Sep 18, 2015

Robotics drive crop, harvest mechanization

In recent years, there has been an increase in the agriculture industry’s level of interest in the development and deployment of driverless tractors, aerial surveying of farmlands, data collection, field management and other technology-based systems.

The demand for robots involved in various agricultural processes like harvesting, pruning, weeding, pick-and-place, sorting, seeding, spraying and materials handling has increased significantly. Driving this sector is the recognition of agriculture’s dwindling labor force.

According to a report from Tractica, annual shipments of agricultural robots will reach 992,000 worldwide by 2024, up from just 33,000 in 2015. The market research firm forecasts that some of the largest application segments will include unmanned aerial vehicles (UAVs) for agricultural purposes, soil management robots, materials management robots, driverless tractors and dairy management robots.

“Demand for agricultural robots is being driven by a number of global trends,” said research analyst Manoj Sahi. “Key economic and demographic factors influencing market development include population growth, increasing strain on the food supply, availability of farm workers, the challenges and complexities of farm labor, the cost of farm workers, shrinking farmlands, climate change, the growth of indoor farming and the automation of the agriculture industry.”

Tractica expects that the market for agricultural robots will develop rapidly in the next five years, as compared with the previous five years, while “providing significant market opportunities to various market participants.” Tractica anticipates that the overall agricultural robot market will reach $3 billion by the end of 2015. The market is anticipated to maintain a healthy growth rate and will reach $16.8 billion by the end of 2020, Sahi said.

Tractica forecasts that the market will continue its momentum, reaching $73.9 billion by 2024.

Specialty crop tools

Harvey (Tony) Koselka, chief operating officer/vice president of engineering for Vision Robotics Corp. of San Diego, recognizes the inroads being made by labor-saving engineering advances for agricultural applications. He was a speaker at the 2015 Mid-Atlantic Fruit & Vegetable Convention in Hershey, Pennsylvania.

“A lot of the non-specialty crops, the row crops, have been hugely mechanized because they can work indiscriminately. So, they don’t have to be smart, they don’t have to work with individual plants or pieces of fruit,” Koselka said.

“(With) specialty crops, we can now do it,” he said. “Computers are smart enough, the power has gone there. Everybody talks about computers, but it’s not just computers. (There are) lots of sensors. We use vision; we use stereo cameras – two cameras side by side to give you depth perception. There are other sensors – Lidar sensors (a remote sensing technology that measures distance by illuminating a target with a laser and analyzing the reflected light). Those are still expensive. Connect, which is a toy that Microsoft put out for their Xbox. It’s a time-of-flight sensor that gives you 3-D perception as well.

“Batteries are better for cars. Small motors are more powerful. All of this stuff is coming together and we sort of have a convergence. This is the right time and a good time for the technology,” Koselka said.

“On the other side, we have people working on it,” he said. “There are a number of companies making intelligent robot systems. There are companies doing facial recognition with computers. Companies in manufacturing are trying to make smart computers to figure how to pick things up or smart robots that can figure out how to pick things up. All that stuff’s being used.

“On top of it, there are already a lot of people working in agriculture, and all that knowledge is being pushed forward with the new technology,” Koselka said. “So it’s all coming together. And we’re at the cusp of doing it.”

Is the evolving technology finding favor with reluctant growers?

“We have a lettuce thinner out that we’ve been selling. We look at the economics of it and it’s a no-brainer to do,” Koselka said. “(Growers) say, ‘Well, is it going to survive five years?’ The payback is six months, so why do you care if it’s going to survive five years? There is still concern. They see it, they get more comfortable. It’s been out there two or three years, we have people calling us now.

“So that’s the problem (of acceptance). The early adapters are comfortable with it. They’re comfortable with technology,” Koselka said. “Everybody wants to see it working on their own farm. They say we’re totally different than the guy next to us, whether they are or aren’t. So the adoption is going to be slower. We thought it would be a lot faster, but right now we’re realizing it’s going to be a longer time for people to make it as widespread as we had hoped.”

Koselka’s company is at the forefront of the new technologies.

“We’ve been involved in a lot of specialty crops,” he said. “Right now, we do have a product that’s in production – a lettuce thinner. About 10,0000 to 20,000 acres have been thinned mechanically with our system. We’re working with a grapevine pruner to automate our pruning of wine grapes. Those are very specific training systems, so we’re doing that. We’re working on an SCRI (Specialty Crop Research Initiative) project, a government-funded project for pruning of apple trees as well. So a lot of stuff is being developed in recognizing and understanding plant structures.”

“For a lot of the easier applications the technology is there, it’s the funding that will stop it,” Koselka said. “Other applications will get there in time.”

Meeting grower needs

But when investment funding is made available, development of harvest-assist mechanization products is quickly expanding.

One example of the evolving technology is a Harvest CROO Robotics’ automated strawberry picker. Harvest CROO has published a patent for its robotic harvesting picking wheel. It’s the first patent pending for the company, which develops and tests robotic harvesting machinery for strawberries.

Machinery manufactured by the company, co-founded by Gary Wishnatzki, owner of Wish Farms, works autonomously in fields, picking and packing berries.

The company, co-founded by Bob Pitzer, the chief technical officer who invented the machine’s harvesting wheel, constructed a prototype in 2014 in Florida and is building another prototype for harvesting in Santa Maria, Calif.

“I charged our engineers with the task of creating a ‘picker’ that does not require a grower to radically change the way they currently grow,” Wishnatzki said. “That is the major reason other robotic harvesters have not yet been commercialized.”

Harvest CROO machines will pick traditional strawberry beds. Pitzer took to the fields to study and observe the way human pickers harvest strawberries. With that information, he began outlining and conceptualizing the first prototype, which mimics the ways humans pick.

“With robotic manipulation, our biggest challenge is minimizing time,” Pitzer said. “Based on our observations, our goal was to develop robots to pick as many berries as possible while utilizing conservation of motion.”

“In the upcoming season, we will have a smaller picker prototype that will be refining our techniques, as well as a motorized platform that we will be working with for the self-driving feature,” Wishnatzki said. “Additionally, we will have a team working on the packaging technology.”

The company, which is producing the machinery in collaboration with investors in Florida and California, has a targeted release date of late 2016. The company plans to lease the machines to keep costs down for growers.

– Gary Pullano