Tillett and Hague Past Projects

Autonomous Crop Treatment Vehicle (1993 - 1996)

Computer vision based, guidance system for non-chemical weed control (1997 - 2001)

Weed mapping in agricultural crops and on railway lines (2003 - 2004)

High work rate vision guidance spanning multiple independently drilled bouts (2002 - 2005)

Mechanical weed control for integrated and organic salad and brassica production (In-row weeding) (2005 - 2007)

Intelligent thinning for direct drilled salads and vegetables using machine vision (2007 - 2008)

Tillett and Hague Home


Autonomous Crop Treatment Vehicle (1993-1996)

One way to reduce agrochemical use is through spatially selective operation; a map of the variability within a field is used to decide how to selectively treat areas of typically 5 m by 5 m resolution from a tractor equipped with GPS. We proposed an alternative but complementary approach based upon real time detection of targets, be they crop or weed. This allowed a much finer resolution, down to individual plants, and required little prior knowledge of the field except an estimate of planting geometry. We named this concept plant scale husbandry, and chose as an example task the selective application of chemical to a transplanted cauliflower crop.

Highly targeted individual plant scale operations may be too slow for manned vehicles. We therefore developed an autonomous vehicle as a platform for experimental plant scale operation. A treatment device consisted of an array of solenoid operated nozzles at 50 mm pitch. Selective treatment was achieved by switching each individual nozzle on or off as the vehicle progressed through the crop.

Autonomous spot treatment vehicle

The planting pattern of the crop, imaged by a vehicle mounted camera, was tracked using an extended Kalman filter. To allow for reliable navigation, including headland turns, the Kalman filter also integrated odometric and inertial sensor data. In addition to providing guidance information, tracking the crop pattern allowed the discrimination of crop - which matched the planting pattern - from weeds which did not.

Sementaion of cauli and weeds

As the vehicle progressed along the row, the results of crop / weed classification were accumulated over several views of each area of ground. The local treatment map constructed was used to select spray nozzles to give selective treatment - in this example, crop plants were treated whilst avoiding weeds.
Traeted Cauli


This project was funded by The Douglas Bomford Trust and the BBSRC.  Click on this link to see a 35 second 3MB Windows Media File format video of the autonomous vehicle in action.

Top of page

Computer vision based, guidance system for non-chemical weed control (1997-2001)

This project was concerned with reducing agrochemical inputs through the use of inter-row cultivation to control weeds. This practice goes back to Jethro Tull's invention of the seed drill in the 18th Century. However, the advent of herbicides all but eliminated inter-row hoeing in the UK. Our objective was to reinvigorate this environmentally friendly technique and make it commercially viable again by the application of modern robotics technology. Defra funded an investigation into the potential for use in cereals and with support from the British Beet Research Organisation we extended the research to cover sugar beet. Within this project we collaborated with ADAS, IACR, Garford Farm Machinery, Robydome Electronics and KRM Ltd.

The principle technical problem was to locate and track the crop rows such that cultivators could be brought into close proximity to crop plants, thus maximising weed kill, without reducing yield. We chose to develop a non-contact sensing system using low cost CCD video cameras and computers. This approach had the advantage of being very flexible in respect of alternative crops and growth stages. However, in agricultural fields we were faced with a number of challenges. These included naturally variable lighting, crop emergence that may be incomplete and weeds that tend to mask crop rows.

Our approach had two principle innovative steps. The first was in the initial image processing in which we exploit prior knowledge of crop row spacing to identify crop rows. The second was in the application of a tracking algorithm to track row location between images. Together these give very reliable performance under a wide range of conditions typically placing cultivators well within +/-3cm at speeds of up to 10kph and beyond. This performance has proved commercially attractive. Working with Garford Farm Machinery and Robydome Electronics we brought a product to market in 2001 under the brand name Robocrop

Experimental 4m cereal inter-row cultivatorExperimental 6m Sugar beet inter-row cultivator

Top of page

Weed mapping in agricultural crops and on railway lines (2003 - 2004)
Weeds segmented from ballast and rails

Weeds present a significant maintenance issue on railway lines.  Normal control methods are based on a uniform application of herbicide even though weeds may be patchy, thus putting more herbicide into the environment than is strictly necessary.  Our vision technology can be used to identify weed patches so that a precise weed map can be created either to guide future spray applications, or control spray nozzles in real time.
Automatically generated map of weeds in wheat
We used our ability to accurately locate crop rows to deduce weed density on the basis that green plant material between rows will be weed.  This technique accurately detects weed density at early stages of crop growth when rows are discrete.  We have used similar techniques to successfully assess crop density.  The illustration above is of an automatically generated weed map in a field of winter wheat grown on a 25cm row spacing.  The technique could be applied to a wide range of agricultural and horticultural crops and linked to either weed map generation or real time sprayer control.
Top of page

High work rate vision guidance spanning multiple independently drilled bouts (2002-2005)

Inter-row cultivation and band spraying improve efficiency of conventional and organic systems alike.  However, they are sometimes difficult to implement due to labour and cost constraints.  Vision guidance has improved this situation, but work rate is still limited due to the need to span a single drill or transplanter bout widths.  In this project we set out to further improve the generic technology for precision vision guidance and to develop new technologies for spanning multiple drill/planter bouts.  We chose to demonstrate the technology on two machines.  The first was a specially constructed 12m wide inter-row hoe spanning three cereal drill bouts.  The second was a 20m wide precision band sprayer spanning five planter bouts based on an adapted trailed sprayer.

The experimental hoe was shown to perform well in a wide range of conditions at speeds of up to 10kph with high accuracy (S.D 10mm).  An investigation into combined hoeing and on-row application of selective herbicides and other agro-chemicals suggests that the potential environmental and economic advantages are substantial.  However, the required very narrow (9cm) spray bands over 25cm cereal rows are difficult to achieve with existing hydraulic nozzle technology due to the very low (10L/h per nozzle) volumes desired.  Alternative spray technologies are available and further work is required to assess their performance in this application.  Click on this link to see a 34 second 5MB Windows Media File format video of the 12m multi-section inter-row cultivator in action.
12m experimental inter-row hoe with tank for combined band spraying

Experimental evaluation of vegetable band spraying indicated that accuracy targets (S.D 25mm at 12kph) were met, providing a cost effective system for reducing chemical inputs. Click on this link to see a 22 second 4MB Windows Media File format video of the 20m vision guided sprayer in action.
20m Vision guided band sprayer with 5 4m sections

Partners
Silsoe Research Institute, HGCA, Garford Farm Machinery, Robydome Electronics, AGCO Ltd, Micron Sprayers, Unilever, Sheepdrove Organic Farm, Robert Montgomery Ltd, Abacus Organic Associates, The Allerton Research & Educational Trust

Government Sponsor
Defra (LK0928)

Top of page

Mechanical weed control for integrated and organic salad and brassica production  (2005-2007)

Diminishing herbicide options, fear of ground water contamination and customer pressure to minimise herbicide use are all pushing the horticultural industry away from reliance on herbicides.  However, product contamination concerns, much of which relate to weeds, necessitate high levels of weed control and have resulted in increasing use of unsustainable hand weeding.

Mixed green and red lettuce with weeds

Weeds growing within crop rows are the major problem because they are missed by conventional inter-row mechanical weeding. THT’s imaging and crop row tracking technology applied to inter-row mechanical cultivators improves the situation by reducing the uncultivated within row band through more accurate guidance than can be achieved manually.  However, it is inevitable that some weeds are missed.  In this project we have developed an adaptable, cost-effective technology for mechanically controlling in-row weeds for a wide range of brassica and salad crops.

Our scientific approach to the project was to develop a fast, two dimensional mathematical template matching techniques, (exploiting periodicity within the planting grid), enabling individual crop plants with some spacing variability to be located.  Regular observations of plant position are passed to a tracking algorithm that can follow plant location from a moving vehicle.

Each plant is individually tracked to cope with variability

A novel shallow cultivation mechanism was also developed. This is synchronised with the plant tracking algorithm enabling weeds to be removed from between crop plants leaving the crop undisturbed.

Sequence of images showing cut out disc rotating about a plant

The sequence of images above shows an experimental cultivator operating above ground on an artificial crop to demonstrate the principle of operation

An experimental implement was constructed based on a standard front mounted Garford vision guided inter-row steerage hoe.  A camera was mounted centrally on the implement looking ahead and down such that the full width of the bed was visible over a length of approximately 2.5m.  Each hydraulically driven rotating disc cultivator was attached to a depth wheel unit mounted on the implement so that cultivation depth could be consistently maintained (normally at approximately 20mm).  The experimental implement had the capacity to take up to five rotating disc units.

In-row weeder in lettuce   In-row cultivation of cabage

Results showed that the machine was effective at removing weeds within the row without removing crop plants.  A field trial has shown that a single pass of the machine incorporating intra and inter-row cultivation removes 80% of weeds.

Untreated area (to left) and treated area (to right) two weeks after treatment showing effectiveness

This project was sponsored by Defra under the Horticulture Link Programme and was conducted jointly with Warwick HRI who provided the weed science and agronomy expertise. The project was part funded by the  Horticultural Development Council and had the following additional industrial partners: Robydome Limited, Garford Farm Machinery , Edwards Brothers, Robert Montgomery Limited, Allium & Brassica Centre , AGCO

Four commercial within-row weeding machines based on this technology were sold to UK growers by Garfords at the start of the  2008 season.  More information on the commercial machines is available on the Robocrop Vision Guidance page.  Click on this link to see a 25 second, 6 MB Windows Media File format video a commercial within-row weeding machine in action

Top of page

 Intelligent thinning for direct drilled salads and vegetables using machine vision (2007 - 2008)

 Due to field variability in the process of drilling and germinating seed some high value row crops are sown at a higher density (typically x3) than that which is ultimately desired, and then post emergence, mechanically thinned to the required stand.  The selective thinning process has the objective of leaving the healthiest plants at as close as possible to the optimum spacing.  The intelligence required to make this selection coupled with the accurate mechanical control required has ensured that this process is almost always conducted manually using hand hoes.  There is worldwide interest in reducing difficult repetitive field work due to labour availability, cost and health concerns.

Automatic thinner with manual thinning gang in background

The aim of this project is to develop an intelligent thinning technology that uses an ability to look ahead and automatically make selection decisions to maximise crop quality and yield.  The project is utilising much of the technology developed in the within row weeding project described above.  Unwanted plants are removed using a selective thinning disc similar to that used for removing weeds between transplanted crops.  The innovation specific to the thinning application lies mainly in detecting the very small closely spaced seedlings and developing algorithms to automatically decide which plants to remove and which to keep.  There are also mechanical challenges relating to the variability in plant spacing that occurs through this process.
Thinning disk in action removing unwanted lettuce plants

Results from initial experimental work funded by the Douglas Bomford Trust in 2007 were encouraging and so we have continued development.. That follow on work has included field testing in California with the assistance of Robocrop importer Solex Corporation  Thinning software has been loaded onto two within-row weeders and tested in lettuce grown on 38" and 80" bed configurations.  The photograph below show the effect of running the automatic thinner on lettuce drilled at a nominal 3.5" spacing set to produce a stand at 11.5".  The area to the left is unthinned and the area to the right has been thinned.  The final spacing is generally regarded as at least as good as that achieved manually, though the machine sometimes has difficulty removing plants ("doubles") that have grown closer (<2.5") to each other than intended.  This problem would not occur often with a well set up drill, though it is accepted that drilling sometimes has to be conducted under less than ideal conditions and so work is continuing to improve performance in this respect.

Lettuce before automatic thinning on left and after on right

A 22 second, 5 MB Windows Media File format video of automatic thinners being trialled on commercial lettuce crops in California can be viewed by clicking here.

 Top of page