John Deere is working with the University of Illinois to create a vision-based navigation system for the remote mower.Photo: Soon-Jo Chung
In order to operate within a specific area, robotic lawn mowers rely on boundary wires installed on a property prior to their use. When homeowners move or make changes to the property, the wires have to be reinstalled, an inconvenience that a new partnership between researchers at the University of Illinois and John Deere is determined to alleviate.
Working with two colleagues, aerospace engineering assistant professor Soon-Jo Chung is creating a new vision-based navigation system that will allow the mowers to operate more autonomously and with greater safety. The university hopes to see the improved product introduced in the United States eventually.
Chung described the new system to the Daily Illini as a combination of vehicles’ GPS navigation and collision avoidance systems. “It’s going to be safer than existing systems because this vision-based navigation system can not only provide the localization information of the mower, but also it’s going to provide much-needed safety features.”
Using its vision-based autonomous navigation system, the mower detects and registers features like windows, doors and trees in order to learn the boundary of the lawn. It can also then detect other features that may obstruct its movement or performance in order to avoid a collision with moving objects, including people.
“I believe this vision-based navigation system will really facilitate the adoption of this kind of robotic mower as mainstream household robots you can buy in addition to vacuum robots,” Chung said.
