Results from recent trials of a spot sprayer developed by Colin Taylor, a young Scottish engineer, show chemical savings of 77% and savings of £46.50/ha, compared with a conventional treatment across the whole area.
The idea to first develop the spot sprayer came from a discussion with a neighbour about the high cost and waste involved in spraying dock weeds in grassland. Indeed, his recent trials showed the weeds accounted for just 19% of the ground cover.
As well as reducing chemical use and costs, spot spraying also increases grass yields, because it is not sprayed and does not have the weeds to compete against it. Moreover, any clover, which is particularly sensitive to the herbicides, is not killed and therefore doesn’t need to be reseeded.
Colin developed the system in his own time using personal savings, without any outside investment. Called the RUMEX system, it employs a combination of computer vision cameras, artificial intelligence software and individual nozzle control.
His computer vision system takes up to 30 pictures every second, which are analysed by the AI software – basically this checks whether it is seeing a weed or not a weed. “We needed the software to be able to recognise docks; and this involved training it using 20,000 different images of docks and grass.
His prototype is fitted to a Hardi 12m wide, 600 litre sprayer, for which he has developed his own electronic control system for the solenoid-operated, individual nozzle control. He fits Hypro E80-30 nozzles because with the 80° fan angle these do not require an overlap to apply a full dose.
With the current 30 frames/sec vision system it allows the sprayer to work comfortably at about6-8km/h, spotting and treating the docks in the field.
“I appreciate that is about half the speed of a conventional sprayer, but it’s producing enormous savings. A normal treatment of 2 litres/active/ha will cost about £70/ha. But if you look a grass field the docks will account for just a percentage of ground cover,” he explains.
Trials show 77% savings
Trials, carried out by Colin this spring on a dairy farm near Lanark in Scotland, confirm is original projections, producing a chemical saving of 77%, compared with a conventional treatment:
|Spot treatment application data|
|Chemical used||20.2 litres|
|Equivalent conventional rate||0.45 litre/ha|
|Spot treatment chemical saving||77%|
|Average weed cover||19%|
Chemical cost savings
As well as the considerable saving in costs from using less chemical, Colin adds there will be a financial benefit from not killing the clover or stunting the grass growth with the selective herbicide. There is also a huge environmental benefit from simply reducing the amount of chemical being applied.
|Spot treatment chemical savings |
(Forefront T @ £30/litre)
|Chemical/Application rate: |
2 litres active/200 litres water
|Conventional application||2 litre/ha||£30/litre||£60/ha|
|Spot treatment||0.45 litre/ha||£30/litre||£13.50/ha|
|Savings||1.55 litres/ha||£30/litre||£46.50/ha (77%)|
“This data is from early trials and testing,” says Colin. “But the figures are a good representation of conditions, although they could change are the testing continues.”
Nevertheless, the trials also showed the weed identification performance was ‘excellent’. “The system identifies a wide range of weed varieties, sizes and shapes as well as from various angles – for example, small docks underneath dense grass swards,” he adds.
While it is ‘trained’ to spot docks, the system also identifies other weeds including thistles, nettles, ragwort and dandelions, which it treated where they were present in the recent trials.
“An important point is the identification system has developed its own level of ‘reasoning’, adds Colin. “The training data cannot possibly have represented every weed and its endless variations in size, shape, colour, presentation etc. Yet, during these trials the identification system has undoubtedly encountered weeds that it has never seen before. It has ‘learnt’ what weeds look like and has positively identified weeds, previously unseen, as weeds. It has done so accurately.”