicle is located at the pump outlet. In this
configuration, the combined flow leaving the system for spray application and return flow from
the IHAS return manifolds are measured. This combined measurement was separated; the
requirements for IHAS indicated that a flowmeter was required at the outlet of the water tank in
order for an accurate measurement of fresh water entering the system and to ensure ratio rate
applications were configured and injected with appropriate herbicide rates.
13
Weed Mapping System
The IHAS uses color machine vision to develop a spray map of the weeds growing along the
roadway shoulder. From the machine vision image captured, objects with a “ green” appearance,
corresponding to the color of living plants, are classified as weeds and their location noted within
the computer system for spray application. IHAS uses eight 3- CCD video cameras for weed
mapping. Four cameras are mounted on each side of the vehicle: two are configured for
capturing close images and two are configured for far images. Three- CCD sensor technology is
traditionally used when high quality color images are desired without the additional
computational cost required to anti- alias filter a single CCD image with a higher resolution
sensor. For use in the IHAS configuration, 3- CCD technology was used to eliminate false
“ green spots” due to aliasing, that occur when imaging non- plant scenes such as black and white
gravel or dark cracks in bare soil surfaces. Additionally, computation time required to analyze
each image was a concern because spray vehicles typically travel at speeds up to 16 kph ( 10
mph) and the machine vision system must analyze the entire roadway shoulder area ( up to 3.66
m or 12 ft) for weeds in real- time. In the future, as more powerful computers become available,
it may be feasible to use less costly 1- CCD technology.
The cameras were mounted in rain- tight enclosures on a frame that was welded to the front deck
of the spray vehicle ( figure 8). Each enclosure contained two cameras ( Hitachi models HV- D30,
Hitachi Kokusai Electric America, Ltd, Woodbury, NY) where one camera was configured for
operation under illumination with direct sunlight and the second camera was configured to
acquire images in the shaded regions of a captured scene. The machine vision computer ( Matrox
model 4- SightII, 1.2 GHz, Matrox Electronics Systems, Ltd., Dorval, Quebec, Canada) was
equipped with two real- time color video frame- grabbers ( Matrox model Meteor- II). There is a
video control switch mounted on front of the camera support frame that informs the Matrox
computer and PLC which cameras and valve towers to use for spraying ( that is, basically the
switch engages the system for driver or passenger side spraying).
The lower camera enclosure (“ near” cameras) on each side of the truck was positioned to capture
roadside images along a 1.8 m ( 6 foot) perpendicular distance from the vehicle footprint. The
upper camera (“ far” cameras) enclosure captured images between 1.8 m and 3.6 m along the
same transect as the near cameras. When the vehicle is in motion at speeds above 1.6 km/ h ( 1
mph) the radar sensor outputs a pulse stream to a high- speed counter in the VCS. The number of
pulses per time captured from the radar sensor is proportional to the distance traveled per time.
The VCS monitors the radar pulse count to determine vehicle speed and distance traveled.
During video capture and nozzle spraying, a region 0.76 m ( 30 in) wide ( in the direction of
travel) by 1.8 m ( 6 ft) long ( perpendicular to travel direction) is analyzed for weeds in each
image captured by the WMS. The near and far cameras are multiplexed ( i. e. they share the same
frame- grabbers) and images not acquired simultaneously, but sequentially in an alternating
pattern ( i. e. near, then far, then near, then far, etc.). Thus, the VCS outputs a trigger or
synchronization signal to the WMS every 0.38 m ( 15 inches) of vehicle travel. At each trigger
signal, a pair of images ( both sun and shadow) from either the near or far camera set are acquired
and analyzed to create the weed map for the corresponding region of shoulder being imaged.
Once analyzed, the weed map is transferred to the VCS for spray application to weed specific
regions.
14
1: Positioning Mechanism
7: Camera
8: Lens
Figure 8. Camera mounting system for IHAS; the left image shows the camera mounting frame
with four camera enclosures and the right image shows one camera enclosure with the
front cover removed revealing the sun camera ( top) and shadow camera ( bottom).
Camera resolution tests. Camera resolution for the actual target size that can be detected and
sprayed were determined for both near and far cameras on the passenger side of the vehicle ( with
the assumption that cameras on both sides of the vehicle would be similar). These tests were
done on a flat, black- top surface on the UC Davis campus. All tests used 0.635 cm ( ¼ in) thick
green scrubbers used for cleaning kitchen ware cut to several different square dimensions.
Based on the 24 nozzles, per side of vehicle, for spraying up to 3.66 m ( 12 ft), it should be noted
that each nozzle is adjusted to spray over a 15.2 cm ( 6 in) square area. Hence, nozzle 1 is
targeted within a square area beginning at a distance of 0.305 m ( 1 ft) and ending at a distance of
0.457 m ( 1.5 ft), which is the starting point for the 15.2 cm square target area for nozzle 2. The
remaining nozzles are targeted in a similar fashion and distances are easily determined from the
first nozzle setting.
For the resolution tests, scrubbers were centered within each 15.2 cm ( 6 in) square area
corresponding to each nozzle along a perpendicular transect from the truck. The spray boundary
for nozzle 1 begins at a perpendicular distance of 30.5 cm ( 1 ft) from the vehicle footprint. Table
15
1 gives the results from the resolution tests. Minimum target size for the near cameras, that is,
up to a distance of 1.83 m ( 6 ft), was a 1.9 cm ( 0.75 in) square. Minimum target size for the far
cameras was a 3.8 cm ( 1.5 in) square.
Table 1. Camera resolution tests; results are indicative of size of material recognized by cameras
for spray application.
Size of scrubber Speed of vehicle Near camera tests Far camera tests
cm ( in) kph ( mph) Percent of targets Percent of targets
sprayed, % sprayed, %
0.635 x 0.635 ( 0.25 x 0.25) 4.8 ( 3) 0 0
1.27 x 1.27 ( 0.5 x 0.5) 4.8 ( 3) 0.05 0
1.9 x 1.9 ( 0.75 x 0.75) 4.8 ( 3) 100 0
12.8 ( 8) 50 0
2.54 x 2.54 ( 1 x 1) 4.8 ( 3) 100 50
12.8 ( 8) 100 0
3.8 x 3.8 ( 1.5 x 1.5) 4.8 ( 3) 100 100
12.8 ( 8) 100 100
7.62 x 7.62 ( 3 x 3) 4.8 ( 3) 100 100
12.8 ( 8) 100 100
Spray deposition tests. Spray deposition assessments of the system were determined for a
general broadcast scenario, with all nozzles spraying, and a random target analysis using 6
targets randomly placed within a predefined grid ( figure 9). The test location was on the UC
Davis campus with a flat black- top ground surface. Two vehicle speeds were evaluated for the
broadcast and random target tests: 3 mph ( idle speed) and 10 mph. The broadcast application
entailed turning all valves on and driving by the targets at the test speed, with three replicates per
test; random target tests were also replicated three times. All targets for all tests used 15.2 cm ( 6
in) square green scrubbers. Deposition, or spray recovery assessment on targets, was determined
by using brilliant sulfaflavine ( BSF). The carrier fluid was mixed and analyzed prior to all tests;
average recovery of the carrier fluid tank was approximately 19.8 ppm ( BSF). Weather
conditions over the test duration are given in Table 2.
16
Figure 9. Grid set- up for spray deposition showing replicate and speed of vehicle for the tests.
Table 2. Weather conditions for spray deposition tests; data were obtained from the California
irrigation management information system ( CIMIS) website for the UC Davis campus.
Air Vapor Wind Wind Relative Dew
Temp Pressure Speed Direction Humidity Point
Time C kPa m/ s 0- 360 % C
1100 11.2 0.8 4.2 348.5 59 3.6
1200 12.5 0.8 2.7 349.4 55 3.6
1300 13.8 0.7 1.3 298.0 47 2.8
Broadcast deposition results from the two test speeds are given in Table 3. The concentrations
are given as a percentage of the spray tank mix concentration ( 19.8 ppm) along with average
concentration, standard deviation and ranges of the actual concentrations ( ppb).
Random target spray deposition results are given in Tables 4- 6. Table 4 gives each deposition
replicate as a percentage of the spray tank mix ( 19.8 ppm). Table 5 gives a comparison of the
average of the replicated target depositions normalized to the average ( all 24 targets) broadcast
deposition for each test speed and the average of the replicated target depositions normalized to
the spray tank mix ( 19.8 ppm).
Table 3. Broadcast depositions as a percentage of spray tank mix ( 19.8 ppm) for two test speeds
and range of actual spray depositions concentrations ( ppb) for all targets.
17
3 mph, broadcast 10 mph, broadcast
Percent of Percent of
Target tank concentration tank concentration
1 1.5 0.9
2 1.6 1.0
3 1.9 1.1
4 2.4 1.3
5 3.7 1.6
6 4.6 1.6
7 5.7 2.1
8 7.9 2.5
9 9.3 1.9
10 6.9 3.6
11 5.4 3.0
12 6.3 4.5
13 3.9 3.3
14 3.4 3.7
15 4.9 2.8
16 4.9 2.1
17 3.9 1.8
18 3.7 2.2
19 4.6 2.4
20 4.2 1.8
21 3.7 1.9
22 1.2 1.8
23 0.7 1.3
24 0.1 0.6
Concentration, ppb Concentration, ppb
Average 796.5 419.0
St. Dev. 446.6 192.1
Minimum 26.0 115.1
Maximum 1836.6 886.8
Table 4. Random target deposition results as a percentage of tank concentration ( 19.8 ppm).
18
Target 3 mph 10 mph
Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3
Percent of tank mix Percent of tank mix
3 3.1 7.6 4.6 1.5 0.7 1.8
8 5.1 6.1 12.3 3.9 2.5 2.4
12 1.8 7.8 10.3 1.3 0.9 3.3
15 7.0 5.8 5.4 1.2 0.4 1.3
18 7.3 7.3 6.1 2.8 0.6 1.8
22 2.5 2.7 1.2 0.4 0.0 0.7
Table 5. Random target deposition averages for two test speeds; deposition for each replicate
target location was averaged and normalized to average broadcast deposition ( all 24
targets) and spray tank mix concentration ( 19.89 ppm).
------- 3 mph --------- -------- 10 mph ---------
Target Percent Percent Percent Percent
of broadcast of tank mix of broadcast of tank mix
3 100 5.1 63 1.3
8 100 7.8 100 2.9
12 100 6.6 87 1.8
15 100 6.1 46 1.0
18 100 6.9 82 1.7
22 53 2.1 18 0.4
19
Data Logging System
An environmental data and as- applied spray actuation location logging system was assembled
from several manufacturers’ standard components and a custom built communication gateway
controller. The communication backbone of the data logging system ( DLS) was a DGPS capable
commercial spray rate controller ( Model Legacy 6000, Midwest Technologies, Wheaton, IL)
coupled with a Control Area Network ( CAN) to allow a user interface console to communicate
with and record data from four Product Control Modules, a Power Speed Module, and a Switch
Sense Module. Analog inputs of the Product Control Modules were connected to sensors to
detect wind speed, wind direction, ambient temperature, and roadside slope. In the instance that
any of these measured conditions fell outside pre- defined minimum and maximum values, the
communication gateway triggered the VCS to prevent (“ lock- out”) spray discharge. The
measured environmental conditions and positions where valves were activated were recorded
with GPS coordinates to produce maps of areas sprayed with corresponding environmental data
at the time of spray. The user manual for the DLS is given in Appendix D.
The communication gateway was constructed to passively listen to the messages on the Control
Area Network and to notify the VCS to disable the spray system if the environmental conditions
exceeded preprogrammed threshold values. The gateway box also monitored inputs connected
to solenoid valve drive lines in order to capture information on when valves were triggered and
spray was being discharged from the vehicle. The gateway controller executed a sample and
hold routine; valve pulses triggered the gateway controller to set an output line to the Switch
Sense Module and hold the line high for 900 ms. In this way, the Legacy 6000, which logged 1
data point per second, could detect any valve triggered during the 1- second period without regard
to the duration of the valve pulse.
A DGPS receiver ( Model AgGPS 132, Trimble Navigation Ltd., Sunnyvale, CA) was connected
to the Legacy 6000 console and was used to reference the environmental sensor data to current
latitude and longitude coordinates. The geo- referenced data was written to a map file on a flash
card for later retrieval and analysis. The GPS antenna was attached on the centerline of the
vehicle between the CCD cameras ( adjacent to the wind speed sensor). Since the antenna was a
few meters in front of the spray valves actual locations of spray deposition were shifted
backward from the referenced locations in software.
A directional wind speed sensor ( Model Wind Sonic, Gill Instruments Ltd., Lymington,
Hampshire, UK) was mounted above the cab of the truck and connected to two of the Product
Control Modules. One output from the sensor indicated the relative wind speed ( as measured
from the moving vehicle). The second output from the sensor indicated the wind direction
relative to the front of the truck ( as measured from the moving vehicle). The Product Control
Modules converted outputs into values that could be transferred over the CAN. The Legacy
6000 console recorded the measured wind speed with GPS coordinates to a map file on the flash
card. The Raven radar- based speed and displacement sensor used to measure vehicle ground
speed was interfaced to the VCS for spray timing purposes, and was also used by the
communication gateway module to resolve absolute wind speed. That is, the gateway module
subtracted the vehicle velocity vector from the relative wind velocity vector to calculate actual
20
wind velocity. Actual wind velocity was compared with a pre- defined maximum wind threshold
to ascertain if spray lock- out was needed.
An RTD temperature sensor with radiation shield ( Model TT- GPL- R- 100, Enercorp Instruments
Ltd., Toronto, ON) was also mounted beneath the camera frame and interfaced to the Product
Control Module. Temperature values were transmitted on the CAN bus and recorded on a flash
card. The communication gateway module compared temperature values with pre- defined
minimum and maximum threshold values for control of the spray lock- out feature on the VCS.
Two ultrasonic distance sensors ( Model UM30- 15113, Sick AG, Waldkirch, Germany) were
attached on the sides of the spray vehicle angled 40o down from parallel to the ground as shown
in Figure 10. The sensor outputs were routed through a SPDT relay switch to a single Product
Control Module. The relay was controlled by a PLC signal indicating whether the sprayer was
operating on the driver’s side or the passenger’s side. A description of the slope calculation is
given in Appendix D. The DGPS antenna, wind speed sensor, ultrasonic sensor ( and locations)
and Legacy interface screen are shown in Figure 10.
Measurement Verification. Conversions from voltage to measured units were programmed into
the Legacy Product Control Modules for each sensing instrument. Accuracy of sensor outputs
and unit conversions were verified by comparing system measurements to those from other,
independent, instruments. Because the DLS was designed to collect environmental conditions
while moving, verification measurements were conducted on the moving vehicle. Included in
the ambient condition tests were verifications of absolute wind speed, absolute wind direction,
ambient temperature, roadside slope, and spray deposition geo- referencing.
All tests for the system interface and capturing of the environmental conditions were conducted
on or near the UC Davis campus. For the wind velocity tests, conditions on the test day
indicated wind direction from due north at an approximate speed of 4.5 m/ s ( 10 mph). Ambient
weather conditions included clear skies and a temperature of 11 C ( 52 oF). Environmental
conditions and GPS latitude, longitude, and time stamp data were collected on the vehicle with
the DLS. Wind conditions and time stamps were also recorded with a stationary weather station
( Model Ultimeter 2000, Peet Bros., St. Cloud, FL). In order to validate the wind velocity
correction algorithm, wind velocity minus vehicle velocity, the vehicle was driven east and west
at varied speeds. Three repetitions of eastbound/ westbound data were collected.
In post- process, GPS coordinates and time stamps were used to calculate the spray vehicle’s
velocity. Measured wind velocity was subtracted from vehicle velocity to calculate absolute
wind velocity. Time- referenced wind speeds and wind directions were compared with those
measured from the stationary weather station for accuracy verification. Table 6 shows the
average measured wind speeds and average measured wind directions during a series of replicate
test runs. Figure 11 displays an example test path with resulting wind vectors. Although the
eastward and westward data points were nearly overlapping, the map shows that wind velocity
was nearly the same for both travel directions. Data from Table 6 show that wind speed
measured on the vehicle was consistently lower than the speed measured by the stationary
instrument. However, two of the three averages were within one standard deviation of the
stationary wind speed measurements.
21
( a)
( b)
Figure 10. DLS instrumentation: ( a) DGPS antenna, wind sensor and ultrasonic sensor, ( b)
Legacy interface screen.
Table 6. Absolute wind speed and direction.
On- board Stationary On- board Stationary
Rep 1
Average 4.582 5.167 339.7 346.8
Standard Deviation 1.138 1.071 18.7 15.6
Rep 2
Average 4.184 6.083 321.8 320.9
Standard Deviation 1.243 0.891 18.9 15.0
Rep 3
Average 4.499 5.328 325.9 351.6
Standard Deviation 1.627 0.889 18.6 16.7
Wind Speed ( m/ s) Wind Direction ( degrees)
Wind sensor
Antenna
Ultrasonic sensor
22
0 50 100 150 200 250 300 350
0
10
20
30
0 to 3
3 to 4
4 to 5
5 to 6
6 to 8
Wind Speed ( m/ s)
Distance ( meters)
Travel direction
Travel direction
Figure 11. DGPS wind map from field verification tests
Ambient temperature measurements were recorded with the DLS and verified with a shaded
stationary thermometer. Measurements on the vehicle were recorded while moving in order to
prevent engine heat from increasing ambient temperature measurements; data are given in Table
7. Because ambient temperature changed slowly over time, visual representations of temperature
data are not very revealing. In order to demonstrate the mapping capability of ambient
temperature, data was collected on the spray vehicle in the morning, the vehicle was parked for a
few hours, and collection was continued in mid- day. Figure 12 displays an ambient temperature
map that indicates a temperature change between morning and afternoon hours. Results indicate
that on- board temperature measurements were within 1 degree C of those obtained with the
stationary thermometer. Because the efficacy of most agricultural chemicals changes over a
large temperature range, the sensor accuracy was deemed sufficient for this application.
Additionally, most maps revealed very little change in ambient temperature, but visual
representations that were collected over large time intervals did indicate temperature fluctuation.
Table 7. Ambient temperature on spray vehicle vs. stationary thermometer.
On- board measurement Actual
21.9 22
25.4 25
25.5 26
28.1 28
30.1 30
Temperature ( degrees C)
23
0 50 100 150 200 250 300
0
20
40
10 to 11
14 to 15
Distance ( meters)
Travel direction Temperature 9: 25 am 1: 10 pm ( degrees C)
Figure 12. Temperature map of morning to mid- day.
Roadside slope assessments entailed measuring three repetitions of distance to the nearest
roadside object with the ultrasonic sensors and recorded with the DLS while driving south on
California State Highway 113 north of Davis, CA. The slope of the roadside varied widely and
ranged from a ditch ( with a negative slope), to flat sections, to inclined sections sloping upwards
approximately 35 degrees. GPS latitude and longitude were also recorded to geo- reference the
roadside distance data. Roadside slopes were also manually measured with a tape measure, to
the top of the vegetation, and locations were recorded with a GPS unit with centimeter accuracy
( Model RTK GPS, Trimble Navigation Ltd., Sunnyvale, CA).
Geo- referenced slope measurements were compared with those measured with a tape to validate
the distance sensor calibration and slope calculations. Results are given in Table 8. The
resulting slope measurements from the DLS averaged 3.2 degrees lower than those measured
with a tape measure. Inaccuracies may have resulted because the ultrasonic sensor did not
measure distance to a point but distance to the nearest object in an area, and with uneven terrain,
roadside slopes were not exact values. However, the system could estimate the slope within a
few degrees, allowing possible spray lockouts for roadside slopes outside those practical for
spraying.
Spray deposition assessments with geo- referencing were also determined. The Legacy 6000
system with Roadway Management Software contained geo- referenced corrections for different
positions on a spray boom. However, the software did not support automatic correction for an
offset between the GPS antenna and the boom. That is, the software assumed that the antenna
was located at the center of the spray boom as is the case in the IHAS vehicle. Thus the
correction for an offset between the GPS antenna and the boom must be done manually using
some type of geographic information system ( GIS) software package.
As the data logging system recorded environmental conditions and spray valve activation, GPS
coordinates were assigned to locations at which valves were triggered for spray. Due to
limitations on the number of inputs that could be recorded by the Legacy 6000 system, the spray
status ( i. e., on or off) of 8 of the 24 IHAS valves were recorded by the Legacy 6000 system.
Valve numbers 2, 5, 8, 11, 14, 17, 20, 23 were interfaced to the Legacy 6000 system
corresponding to target locations 0.5 m, 1 m, 1.4 m, 1.9 m, 2.4 m, 2.8 m, 3.3 m, and 3.7 m from
24
Table 8. Measured roadside slope versus actual slope.
Distance Slope Actual Slope
( cm) ( degrees) ( degrees)
321.1 - 4.3 - 1.2
319.7 - 4.1 - 0.6
319.3 - 4.0 - 1.2
341.9 - 6.9 - 5.4
339.3 - 6.6 - 4.8
339.3 - 6.6 - 4.8
210.8 18.0 18.4
215.2 16.8 20.0
194.9 22.8 23.6
185.8 25.7 27.5
195.4 22.6 30.3
214.9 16.8 26.7
On- board measurement
the edge of the truck. Normally, the IHAS is operated in “ expanded pattern” mode where three
valves ( the one directly targeting the weed plus the two adjacent valves) are activated for each
weed to be sprayed. Thus in expanded spray pattern mode, when a weed is to be sprayed by one
of the 16 valves not interfaced to the Legacy 6000 system, one of the 8 valves that is interfaced
to the Legacy 6000 system will also be activated because it will be adjacent to an activated valve
and thus will also be activated as part of the expanded spray pattern. This allows the complete
GPS spray logging of the entire 3.6 m ( 12 foot) region scanned by the IHAS when operated
normally. If the operator deactivates the expanded pattern mode, the spray activation of the 16
valves not interfaced to the Legacy 6000 system will not be recorded.
The DLS recorded the GPS location of the GPS antenna when a valve was activated. The GPS
location of the actual spray deposition must be calculated from the truck location at the time of
valve actuation, the distance from the GPS antenna to the spray nozzle, the time of flight of the
spray packet, and the vehicle travel speed. Since the GPS antenna was mounted on the camera
frame, the distance in the direction of travel from the GPS antenna to the spray nozzle was the
same as the distance from the camera to the spray nozzle stored in the VCS. The spray packed
time of flight was also stored in the VCS for each nozzle.
A spray test was conducted to verify the geographical relationship between the logged locations
and the actual locations of spray deposition. GPS antenna locations were recorded with the on-board
system as solenoid valves were triggered for green spray targets under normal IHAS
operation. A centimeter- accurate GPS sensor ( Model RTK GPS, Trimble Navigation Ltd.,
Sunnyvale, CA) was used to determine the actual locations of spray deposition on the ground.
Three repetitions at three different travel speeds ( 0.9 m/ s, 2.7 m/ s, and 4.5 m/ s) were executed to
gather data from spray deposition locations and valve- trigger locations from the data logging
system; data results are given in Table 9.
25
Table 9. Geo- referenced offset between valve triggers and actual spray deposition.
Offset between GPS map location and actual spray deposition
In the direction of travel ( m) Perpendicular to the
Vehicle Speed ( m/ s) Raw Data Post- Process Corrected Direction of travel ( m)
4.47 6.93 0.10 0.12
4.47 6.70 - 0.13 - 0.17
0.89 3.51 0.23 0.16
0.89 3.42 0.14 - 0.21
0.89 3.48 0.20 0.26
0.89 3.27 - 0.01 0.02
2.68 5.05 0.00 0.50
2.68 4.93 - 0.12 0.16
2.68 5.15 0.10 0.51
2.68 5.15 0.10 0.18
0.89 2.96 - 0.32 0.54
0.89 3.04 - 0.24 0.26
4.47 6.78 - 0.05 0.62
4.47 6.89 0.06 0.07
4.47 6.91 0.08 0.59
4.47 6.73 - 0.10 - 0.25
As Table 9 indicates, large discrepancies existed between the raw ( uncorrected) recorded valve-trigger
locations and the locations of spray deposition in the direction of travel. A portion of
these differences resulted from the physical offset between the GPS antenna and the location of
the spray valves and the time of flight for spray packets. Additionally, because the valve- trigger
data was recorded once per second while spray valves were activated at up to a 10 Hz rate, the
location differences also varied as a function of vehicle speed. To correct for both of these
sources of error, a plot of the raw position offset error versus vehicle travel speed was made
( figure 13). A linear regression analysis was used to develop a correction equation based upon
travel speed. When this equation was used to correct the spray deposition location in the
direction of travel, the resulting differences were within the expected uncertainty of the system.
26
y = 0.9909x + 2.3979
R 2 = 0.9906
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Vehicle Speed ( m/ s)
Position Offset ( meters)
Figure 13. Plot of position offset in the direction of travel of GPS map data from actual spray
deposition versus vehicle travel speed.
The IHAS/ Legacy spray lock- out function was tested for each of the predetermined and
programmed thresholds: maximum wind speed, minimum temperature, maximum temperature,
minimum roadside slope, and maximum roadside slope. The thresholds were set to values
typically achievable on the day of the test. Note that Appendix D provides programming
information for the Legacy.
The Legacy system is capable of alarming the driver if the truck is within 1000 ft of a “ hazard”,
for example a school location. The locations of the hazards must be pre- defined in a road
markers map. For use with the IHAS truck, hazard markers are placed around the perimeter of
schools in a road- markers map. This map file is then used by the Legacy system to alarm the
driver when the spray truck is too close to a school. The driver must then shut off spray until
there is an acceptable distance between the truck and the school ( the alarm will stop). Note that
the truck will not automatically turn the spray off. It is the responsibility of the driver to act
when the alarm sounds.
27
Conclusions
A commercial prototype displacement- based precision valve control system ( VCS) was
successfully developed and installed on a Caltrans herbicide spray vehicle. The VCS was
designed to serve as the main controller in the commercial IHAS prototype completed and
described in this document. The VCS was designed to function in a parallel “ add- on” mode and
was compatible with the existing Raven variable- rate herbicide injection system. Additionally,
the interface allows Caltrans to operate the vehicle in a non- IHAS mode if desired.
A fluid handling system was designed for the IHAS and functions in a parallel " add- on" mode to
allow compatibility with, non- IHAS, variable- rate herbicide injection systems currently used on
Caltrans herbicide spray vehicles. An IHAS valve tower with 3- way valves ( one per spray
nozzle) served as the basis for the IHAS fluid handling system. This design has two benefits.
First, provides a means of maintaining a constant pressure at the valve and is independent of the
number of valves activated. Second, it allows bypass spray mix to be recycled in a small loop
through the pump and 3- way valves. This provides a small fluid buffer to minimize variation in
herbicide concentration as the number of spray valves activated varies.
Two different control methods were developed ( boom switch and ration rate) for the IHAS and
was compatible with the existing Raven chemical injection controller. The response times of the
Raven controller in IHAS mode were characterized, and results indicate that the system was able
to maintain spray mix concentrations within 10% of the desired level throughout the range of
flowrates possible in IHAS mode. Also, results found that fluid pressure at the valves was very
consistent over the range of flowrates possible in IHAS mode. In general, the boom switch and
ratio rate control modes gave comparable performance, with the ratio rate mode showing a
reduced ability to respond to rapid demand changes at high application rates. The boom switch
mode had the advantage of “ looking ahead”, so the VCS could anticipate upcoming spray
demand changes and reduces concentration delays associated with the Raven response time.
A commercial prototype weed mapping system ( WMS) for IHAS was successfully developed.
The WMS uses color machine vision to map weeds growing along the roadway shoulder. Eight
3- CCD video cameras were deployed, four on each side of the spray vehicle for detecting weeds
in both direct sunlight and in shadows along the roadway shoulder. The WMS successfully
identified weeds of 3.81 cm2 or larger. 3- CCD technology successfully eliminated false
detection non- plant scenes such as black and white gravel or dark cracks in bare soil surfaces.
A DGPS data logging system ( DLS) was installed for capturing location and environmental
information while actively spraying roadside shoulders. Environmental conditions can affect the
quality of herbicide application due to excessive wind, and ambient temperatures may degrade
applied chemicals. Additionally, roadside slopes result in the spray trajectories missing intended
targets. These conditions are now monitored, with later data retrieval capabilities, to ensure
accurate spray applications in addition to having the capability to lock- out automated spray
applications based on pre- determined off- limit areas.
The prototype IHAS can enhance Caltrans effectiveness at minimizing herbicide release into the
environment and providing protection along areas that are environmentally sensitive.
28
Appendix A: Adaptive Equipment VCS Manual
29
Valve Control System
Documentation r1
August 24, 2001
Customer: University of California
Davis, CA
AE Contacts: Ward Simonton
Roy Harrell
2512 NE 1st Blvd
Unit 400
Gainesville, FL 32609
352- 372- 7821
30
I. SAFETY
A. Basic Safety Practices
• All personnel responsible for servicing or operating this system should read this documentation.
• Electrical power should be turned off and locked out before servicing the system. The main electrical cabinet
contains 120 VAC connections. Service should not be performed without removing power
beforehand. Reference wiring diagrams UC. WD. 01- 08.
• One system stop is provided: a red palm button mounted on a remote electrical box referred to
as OI/ S ( Operator Interface/ Switches). When the red palm button is pressed, the Master Control
Relay in the control panel will be de- energized and power to the control relays for the spray
nozzles will be removed.
• On a system stop, power will not be removed from sensors, pilot lights, or interface electronics.
II. SYSTEM FUNCTIONALITY
A. Overview
The Valve Control System ( VCS) is a truck mounted system designed to energize and de- energize
12 valves at rates up to 30 Hz each for the purpose of road side spraying. Valves can be placed
into one of three modes: OFF, ON ( valves are on continuously), and AUTOMATIC ( valves are
controlled based on a dynamic spray pattern). Spray patterns must be transferred to the control
system through an Ethernet communication port using direct addressing of controller memory.
Travel speed and distance is measured using a radar sensor with displacement output. The effects
of a range of operating speeds, spray flight travel times, and distance between spray pattern sensor
and valves are accommodated for in the controls.
B. Specifications
1. Max operating travel speed = 10 mph
2. Range of spray pattern band1 widths = 6” to 12”
3. Range of spray pattern bands per block or image = 3 to 7
4. Max number of bands between spray pattern sensor and spray boom = 40
5. Max time of flight for nozzle stream = 0.5 s
6. Programmable minimum valve open time ( for entire valve set)
7. Programmable valve lead time ( for entire valve set)
1 A band is a columnar section of a spray pattern block or processed image.
31
III. SYSTEM OPERATION
A. Cabinet Startup
All circuit breakers should be in the On position ( UC01. WD. 01 and UC01. EL. 01). The SLC505
PLC ( programmable logic controller) is the primary control device in the system. When 120 VAC
power is applied to the main control panel, the SLC505 will initialize provided that it is in Run
Mode. The mode of the SLC505 is set by a key on the unit. OI/ S can be used to power up the PC
as described in Section III. C.
B. Cabinet Shutdown
Before removing 120 VAC power from the cabinet, the PC should be made to go through a
controlled shutdown as described in Section III. C below. Once the white status light on OI/ S goes
from ON to OFF indicating that the PC has completed a shutdown, then power can be removed
from the cabinet.
C. PC Power up and Shutdown
OI/ S ( see Section III. D below for a complete description of OI/ S functionality) can be used to
remotely power up and shutdown the PC using the three position selector switch. For this switch
to be active, ( a) the red pushbutton ( for Stop Spraying) must be depressed, ( b) the PLC must be
powered and in Run Mode, and ( c) 24 VDC control power must be available. Turning the switch
momentarily to the right will initiate power up on the PC; the PC will boot, load the WinNT
operating system, perform an Auto Login, and load two applications ( an operator interface
( OI/ PC) and a sample program for PLC interfacing). The status light on OI/ S will go from ON to
OFF when the PC has completed the boot process ( 2 minutes) and is ready for operation.
To shutdown the PC, the selector switch on OI/ S can be turned momentarily to the left. The PLC
then sets a flag that is read by the sample interfacing program running on the PC noted above; this
PC program resets the PLC flag as an acknowledgement, then it launches a WinNT shutdown
application. When the status light on OI/ S goes from ON to OFF ( 50 seconds after the PC
acknowledgement), the PC should have completed the shutdown. Note that there is no positive
feedback on this process.
Note: If the PC is booted using the OI/ S switch, then it should be shut down with this switch also. This is due to PC
power up and shutdown interlocks embedded in the PLC control software.
The PC can also be booted using the switch located on the PC enclosure. This switch ( when
toggled to the right) is identical to many off- the- shelf PC power buttons: ( a) momentary action
when PC is not powered= power up signal; ( b) momentary action when PC is powered= sleep
signal; ( c) action of greater than 4 seconds= power down signal. There is no connection with this
switch and the PLC control system.
Note: If the PC is booted using the switch on the PC enclosure, it should be shut down with this switch also. This
switch may be used if the PLC is not powered or not in run mode.
32
D. Operator Interface/ Switches ( OI/ S)
OI/ S is a hard wired PLC interface. There are four panel devices on OI/ S.
1. The three position selector switch requests the PLC to control PC power up and shutdown as
described above.
2. The green pushbutton enables spraying.
3. The red palm button disables spraying and enables PC power control when depressed.
4. The panel light is used to signal several events:
( a) PC power up in progress;
( b) PC shutdown in progress;
( c) spraying has been initiated but there is no spray pattern data available; or
( d) invalid parameter data ( e. g. bandwidth, number of bands, etc) has been entered into
the controller.
E. Operator Interface/ Touch Panel ( OI/ TP)
OI/ TP is a PLC interface and operates over a continuous RS232 communication channel. Its
primary function is to allow the truck operator to individually set valves to either OFF, ON, or
AUTO mode. OI/ TP also displays to the operator the following information: spraying status, the
relative spray rate averaged over a 1.0 second period, and whether or not the main control cabinet
is overheated. OI/ TP is a programmable interface and can be configured for many other operator
functions.
F. Operator Interface/ PC ( OI/ PC)
OI/ PC is a software application on the PC that allows for configuring both valve/ nozzle and vision
parameters in the PLC. When OI/ PC loads, the current PLC parameters are read and displayed.
Parameters can be edited, and when the proper values are entered, can be updated to the PLC.
OI/ PC operates over Ethernet and performs PLC communication on an event- driven basis.
G. Spraying
The spray pattern for the set of 12 valve/ nozzle pairs will depend upon the mode configuration for
each valve ( OFF, ON, AUTO) and the spray pattern sent to the PLC. To begin spraying, the red
palm button on OI/ S must be released and the green push button pressed momentarily. Spraying is
then enabled. Nozzles whose valves have been configured in ON mode will begin spraying
immediately. AUTO mode for those nozzles so configured will be initiated when the truck speed
is greater than approximately 1.5 mph. Control parameters used for AUTO mode are updated
dynamically as the truck speed changes.
33
IV. ELECTRICAL OVERVIEW
A. Electrical System Documentation
Refer to UC01. WD. 01- 08, UC01. EL. 01- 03, and the Cables & Connectors table for a complete
description of the electrical system. All are bound together in a separate binder.
B. Power
The following table outlines power requirements for the control system.
Qty Device Voltage Load ( A) Power ( W) Cable # Service Manufacturer
1 input power 120 VAC 10.0 1200 1 10
1 PC 120 VAC 1.7 200 2 6 Dog Bytes
1 monitor & cabinet fans 120 VAC 1.3 150 3 6
1 PLC 120 VAC 0.6 75 4 2 Allen- Bradley
1 24 Vdc power supply 120 VAC 0.6 76 5 2 idec
AC load = 501
12 valves ( set 1) 12 VDC 12.0 144 7 16 UCD
1 ground speed sensor 12 VDC 0.3 3.6 " Raven
12 valves ( set 2) 12 VDC 12.0 144.0 8 16 UCD
1 boom switch interface 12 VDC "
12V load = 292
1 24 Vdc output 24 VDC 1.3 30 6 2
24 solid state relays 24 VDC 0.48 11.5 Phoenix
4 solid state relays 24 VDC 0.06 1.4 Weidmuller
5 solid state relays 24 VDC 0.10 2.4 Phoenix
1 touch panel 24 VDC 0.75 18.0 AD
24V load = 33
34
C. Controller
The controller for the VCS is an Allen- Bradley SLC505 programmable logic controller. This
controller has one RS232C serial port and one 10 Mb Ethernet port to support processor
communications. Ladder logic is used to program the controller.
D. Valve Control
Twelve PLC 24 VDC outputs are used for independent valve/ nozzle control ( UC01. WD. 03).
Each output controls a solid state relay, the output of which is 12 VDC. For future expansion,
additional outputs have been installed in Slot 2.
E. Boom Switch Interface
Five PLC 24 VDC outputs are used to interface with the Raven boom switch box ( UC01. WD. 03).
Each output controls a solid state relay, the output of which is 12 VDC. The five outputs are used
to transmit a 0- 31 decimal value to indicate a normalized rate of spraying over a one ( 1) second
period.
F. Spray Pattern Sensor Triggering
Two PLC 24 VDC outputs are used for independent sensor triggering and vision frame acquisition
( UC01. WD. 03). The sensor, or camera, trigger output controls a solid state relay, the output of
which is a 5 V TTL pulse signal. The frame acquisition output also controls a solid state relay, the
output of which is a 5 VDC pulse signal. A spare TTL and a spare generic 5 VDC solid state
relay are included to match sensor and frame electronics needs.
G. PC Control
One PLC 24 VDC output is used for controlling the power up and shutdown of the PC
( UC01. WD. 04). The PLC output controls a mechanical relay, the output of which is used to
connect the Power Control input on the PC motherboard to PC common ( O VDC).
H. Raven Radar Sensor
The pulse output of the Raven radar sensor is used to track truck position change and to calculate
truck speed. The output of the sensor is used to drive a high speed counter module located in slot
5 of the PLC ( UC01. WD. 05). Due to the sensor having a 12 VDC output, an 1800 ohm resistor is
used to reduce the voltage that is input to the counter module.
I. Cabinet Fan
An electrical cabinet fan and vent is included to assist in cooling during operation or system
testing. The fan may or may not be required depending upon future testing and environmental
conditions. Regardless of whether the fan is actively used, periodic cleaning and replacing of the
vent filter elements will be required. Frequency will depend also on environmental conditions.
35
V. SOFTWARE OVERVIEW
A. PLC
The primary control software for the VCS is sprayer. rss, a ladder logic program for the AB
SLC505 PLC. RSLogix500, a Rockwell Software development application, was used to write and
debug the software. The program is divided into three separate routines: MAIN, which contains
the majority of the control logic; UPDATE, which performs calculations when system parameters
has been updated; and NEW_ IMAGE, which performs the required calculations and logic upon
each instance of an image boundary being crossed.
B. PC Operator Interface ( OI/ PC)
The PC operator interface, ValveControlOI. exe, is a Visual Basic program that provides a means
to configure valve/ nozzle and vision parameters in the PLC. This program uses four screens for
displaying data, editing data, and sending data to the PLC.
C. Sample Power Up and Shutdown PC Application
The Visual Basic program ShutdownTest. exe is a sample method of how to interface with the PLC
regarding PC startup and shutdown. This program is designed to load during the WinNT
automatic login procedure and communicate to the PLC that the PC is ready for operation. Every
two seconds afterwards, the program checks to determine if the PLC has requested a PC shutdown
event. If so, the program will acknowledge the request and then initiate a WinNT shutdown
procedure.
D. Spray Pattern Emulator
The Visual Basic program Emulator R3. exe generates various spray patterns to emulate processed
vision data. Data generated to emulate a spray map is displayed graphically. Patterns include
random, checkerboard, vertical striping, and sinusoidal. Each pattern can be configured for
frequency, etc. Patterns can be dilated. These patterns can be transmitted in real time to the PLC
for live spraying tests.
E. PLC- PC Communication
Each of the Visual Basic programs described above use OPC software for PLC- PC
communication. OPC, “ OLE for Process Control,” is a software standard providing a means for
applications on personal computers to exchange data with other personal computer applications
and with control computers such as programmable logic controllers. OPC is analogous to TCP/ IP
in that it is a layered software protocol independent of the hardware. However, unlike TCP/ IP,
OPC has a user layer that is designed for embedded configuration and communication in
applications such as those written in Visual Basic and C.
The VCS uses two purchased programs to embed OPC functionality: AB Ethernet Suite Top
Server and OPC Data ActiveX Control. The former contains both an Ethernet driver to establish
communication with the SLC505 PLC and an OPC server. The latter encapsulates the many OPC
function calls into an ActiveX container to greatly simplify programming. The ActiveX control
can be used in C programs in addition to Visual Basic programs. Both the AB Ethernet Suite Top
Server and the OPC Data ActiveX Control were purchased from Software Toolbox, Matthews,
NC.
36
Appendix B: VCS & WMS Electrical Schematics from Adaptive Equipment
53
Appendix C: IHAS User Manual
54
1. Pre- spray checklist and System Startup
• Open carrier valve on undercarriage of truck.
• Start truck, allow several minutes warm up time and air- brake charging.
• Place diesel engine switch ( for driving pump) in truck cab, to “ ON” position.
• Toggle video switch box for spray application side ( passenger or driver).
• Prime centrifugal pump.
• Place kill switch ( on outside of truck, passenger side, next to pump) for diesel engine
driving pump to “ UP” position.
• Start motor for engine pump, ensure pressure stabilization.
• Once pressure stabilization has occurred, kill diesel motor for driving pump, replacing
kill switch to “ UP” position.
• Open main AE control cabinet vents.
• Switch generator power switch to “ ON”, start generator.
• Power system up by turning “ 10A” switch to “ ON” position.
• Touch panel indicates system is initializing, and this message is removed once system has
initialized and is ready to spray.
55
2. Spray applications from inside driver cab and touch panel interface
• Turn diesel motor ( driving pump) on with remote switch ( inside driver cab).
• Pull red stop spraying button out.
• Direct truck to spray location.
• Start moving, engage green spray button.
• Speed boundaries are 2 – 10 mph.
• Check touch panel occasionally for speed, spraying messages, etc.
• Press red stop spraying when completed spray application.
• Can re- engage green button after stop spraying has been pushed for continued
application, if desired ( pull out red stop button for this activity).
• Recommendation: power system down when switching side to spray ( with video switch
box), toggle red switch, power system up.
3. System Shutdown
• When completed spraying, turn switch for diesel engine driving pump to “ OFF” position.
• Press System Shutdown button in top right hand corner of Main System Menu.
• When message flashes “ System Must Now Be Shut Down”, turn 10A switch on generator
to “ OFF” position.
• Turn generator power switch to “ OFF” position.
Notes:
If using touch screen to go through a series of spray application tests, then must shut down
system through the touch screen. Once the touch screen indicates “ System Must Now Be
Shut Down”, then cycle AC power off. Process takes a few minutes.
If use monitor to re- configure the system and/ or re- compile the system vision program, then
must shut down system through the computer ( ctrl- alt- del, choose shut down). After monitor
indicates it is safe to power down, then cycle AC power off.
56
4. Touch panel overview
Figures C1 - C7 show the majority of the touch panels that the driver interfaces with during
system spray applications. These figures show the main features and will be discussed starting in
the upper right of Figure 1 and proceeding clockwise. Italics indicate a system message to the
user. No italics indicate the user must press the button for further access. Some menus require a
4 digit code for access. The code is currently set to 3791. Pressing the “ ENTER” button after
entering the password allows access to embedded menus.
Figure C1. Main system menu after system has been powered up.
During the initial power up sequence, a message flashes below the System Shutdown button
indicating that the system is initializing.
Active buttons for user interfacing are: System Shutdown, Nozzle Setup ( On, Off, Auto), Setup
Nozzle TOF, Set Parameters, Debug, Status, Enable/ Disable Expanded Pattern, and Update
Parameters.
If any parameter or sets of parameters are changed, press Update Parameters for these to become
the default parameters for the current spray application.
System Fault Number
Update INVALID or
Update VALID
Please Wait… Shutting Down or
System Must be Shut Down or
System Initializing
57
Figure C2. Nozzle buttons for manual or automatic control.
Each nozzle has the option to be turned on, off, or set to automatic mode. Automatic mode is the
default spray mode for the vision system. Turning a nozzle on will activate the nozzle to the
open position when spraying is initiated, turning a nozzle off will remove the nozzle from the
spray application.
Pressing Nozzle On/ Off/ Auto Nozzles 13- 24 gives access to nozzles 13- 24.
Back to Main takes the user back to the Main System Menu.
OFF ON AUTO
58
Figure C3. Nozzle time of flight menu.
Each nozzle time of flight can be adjusted. Pressing the respective nozzle button takes the user
into the menu requiring a password. Once the password is entered correctly, the nozzle time of
flight can be adjusted ( see Table 1 for nozzle time of flights and boom spacing).
59
Figure C4. Main spray parameters menu.
Access to these menus requires the password.
Table C1 shows boom spacing for the current configuration. Minimum boom spacing is 108”.
Speed parameters are set to 125 and 625 Hz ( minimum and maximum speeds for pulse train used
to monitor speed from the Raven sensor). Valve Lead Time ( 0.015 s) and Minimum Spray
( 0.01 s) are options for targeting nozzle spray to targets.
Vision modes options: 1 = vision spraying, 2 random spray, 3 = checkered spray, 4 = sinusoid
spray, 5 = vertical spray. Vision Mode 1 is the default spray mode with the system. All other
modes are for demonstration purposes, for system adjustment or system demonstration.
Bands Per Image is used to adjust the number of cells around the target that are sprayed in
addition to the active target.
60
Table C1. Finalized time of flight for nozzles and noted boom distances.
Passenger side
Boom 1 2 3 4
Nozzles 1- 6 7- 12 13- 18 19- 24
------------------ time of flight, ms -----------------------
127 159 171 252
132 169 190 266
142 184 211 280
152 189 232 294
162 209 254 308
172 219 276 324
Distance to
nozzle orifice from 8.95 9.95 10.95 11.95
mid- camera lens, ft
Driver side ( equivalent nozzle heights with passenger side are assumed)
Boom 1 2 3 4
Nozzles 1- 6 7- 12 13- 18 19- 24
------------------ time of flight, ms -----------------------
131 112 275 286
107 275 287 281
124 127 280 295
101 138 253 249
110 149 266 253
119 152 281 279
Distance to
nozzle orifice from 10.98 11.98 10.98 12.98
mid- camera lens, ft
61
Figure C5. System debug menu.
Temp Trigger Period sets the camera trigger for capturing images. During stationary analysis of
camera images, cycling the Auto Trigger OFF and Auto Trigger ON will initiate image capture.
PC test ON sprays without vision mode. However, this will give a diagonal pattern, and will
override whatever setting is made for vision mode. This is basically used to check system
targeting and time of flight values.
PC Test OFF should be activated for all other testing procedures.
Update Parameters after selecting the PC Test option for system to acknowledge change.
PC test OFF is default upon system power up.
NOT in BEGIN Spray Mode or
BEGIN Spray Mode OK
Start up NOT Complete or
Spray Startup Complete
System Fault Number
62
Figure C6. System status menu.
Basic information on system operation during spraying and stationary testing. Each indicator
will highlight a single message.
Cabinet Temp OK or
Cabinet Temp HIGH
NOT Spraying or
SPRAYING
DRIVER Side Mode or
PASSENGER Side Mode
Spray Request OFF or
Spray Request ON
Driver Relays OK or
Driver Relays FAULT
Passenger Relays OK or
Passenger Relays FAULT
System Expanded Pattern DISABLED or Fault Number
Expanded Pattern ENABLED
63
Figure C7. System runtime menu.
This menu appears when the green start spraying button has been pushed and relays information,
on the indicators, to the driver during the spray operation.
Cabinet Temp OK or
Cabinet Temp HIGH
NOT Spraying or
SPRAYING
DRIVER Side Mode or
PASSENGER Side Mode
Expanded Pattern DISABLED or
Expanded Pattern ENABLED
Speed OK or
Speed OUT OF RANGE
System Fault Number
64
5. Additional information and discussion
As requested from AE, the 5th bit on the boom switch control cable is available for monitoring
passenger side spraying ( on = 1) versus driver side spraying ( off = 0). This option implies
previous data collection with LabView could be re- initiated. This bit is available on Cable 4, Pin
5.
Spraying without vision
If spraying without vision and with forward movement, e. g., for demonstration purposes, set
vision mode to option desired and run system under normal operation conditions.
The current option for spraying without vision and stationary is as follows ( this will only send
signals to the main AE control cabinet versus sending to main AE control cabinet and Raven
control system in driver cab):
 Truck battery power off and 24 Vdc switched off in main AE control
cabinet.
 An additional AE box ( FG- AE) is required to connect to the white AE box
( Raven- AE) that resides on top of the Raven speed sensor box. Inside the
FG- AE is a cable for interfacing to the Raven- AE and an external function
generator.
 Use the function generator to select square wave for artificial speed ( pulse
train).
 Truck battery power on, 24 Vdc switched on in main AE control cabinet.
 Function generator needs 12 V peak- to- peak square wave ( with 6 V off-set)
output to mimic Raven radar sensor.
 Set function generator to 550 Hz for 10 mph, 300 for 5 mph, etc.
 Power system up and perform stationary testing as needed.
65
Appendix D: Data Logger User Manual
66
System Overview
The RMS version of the Mid- Tech Legacy 6000 was designed for use with roadside spray trucks
but was intended to control chemical application and map the controlled output. On the IHAS
truck, we have implemented the Legacy system for measuring and mapping environmental
conditions including wind speed, wind direction, ambient temperature, and roadside slope. In
the case that any of these measured conditions falls outside specified minimum and maximum
values, the data logging system triggers the spray controller to shut off spray. The measured
environmental conditions and points where valves have turned on are recorded with GPS
coordinates to produce maps of areas sprayed with environmental data.
This manual is to be used with the Legacy 6000 RMS- EXT Users Guide ( Mid- Tech part num: 98-
05064) for configuration and use of the data logging system. Information on the system
components can be found in the Legacy 6000 Fieldware Users Guide ( Mid- Tech part num: 98-
05053) and PLC spray controller information can be found in the IHAS Spray Checklist & PLC
Operator Users Guide.
System Details
The data logging system consists of several pieces connected through a Control Area Network
( CAN). The CAN is a communication backbone between individual modules. Each sensor
plugs into a module that measures the sensor’s output voltage and communicates the resulting
data on the network. Wind speed, wind direction, temperature, and roadside distance sensors
connect to Product Control Modules ( PCM). PLC controlled spray valves interface with a
Switch Sense Module. A radar measuring vehicle speed connects to a Power Speed Module. A
Global Positioning System ( GPS) connects to an operator interface console module. The PLC
spray controller connects to the CAN through a Gateway Module. Most of the modules in the
system are Mid- Tech products and are described further in the Legacy 6000 Fieldware Users
Guide. Connection diagrams are pictured on pages 20 and 21.
Each PCM setup and calibration is critical for the PLC Lockout line to work properly and for
data to be logged correctly. Although repeated setup and calibration procedures are not required
for daily operation, a condition may arise that would require a reentry of these values. Detailed
step by step processes are outlined in the PCM Setup and PCM Calibration sections of this
manual.
Spray Configuration
Spray Configurations have been created in the RMS Office program and imported into the
Legacy 6000. Three spray configurations have been loaded for truck operation. " Driver Side" is
a spray configuration that has valve configurations set to map while spraying on the driver's side,
" Passenger Side" is a spray configuration in which valves are configured for mapping the
passenger's side, and " Alternate Sides" is configured for mapping on either side with channel 10
indicating the side that is being sprayed ( OFF is drivers side, ON is passenger side).
67
The mapping configurations in the Legacy are to be used with the expanded spraying pattern in
the PLC. This is because the Legacy module is connected to every third valve in the bank and
will sense every trigger only if three valves are fired at a time.
The preset spray configurations may be selected in the Vehicle Setup menu in the Legacy
console. Note that when post- processing data from the " Alternate Sides" configuration, the
spray pattern must be mirrored and translated to the other side of the truck if channel 10 is
indicated ON.
See the Legacy 6000 RMS- EXT Users Guide for more detail on Spray Configuration selection.
Legacy 6000 PCM Setup
Chapter 2 of the Legacy 6000 Fieldware Users Guide outlines setup procedures for the console,
vehicle configuration, and Product Control Modules. However, specific PCM values may be
entered to ensure proper operation. Listed below are specific values to be input for PCM setup.
PCM # 1 Configuration
Favorite: Pump
Application: Liquid
Application Name: W_ SPEED
Configuration: Standard
PCM Link: None
Drive Type: No Drive
Units: Gal/ Min
Basis: Time
Primary Sensor: Pressure Analog
Input: E
Sensor Name: W_ SPEED
Cal # Basis: None
Nozzle Const: 1.35
Alarm Units: psi
Min Alarm: OFF
Max Alarm: OFF
Alarm Delay: 10s
Sensor Output: 0- 5.0V
Secondary Sensor: None
Monitor 1: None
Monitor 2: None
Monitor 3: None
Monitor 4: None
PCM # 2 Configuration
Favorite: Pump
Application: Liquid
Application Name: W_ DIR
Configuration: Standard
68
PCM Link: None
Drive Type: No Drive
Units: Gal/ Min
Basis: Time
Primary Sensor: Pressure Analog
Input: E
Sensor Name: W_ DIR
Cal # Basis: None
Nozzle Const: 1.35
Alarm Units: psi
Min Alarm: OFF
Max Alarm: OFF
Alarm Delay: 10s
Sensor Output: 0- 5.0V
Secondary Sensor: None
Monitor 1: None
Monitor 2: None
Monitor 3: None
Monitor 4: None
PCM # 3 Configuration
Favorite: Pump
Application: Liquid
Application Name: TEMP
Configuration: Standard
PCM Link: None
Drive Type: No Drive
Units: Gal/ Min
Basis: Time
Primary Sensor: Pressure Analog
Input: E
Sensor Name: TEMP
Cal # Basis: None
Nozzle Const: 1.35
Alarm Units: psi
Min Alarm: OFF
Max Alarm: OFF
Alarm Delay: 10s
Sensor Output: 0- 5.0V
Secondary Sensor: None
Monitor 1: None
Monitor 2: None
Monitor 3: None
Monitor 4: None
PCM # 4 Configuration
Favorite: Pump
Application: Liquid
69
Application Name: SLOPE
Configuration: Standard
PCM Link: None
Drive Type: No Drive
Units: Gal/ Min
Basis: Time
Primary Sensor: Pressure Analog
Input: E
Sensor Name: SLOPE
Cal # Basis: None
Nozzle Const: 1
Alarm Units: psi
Min Alarm: OFF
Max Alarm: OFF
Alarm Delay: 10s
Sensor Output: 0- 5.0V
Secondary Sensor: None
Monitor 1: None
Monitor 2: None
Monitor 3: None
Monitor 4: None
Calibration
Calibrations must be executed for the Power Speed Module and each of the Product Control
Modules before the first use and after any electrical modification of the Legacy 6000 system.
Calibrations are not required each time the truck is used or each time a new data file is generated.
To begin, simply enter the Calibration menu.
Calibration of the Power Speed Module can be executed as stated in Chapter 3 of the Legacy
6000 Fieldware Users Guide. In the initial calibration, the speed sensor had a Quick Cal number
of 770, resulting with a frequency of 580 Hz equal to a ground speed of 10 mph.
The voltage output of each analog sensor is displayed by the Legacy console as a pressure value
( psi). When the data is written to a file, each value is recorded as a flow rate ( gal/ min). In order
to simplify the post- process calculations, a specific calibration routine was formulated so that
each sensor must be calibrated to a specific pressure value and to a specific nozzle constant.
In the case of wind speed measurement, the calibrated pressure displays the numerical value of
the wind speed in miles per hour ( displayed value 25.0 psi = measured value 25.0 mph). Given
the correct nozzle constant, the flow rate recorded in a map by the Legacy 6000 is simply the
square root of the measurement ( logged value of 5.0 gal/ min = measured value 25.0 mph).
The routine is executed by first calibrating the ' pressure sensor'. When asked to relieve all
pressure from the system, simply connect the Calibration Zero Module to Input E of PCM # 1,
and then hit Enter. Then connect the 2.5V Calibration Set Module to Input E of PCM # 1 and hit
Begin Pressure Set. Pause a moment and then hit Enter. The actual value that must be entered is
33.6 psi.
70
Because nozzle constants of 1.35 were entered in each of the PCM setup procedures, flow rate
calibrations are not necessary. The nozzle constant already simplifies the post- process
calculation by equating the actual measured value ( in mph, degrees, cm, etc.) to the numerical
value of the square of flow rate.
In order to calibrate wind direction, repeat the steps conducted for wind speed. In this case, the
calibrated pressure displays the numerical value of the wind direction in degrees from the front
of the truck ( displayed value 121 psi = measured value 121 degrees from truck front). Note that
the angle measurements may wrap around to 540 degrees. To achieve these calculated values,
connect the calibration modules to Input E of PCM # 2, conduct the procedure as described for
PCM # 1, and enter an actual pressure value of 270.
To calibrate the temperature sensor ( PCM # 3), the routine is similar to those of the wind speed
sensor. Connect the Current Calibration Module to Input E of PCM # 3. Enter the pressure
sensor calibration for PCM # 3. Press enter to zero the sensor input on 4 mA. Connect the 2.5V
Calibration Module to Input E of PCM # 3. Press the set pressure button on the Legacy and enter
an actual pressure value of 25.6 psi. For this particular sensor, the numerical value of pressure is
equal to the numerical value of temperature in degrees Celsius.
The roadside distance sensors’ input ( PCM # 4) is calibrated using the same method as PCM # 1.
Connect the Calibration Zero Module to Input E of PCM # 4 and zero the input. Connect the
2.5V Calibration Module to Input E and press the set pressure button. Enter an actual pressure
value of 762.5 psi. This numerical pressure value is offset from the numerical measured value
by 50. ( 30 psi actually represents 80 cm from the nearest object, 762.5 psi actually represents
812.5 cm from the nearest object).
Setting Lockout Thresholds
The Environment Data Logging System is equipped with the ability to disable the spray valves in
the case that one of the environmental conditions falls outside some preset threshold values. In
order to set these values, a monitor and keyboard must be connected to the Matrox computer.
The Matrox is linked to the PLC/ CAN Gateway Module via serial port COM 1. Using
HyperTerminal, the operator can change the threshold values. Settings for the terminal program
include: 9600 baud, 8 data bits, no parity, 1 stop bit, and no flow control.
When connected, the following menu title will be displayed:
Spray Lockout Threshold Setup Menu
Select ' w' for wind, ' t' for temperature, or ' s' for slope
When wind is selected, the following prompt is displayed:
Old setting for maximum wind speed ( mph)
50
Enter new maximum wind speed ( mph)
71
The minimum accepted wind speed threshold is 1, the maximum accepted value is 67.
When temperature is selected, the two following prompts are shown:
Old setting for minimum temperature ( F)
40
Enter new minimum temperature ( F)
Old setting for maximum temperature ( F)
105
Enter new maximum temperature ( F)
The minimum accepted temperature threshold is 32, the maximum accepted threshold is 132.
When roadside slope is selected, the following two prompts are displayed:
Old setting for minimum roadside slope ( degrees)
- 10
Enter new minimum roadside slope ( degrees)
Old setting for maximum roadside slope ( degrees)
40
Enter new maximum roadside slope ( degrees)
The minimum value for the roadside slope thresholds is - 23 degrees, the maximum is 65 degrees.
After the threshold values are set, they are retained even during loss of power.
System Operation
Normal Operation ( without school avoidance map)
The following is a checklist of tasks to operate the Environment Data Logging System. A more
detailed discussion can be found in the Legacy 6000 RMS- EXT Users Guide.
1. Pre- spray checklist for everyday operation without school maps
• Insert Flash card in Legacy Console.
• Turn on Battery Power Switch on side of truck.
• Turn on Legacy console.
• Press the “ ARM” button on the upper right corner of the console.
• Press “ Create new job using settings from the previous job”
• Select a new job name ( easiest way is to use second button from top right)
• Enter ARM setup and select a “ Map File” name
• Press the begin ARM button on the upper right corner
72
2. Post- spray checklist and System Shutdown
• Exit ARM Operation.
• Turn off Legacy console.
• Turn off Battery Power switch on side of truck.
• Remove Flash Card from Legacy Console.
ARM parameters may be changed in the ARM setup pages. The first time a new map is to be
created, a new job must be created along with several items of information that must be input
into the Legacy file. Unless there has been an electrical change in the system, there should be no
reason to run a Calibration. Additionally, no fluid is controlled by the Legacy so there is no need
to Prime or Agitate.
The remaining three menus must be completed before data can be mapped. Listed below are
parameters for two of the three menus. Critical values are underlined.
Product Setup
W_ SPEED
In Use: Yes
Product: Wind Speed
Correction Factor: 1.0
Initial Quantity: 0.0
W_ DIR
In Use: Yes
Product: Wind Direction
Correction Factor: 1.0
Initial Quantity: 0.0
TEMP
In Use: Yes
Product: Temperature
Correction Factor: 1.0
Initial Quantity: 0.0
SLOPE
In Use: Yes
Product: Roadside Slope
Correction Factor: 1.0
Initial Quantity: 0.0
ARM Setup
Map File: File Name ( Any name is OK, just don’t leave blank)
Base Map File: None
Collection Interval: 1s
Alarm: Off
Range: 1000 ft
Slope Inc/ Dec: 15 degrees
Speed Source: Radar
GSO Speed: 0.1 MPH
73
As stated in the checklist above, after these values have been entered for the first time, a new file
can be created by simply pressing the button “ create new file using settings from the previous
file”. However, the map file name, in the ARM setup, must be added for each new job that is
created.
Hazard Operation ( with school avoidance map)
The Legacy system is capable of alarming the driver if the truck is within 1000 ft of a “ hazard”.
The locations of the hazards must be pre- defined in a road markers map. For use with the IHAS
truck, hazard markers are placed around the perimeter of schools in a road- markers map. This
map file is then used by the Legacy system to alarm the driver when the spray truck is too close
to a school. The driver must then shut off spray until there is an acceptable distance between the
truck and the school ( the alarm will stop). Note that the truck will not automatically turn the
spray off. It is the responsibility of the driver to act when the alarm sounds.
The following is a checklist of tasks to operate the Environment Data Logging System with a
school avoidance map. When a new job is created in the Legacy console, there is no method to
transfer a predefined school map to the new job. Thus, this file transfer must be done on a PC
before the flash card is inserted into the Legacy Console.
1. Pre- spray checklist for operation with school maps
• On a PC, create all new job folders required for the day on the Legacy Flash Card.
• On a PC, transfer the school map file and other ARM setup files to each of the new job
folders.
• Insert Flash card in Legacy Console.
• Turn on Battery Power Switch on side of truck.
• Turn on Legacy console.
• Press the “ ARM” button on the upper right corner of the console.
• Select a new job that was created on the PC
• Press the begin ARM button on the upper right corner
2. Post- spray checklist and System Shutdown
• Exit ARM Operation.
• Turn off Legacy console.
• Turn off Battery Power switch on side of truck.
• Remove Flash Card from Legacy Console.
Data Conversion
After collected with the Legacy system, the data may be processed with the Mid- Tech RMS
Office software and later processed with ArcView, SSToolbox, or some other GIS package.
Because the Legacy system will only log data from the Product Control Modules ( PCM’s) in one
unit: flow rate, every measurement made is recorded in units of gallons per minute. However,
74
with a few simple conversions, the recorded numerical values can be converted back to the
measured values with the desired units.
Wind Velocity
In order to calculate the absolute wind velocity, collected data must be used to determine wind
velocity relative to the truck and the truck velocity relative to the ground. The following
equations may be used to calculate absolute wind velocity in x- y coordinates where positive y is
north and positive x is east.
Relative wind velocity may be determined by finding wind speed and wind direction relative to
the truck. Measured values of relative speed and direction are:
2
1 1 s = Q
2
2 ! = Q
where: s1 is relative wind speed ( mph), φ is the direction from which the wind is blowing relative
to the front of the truck ( degrees), Q1 is the recorded flow rate from PCM # 1, and Q2 is the
recorded flow rate from PCM # 2.
The resulting vector is:
v s ( sin( ! ) x cos( ! ) y)
1 1 = " +
Truck ground displacement may be determined by the change in latitude and longitude
coordinates from adjacent GPS samples.
( )( ) 2 1 " y = N + h lat ! lat
( )( ) cos( ) 2 1 1
" x = N + h lon ! lon lat
where: N is the radius of the earth ( miles), h is the altitude above sea level ( miles), lat1 and lon1
are the latitude and longitude of the first point ( radians), and lat2 and lon2 are the latitude and
longitude of the second point ( radians). ( Note that longitude values in the Western Hemisphere
must be negative).
Truck velocity is the resulting vector of:
t
x y
v
! + !
= 2
where: v2 is the resulting velocity vector and t is the time interval between GPS data points
( hours). Adding the vectors of truck velocity and relative wind velocity yields absolute wind
velocity.
75
1 2
v = v + v
where v is the absolute wind velocity in mph.
Temperature
Ambient temperature can be calculated easily from the recorded flow rate value from PCM # 3.
The basic equation is:
2
3 T = Q
where: T is the ambient temperature ( degrees C) and Q3 is the numerical flow rate from PCM # 3
( gal/ min).
Roadside Slope
Roadside slope can be determined from distance measurements made to the nearest point from
ultrasonic distance sensors on the sides of the truck. The distance measurement is calculated as:
50
2
4 d = Q +
where: d is the distance from the sensor to the nearest roadside point ( cm) and Q3 is the
numerical flow rate from PCM # 4 ( gal/ min).
The angle of the roadside slope 􀀀 may be calculated as:
! "
#
$ %
&
'
( '
=
cos 40
188 sin 40
arctan
d
d
)
where, 188 is the height of the ultrasonic sensor ( cm), and 40 is the angle of the sensor relative to
the ground plane ( degrees). Note that angles defining roadside slope calculation are given in
figure D1.
76
Figure D1. Diagram of spray vehicle with roadside slope geometry.
Wire Diagrams
Figures D2- D5 show electrical connections for the IHAS Data Logging System. The number
values correspond to the numbering system defined in the PLC documentation. Example values
include ( For more information on the Mid- Tech Legacy wiring, consult the Legacy 6000
Fieldware Users Guide). :
000 = Truck Chassis
001 = PLC Ground
003 = Truck Battery Ground
014 = Truck Battery Power ( 13.8V)
024 = PLC Power ( 24V)
300- 331 = Valve Control Lines
400 = Ground Speed Signal Output
77
Figure D2. Wiring diagrams for Legacy CAN.
78
Figure D3. Wiring diagrams for interfacing Legacy to PLC.
79
Figure D4. Wiring diagram for Raven splitter.
80
Figure D5. Wiring diagram for Legacy connections to PLC control box.
81
Appendix E: Sensor Specifications
ALL WEATHER SENSING TECHNOLOGY
WindSonic
Wind Speed & Direction Sensor
LOW START SPEED
CORROSION FREE, UV STABLE MATERIAL
NO CALIBRATION REQUIRED
ROBUST CONSTRUCTION
TRUE 0- 359 º OPERATION ( no dead band)
WIND SPEED & DIRECTION FROM A
SINGLE UNIT






w w w . g i l l . c o . u k WINDSONIC
AGRICULTURE
HVAC
POLLUTION CONTROL
PORTABLE WEATHER STATIONS
ROADSIDE WEATHER STATIONS
TUNNELS
MARINE







MAINTENANCE FREE - 2 YEAR WARRANTY
GILL INSTRUMENTS LTD
Saltmarsh Park, 67 Gosport Street,
Lymington, Hampshire, SO41 9EG, UK
Tel: + 44 ( 0) 1590 613500
Fax: + 44 ( 0) 1590 613555
E- mail: anem@ gill. co. uk
Website: www. gill. co. uk
© Gill Instruments 2005
The WindSonic is part of the Solent range of ultrasonic
anemometers. The range is in continuous development and therefore
specifications may be subject to change without prior notice.
At last, a real low cost alternative to conventional cup/ vane/
propeller wind sensors in a single unit - WindSonic from Gill
Instruments. Utilising our expertise as the world's leading
sonic manufacturer, WindSonic is based on our existing, highly
successful, proven ultrasonic technology. Ideal for applications
that demand economic wind sensing, WindSonic is suitable for
land- based and marine environments.
A lightweight unit, WindSonic is of a robust, high strength
construction designed to withstand installation and use with no
fear of the damage commonly experienced with more fragile
cups, vanes or propellers. Without the need for expensive
on- site calibration or maintenance and with a corrosion free
exterior, WindSonic is a true fit and forget unit.
The flexible design enables you to easily configure WindSonic
to deliver the information you require. By using the software
provided it is possible to select the output rate and choose
the units of measurement that suit your application. Ensuring
accuracy and reliability, WindSonic automatically transmits
an anemometer status code with each output to indicate its
operating status. Available in three options, providing a number
of different digital and analogue outputs.
Maintenance free, quick and easy to install, WindSonic is
designed to be mounted using a standard pole fitting and
comes complete with all screw fittings, a mating marine grade
connector and comprehensive user manual.
The unit is supplied with a 2 year warranty as standard.
WINDSONIC - ULTRASONIC WIND SENSOR
CUSTOMER SELECTABLE
Output 1, 2 or 4 outputs per second
Parameters Wind Speed & Direction or
U and V ( vectors)
Units of Measure m/ s, knots, mph, kph, ft/ min
WIND SPEED
Range 0 – 60 m/ s ( 116 knots)
Accuracy +/- 2%
Resolution 0.01 m/ s ( 0.02 knots)
WIND DIRECTION
Range 0 to 359° – no dead band
Accuracy +/- 3°
Resolution 1°
ANEMOMETER STATUS
Message supplied as part of standard output
POWER REQUIREMENT
Anemometer 9- 30Vdc @ 14.5mA typical
Start up time < 1 second
OUTPUTS
Option 1 RS232
Option 2 RS232 + RS422 + RS485 + NMEA*
Option 3 RS232 + RS422 + RS485 + NMEA*
+ 0- 5V or 4- 20mA
Option 4 SDI- 12 + RS232
* NMEA 0183 Version 3
ENVIRONMENTAL
Ingress Protection IP65
Operating Temperature - 35° C to + 70° C
Storage Temperature - 40° C to + 90° C
Operating Humidity < 5% to 100%
EMC EN 61000- 6- 2 : 2001
EN 61000- 6- 3 : 2001
MTBF
15 years
MATERIALS
External Construction LURAN S KR 2861/ 1C ASA/ PC
DIMENSIONS
Size 142 x 160 mm
Weight 0.45 kg
WARRANTY
2 years
OPTIONAL FACTORY CALIBRATION
Traceable to national standards
ACCESSORIES
Pipe Mounting 44.45 mm ( 1.75 in) diameter
WindCom - Display & logging software *
Cables
Display
* download WindCom free from www. gill. co. uk
D300
ENERCORP instruments ltd
25 Shorncliffe Rd, Toronto, ON, M9B 3S4 Tel 1( 800) ENERCORP or ( 416) 231- 5335 Fax 1( 877) ENERCORP or ( 416) 231- 7662
Visit our on- line catalogue at www. enercorp. com our e- mail address is info@ enercorp. com
- 22-
• Accurate Platinum RTD
• Rugged construction
• Splashproof
ORDERING DATA
TS- GPS- R- 100- ( ) - ( ) - ( )
stem connection temp( C)
inches 4 = 1/ 4" NPT 400
8 = 1/ 2" NPT 600
e. g. TS- GPS- R- 100- 6- 4- 400 general purpose industrial
probe with small head and 100 ohm RTD, 6" long stem and
1/ 4" NPT process thread rated for 400C operation.
General purpose small version
This is the all purpose
model for general light duty
industrial or commercial
temperature measurement.
It features a small threaded
cast aluminum head with
O- ring seal. Electrical
connections are made
through the cable gland
with rubber grommet.
The standard process
connector is 1/ 4", although
1/ 2" NPT is available if
specified at order time. The
sheath is 1/ 4" O. D. stainless
steel. The sheath length
must be specified at order
time, although 4" and 6"
lengths are normally
available from stock.
The assembly is rated for
measuring temperatures up to 200C and uses a thin
film RTD sensor to DIN 43 760 or IEC 751.
• Accurate Platinum RTD
• Rugged construction
• Splashproof
General purpose large version
This is the all purpose model for general heavy duty
industrial or commercial
temperature measurement.
It features a large threaded
cast aluminum head with
gasket. Electrical
connections are made
through the 3/ 4" NPT female
opening suitable for piping or
standard electrical fittings.
The standard process connector is 1/ 2",
although 3/ 4" NPT is available if
specified at order time. The standard
sheath is 1/ 4" O. D. stainless steel and
other sizes are available to special order.
The sheath length must be specified at
order time.
The standard assembly is rated for
measuring temperatures up to 200C.
400C and 600C versions are available to
special order. We use a thin film RTD
sensor to DIN 43 760 or IEC 751 or wire
wound if requested.
ORDERING DATA
TS- GPL- R- 100- ( ) - ( ) - ( )
stem connection temp( C)
inches 8 = 1/ 2" NPT 400
12 = 3/ 4" NPT 600
e. g. TS- GPL- R- 100- 4- 8 general purpose industrial probe
with large head and 100 ohm RTD, 4" long stem and 1/ 2"
NPT process thread rated for standard 200C operation.
GENERAL PURPOSE INDUSTRIAL
970220
TEMPERATURE
UM 30 Ultrasonic sensor
12 SENSICK © SICK AG · Industrial Sensors · Germany · All rights reserved 8 010 312/ 12- 03- 03
High measurement accuracy
thanks to time- of- flight
measurement
Independent of material shape
( including films, glass and bottles)
Teach- in
Insensitive to dirt, dust and fog
Operating scanning range up
to 6,000 mm
Binary outputs or analog output
ø 65
5 5
138.5
M 30x1.5
48
Fastening nuts, width across 36 mm
Connection plug M 12
Control and display panel
Setting key 2
Setting key 1
2
1
1
L+
Q, Q
NC
4
3
2
M
brn
blk
blu
5
NC gra
wht
1
L+
Q , Q
Q , Q
4
3
2
M
brn
blk
blu
5
NC gra
wht
2 2
1 1
1
L+
NC
Q
4
3
2
M
brn
blk
blu
5
NC gra
wht
A
Operating
scanning range
800 . .. 6000 mm
Ultrasonic sensor
UM 30- 15111
All types
Connection types
Adjustments possible
Mounting systems
Dimensional drawing
Accessories
1
2
3
4
5
6
7
UM 30- 15112 UM 30- 15113
5- pin, M 12 5- pin, M 12 5- pin, M 12
P2 D2 D1 P1
3
4 5
6 7
1
2
3
4 Limiting scanning range
3 Operating scanning range
2 Pipe diameter 27 mm
1 Aligned plate 500 x 500 mm
8 010 312/ 12- 03- 03 © SICK AG · Industrial Sensors · Germany · All rights reserved SENSICK 13
UM 30
Operating scanning range
( limiting scanning range) 800 ... 6000 mm ( 8000)
Ultrasonic frequency 80 kHz
Resolution 1 mm
Reproducibility ± 0.15 % of final value
Accuracy ≤ 2 % of final value
Supply voltage VS 12 ... 30 V DC 1)
Ripple ± 10 %
Current consumption 2) ≤ 70 mA
Switching outputs, reversible3) Q: PNP, VS – 2 V, Imax = 500 mA
Q1, Q2: 2 x PNP, VS – 2 V, Imax = 500 mA
Analog output, reversible3) 4) QA: 4 ... 20 mA/ 0 ... 10 V
Response time 5) 240 ms
Switching frequency 2/ s
Switching hysteresis 100 mm
Standby delay 2 s
Connection type Plug M 12, 5- pin
Enclosure rating IP 65
Ambient temperature 6) Operation – 20 ° C ... + 70 ° C
Storage – 40 ° C ... + 85 ° C
Weight 360 g
Housing material Nickel- plated brass
1) Limit values
2) Without load
3) Outputs short- circuit protected
4) Automatic switching between voltage
and current outputs dependent on load
Current output 4 ... 20 mA:
RL ≤ 500 Ω, VS ≥ 20 V;
RL ≤ 100 Ω, VS ≥ 12 V
Voltage output 0 ... 10 V:
RL ≥ 100 kΩ; VS > 15 V
5) Only with UM 30-_ _ _ _ 3: Recovery time
according to EMV EN 50 319
6) Temperature compensation
at – 20 ... + 50 ° C
Technical data UM 30- 15111 15112 15113
Detection ranges
Type
UM 30- 15111
UM 30- 15112
UM 30- 15113
Part no.
6 025 659
6 025 664
6 025 669
Order information
2400 1200 0 1200 2400
[ mm]
1200
2400
3600
4800
6000
7200
8400
9600
UM 30- 1511_
6000 mm
1
2
3
4
86
Appendix F: IHAS Direct Nozzle Injection Research Study
Transactions of the ASABE
Vol. 49( 4): 865− 873  2006 American Society of Agricultural and Biological Engineers ISSN 0001− 2351 865
DIRECT NOZZLE INJECTION OF PESTICIDE CONCENTRATE INTO
CONTINUOUS FLOW FOR INTERMITTENT SPRAY APPLICATIONS
D. Downey, T. G. Crowe, D. K. Giles, D. C. Slaughter
ABSTRACT. A direct nozzle injection system was developed to intermittently inject concentrated solutions into continuous
carrier liquid flow through a straight− stream spray nozzle used for targeted roadside spraying of post− emergent herbicide
during pre− emergent herbicide application. The injection system was based on a 12 VDC direct− acting electrical solenoid
valve with a 0.56 mm valve orifice and metering plate with a 0.2 mm diameter orifice. A conductivity− based sensor was used
to measure the instantaneous concentration of NaCl tracer simulating a pesticide solution. Injection pulse durations ranged
from 10 to 100 ms into carrier flows of 1.5 and 2.6 L/ min from 1.55 and 2.18 mm nozzle orifice diameters, respectively. Lag
times between initiation of the injection valve actuation and emission of the concentrate material from the spray nozzle were
on the order of 25 ms. Concentrations of 1% ( v/ v) from injected solution into the flow emitted from the nozzle could be achieved
within 100 ms after valve actuation. Increasing the injection pulse duration did not reduce lag time nor increase the temporal
rate of concentration increase in the emitted spray; however, increasing the injection pressure increased the rate of
concentration increase. Analysis of the injection event, using standard mixing criteria, determined that the injection mixing
events were Gaussian in nature and did not represent ideal plug flow or short− circuiting events. For an intermittent,
target− detecting system with a detector− to− nozzle distance of 4 m and a ground speed of 5 m/ s, the direct nozzle injection
system is a feasible configuration for spot spraying if the sum of detection time and time of flight for the emitted spray is less
than 800 ms. For a prototype machine vision− based roadside sprayer, detection and spray flight times were less than 67 and
400 ms, respectively; therefore, feasibility of spot spraying using at− nozzle injection was established.
Keywords. Injection, Microliter, Nozzle, Orifice plate, Pesticide, Precision agriculture, Spray.
n arid climates, roadside weed control reduces fire haz-ards,
decreases weed seed production, and improves
driver visibility; the operational goal is removal of all
vegetation in a narrow strip adjacent to a road or free-way.
Targeted application of foliar− active pesticides is a