A field with a future

Just a thin mast with a black pipe rocks to and fro between the ears of wheat – that’s all that can be seen of Terra Sentia from the edge of the field. The knee-high robot, which is electrically powered and guided by GPS, moves through a precisely laid-out trial field in Champaign near Chicago in the U.S. state of Illinois.

Jana Murche, the head of the North American wheat breeding program at KWS, is also in the field, collating phenotypic traits manually. But she can only watch over a small part of a plot. “It would be better to keep a constant eye on the plants and see how they fare in the field,” says Jia Yan, the project manager responsible for digital innovations at KWS in the U.S.

Robot takes pictures of the field

That is precisely where the robot helps breeders in their work. “The robot has several sensors and cameras on board to measure and map the field,” says Jia Yan. It also stores the exact position from where the pictures come and detects whether the plants are already flowering.

The objective is for Terra Sentia to travel over the field every day so as to supply even more data. And twice if necessary. Or three times.

It’s also relatively easy to make replicas of the robot. The predecessor model even came out of a 3D printer.

The technology “will allow us to make more soundly based decisions as part of selection,” says wheat breeder Mark Christopher. “Especially in our breeding nursery with its hundreds of thousands of individual rows, where it hasn’t been easy for us to collect that data up to now.”

The heart of the system: Software with artificial intelligence

Yet the heart of the system is not the four-wheeled robot, but an artificial intelligence software solution on the computers of KWS and the start-up Earth Sense, which has developed the robot.

The software analyzes the robot’s pictures and identifies things of interest to the breeders – for example if an ear on a stalk is already fully developed or if it is still partly enclosed by protective leaves.

The software must first be trained for that to succeed, i.e. it has to learn like a human. Wheat breeders like Jana Murche and Mark Christopher screen the images taken by the robot.

And then data scientists “feed” the software with information from the breeders. Once the artificial intelligence has obtained enough knowledge through this training, it then uses it to compare new images. It evaluates the pictures of plants without the need for human assistance.

The results of the system’s first version show that the algorithm already works accurately. The artificial intelligence detects complete ears with 96-percent reliability. It can identify whether an ear is completely awned with an accuracy of 92 percent.