Adaptable robots entail a leap for rescue operations and more.
Autonomous robots are one of contemporary engineering’s most amazing inventions. They excel in factories and other manmade spaces. But amid the randomness of nature, they struggle.
To help these machines overcome uneven terrain and other obstacles, University at Buffalo researchers have turned to beavers, termites and other animals that build structures in response to simple environmental cues, as opposed to following predetermined plans.
“When a beaver builds a dam, it’s not following a blueprint. It’s reacting to moving water. It’s trying to stop the water from flowing,” says Nils Napp, assistant professor of computer science and engineering at UB. “We’re developing a system for autonomous robots to behave similarly.”
The work could have implications for everything from search-and-rescue operations to planetary exploration for Mars-rover-style vehicles.
While the project involves animals and robots, its main focus is math: specifically, developing new algorithms, or the sets of rules that self-governing machines need to make sense of their environment and solve problems.
Creating algorithms for an autonomous robot in a controlled environment is relatively straightforward. It’s much more difficult to accomplish for the natural world, where spaces are unpredictable, Napp says.
To address the issue, he looked to the type of indirect coordination used by termites to build a nest, and even by humans to create Wikipedia. In both cases, new users react to and modify what’s already there, ultimately building something complex.
Using off-the-shelf components, Napp and his students outfit a mini-rover vehicle with a camera, custom software and a robotic arm to lift and deposit objects.
They then create uneven terrain out of rocks, bricks and broken concrete to simulate an environment after a disaster such as an earthquake, and place small bean bags around the simulated disaster area.
Finally they activate the robot. Using the algorithms Napp developed, it picks up bean bags and deposits them in holes and gaps in the broken terrain. Eventually the bags form a ramp, which allows the robot to reach its target location, a flat platform.
“It’s like a beaver using nearby materials to build with,” Napp says. “The robot takes its cues from its surroundings, and will keep modifying its environment until it has created a ramp. That means it can fix mistakes and react to disturbances, just like beavers that fix leaks in their dams.”