The Andweeder

March 2015

Automated precision weeding for squash and other crops developed by Andy Lysaght

Prompted by farmer Gareth Holder, Hawke’s Bay’s Andy Lysaght invented a machine to weed squash mechanically, saving money and giving more accuracy than hand-weeding.

Andy has won two major national awards with his Andweeder, designed to eliminate hand weeding of crops. They are the Ravensdown innovation award at the national horticultural field day in Hastings and the Launch NZ innovation award at the national fieldays at Mystery Creek.

Fieldays judges said the Andweeder brings a step change to the industry, converting an intensive manual process to an automated and precision process. In doing so, its solved a long-standing problem that has bamboozled hi-tech machine vision technology.

The Andweeder uses touch triggered mechanics to weed among squash. A 20ha trial where 10ha was weeded by the Andweeder and the other 10ha was manually weeded, saw the three-row machine take 6 hours 45minutes, compared with 125 hours by hand.

Andy, an agricultural contractor was encouraged by Gareth, who was sick of having to hand weed his squash crops, to design something mechanical which would do the job better. “You are pretty good at knocking things up,” he said. About a week later Andy came back to Gareth, and said: “Come and play with this.” Gareth says Andy does a lot of thinking about things, and when he sees a problem, he thinks about how he can solve it.

Andy says: “Chemicals aren’t the answer. We are doing our best to get away from chemicals. We can’t keep doing what we are doing and the rest of the world doesn’t want us to keep doing it either. Hand labour is expensive.” He says he hates wasting energy so looks for ways to make things easier. Eight years before, he had looked at using hydraulics to solve the weed problem in squash, but that didn’t work, so he forgot about it.

He started from first principles and trialled the touch-triggered mechanical idea, taking advice from an auto electrician. “Touch triggered mechanics works the same way as a hair trigger operates on a rifle. As soon as it touches the plant, it sends a signal to an electric over-air activation system to pull the weed knives away from the plant. There are no moving parts apart from the mechanics.”

“I pulled a test unit down the row with a rope. I was the horse, and Gareth was holding the camera and following.”

Then they developed the weeder for a machine. When Andy took the idea to Mystery Creek, judges told him it was a nice idea, but to carry on working on it.

“We came back from that and put the unit on a test bed and ran it 24 hours a day to see how long it would last. It worked for 11 days. At three times a second, that was 3,200,000 clicks as a minimum. Each row could cover 270ha, so the machine could handle 760ha based on the current row spacing in Hawke’s Bay. Then we pulled our air rams apart, checked they were in good condition, put them back together and carried on experimenting.”

Now they have the original prototype, a single row weeder and a three row weeder.

“We are going back to a simple, basic manual system. The new generation of designers haven’t done work in the field and got their hands dirty or seen how a crop grows. I’ve got 45 years of accumulated knowledge. You have to look, learn and listen, and understand how plants grow and how machinery reacts to plants. We have solved how to get around the plant at speed and do a good job.”

Gareth provides agronomic advice about the practicability of ideas in the field. He says: “We have found which crops it works on and which it doesn’t.”

With squash they have measured a lower than 2% mortality rate, compared to a hand weeding rate of 5-10% mortality. “We reckon if you get it above 3% mortality you have a problem and you have to get off the tractor seat and sort it out. If you have 3% damage, something is not quite set up right.”

Andy is charging out the machine at $225/ha as a contractor. The rest of the year he works full time as a contractor and says this work is “just another string in my bow”.

Gareth says his view is that production costs for cropping will continue to increase, particularly labour charges. He thinks the scope for the Andweeder is huge, not just in New Zealand but also in Australia, where hourly labour rates are higher than in New Zealand.

“Being able to reduce the labour management and safety issues around having a crew of 15 to 20 people working with push hoes, to potentially three people doing the same job in a more consistent way with the Andweeder, is also very attractive” he says. You don’t have the hassle of whether staff will turn up or not and you have a known fixed rate of costs. Labour costs for handweeding crops vary from $150 to $600/ha depending on the weed infestation, while the machine is being charged out at a flat rate of $225/ha.”

And the results are consistent across the whole field. If one out of 20 staff is not pulling their weight the results can be significant in terms of weed growth and crop suppression.

“We are also seeing an ever-increasing pull from consumers for less chemicals used in producing food. Mechanisation will reduce the need for chemicals and the likelihood of resistance to chemicals building up.”

Overseas universities have already been in contact about buying the machine and the company is patenting its weeder in 148 countries. They’ve just been asked to test the weeder in tomato crops, and they recently trialled weeding transplants. They’ve also trialled it on corn, maize and brassica crops. They’ve also been asked to adapt it to nursery applications, and for boysenberry and blackcurrants.

They’re hoping to make the first sales of the Andweeder soon.