Researchers develop robots for disaster zones
When responding to a natural or man-made disaster means putting human lives at grave risk—as during the Fukushima Daiichi nuclear disaster—wouldn’t it be ideal to send in the first-responder robots instead?
The Ohio State University is part of a 10-school collaboration, led by Drexel University, working to advance robotics technology for disaster relief as part of the U.S. Defense Advanced Research Projects Agency Robotics Challenge. Teams from academia, industry and the private sector will attempt to design and deploy a robot capable of disaster response in radioactive or bio-contaminated areas. The robot must drive vehicles, navigate human-centered environments, use tools and manipulate equipment.
Yuan Zheng, professor of electrical and computer engineering, leads Ohio State’s efforts in the challenge.
See a video of the challenge.
Researchers from each of the 10 partner schools are working to tackle specific aspects of the challenge, which is broken into eight events related to disaster mitigation. Robots must mount, drive and dismount a vehicle; travel across rubble; remove debris; open a door, climb a ladder; use a tool to break through a concrete wall, locate and shut off a leaky valve; and remove and replace a pump.
According to Zheng, having a robot capable of accomplishing even one of these tasks would be a giant leap forward from the current state-of-the-art in robotics technology.
“A robot can’t do any of these tasks today. Just getting a robot to climb into a vehicle is very difficult, almost impossible, to perform,” he said.
The Ohio State team is responsible for equipping the robot to travel across rubble, which might be anything from rocks to large piles of debris. The group developed what they consider to be an innovative gait, inspired by the poles skiers use to aid in balance.
“We developed an innovative update, called a “ski type” gait, which adds two canes to the robot to increase its support area and stability,” Zheng said. “Using this approach, if the surface becomes sloped or uneven the robot still has some margin to maintain stability. It’s much better than a two-legged robot.”
The removable canes increase the robot’s stability without making permanent changes to its structure, which could affect its performance in other challenge events.
The Drexel-led group, dubbed the DRC-HUBO team, is one of seven selected for Track A funding by DARPA and must develop their own robot and operation software. The team derives its name from the HUBO humanoid robots they will employ for the challenge. The robots were developed by another challenge partner, Korea Advanced Institute for Science and Technology, and funded by an earlier collaboration of universities, which is also led by Drexel University and includes Ohio State. That collaboration resulted in a $6 million National Science Foundation grant to develop a common platform (HUBO+) for humanoid robotics research. The team is also using HUBO+ in the challenge.
The team passed the first critical design review in June, a virtual competition that tested the teams’ concepts via simulation. In December 2013, the conceptual design and simulation phase will culminate with a live competition to test the robots’ ability to complete the eight events. DARPA will then select teams to continue into Phase 2 for another head-to-head competition 12-months later.
Another prerequisite to engineering robots capable of responding to nuclear disasters is making them able to withstand high levels of radiation. Zheng is working with Lei (Raymond) Cao, assistant professor of nuclear engineering, on a separate $450,000 Department of Defense Threat Reduction Agency project to study the radiation sensitivity and failure mechanisms of robots working in harsh radiation environments.
“We are focusing on three fundamental components of robots—the battery, the magnet in the motor and the harmonic drive—under radiation,” said Zheng. “These studies have never been performed.”
The renewed interest in humanoid robotics has engaged Zheng—who was one of the first U.S. researchers working in humanoid robotics and developed the first humanoid in 1986—into the real-world application of the research area he embarked on more than 30 years ago.
“Robots are just cool and are always very challenging. Once you solve one problem, there is always another one to solve.”