ANN ARBOR – The National Science Foundation has awarded the University of Michigan a new $1 million research grant to “train” robots to navigate disaster zones and wilderness areas in real time.
U-M has been conducting research with bipedal (or two-legged) walking robots for years, and the new, three-year project aims to get the robots to traverse terrain freely without relying on a preexisting map of the area.
Researchers will equip the robots to dodge obstacles, adapt on treacherous ground and determine whether an area is safe to traverse.
The goal is to achieve a technology that could enable robots to enter areas that are too treacherous for first responders, like collapsed buildings. The technology could also have implications for improving prosthetics, making them more intuitive for those who wear them.
“I envision a robot that can walk autonomously through the forest here on North Campus and find an object we’ve hidden,” principal investigator on the project and the Elmer G. Gilbert Distinguished University Professor of Engineering at U-M, Jessy Grizzle, said in a release. “That’s what’s needed for robots to be useful in search and rescue, and no robot right now can do it.”
An expert in walking robots, Grizzle is partnering with Maani Ghaffari Jadidi on the project. Ghaffari is an expert in robotic perception and an assistant professor of naval architecture and marine engineering. Grizzle said the pair’s areas of expertise complement each other and will expand the technology they work with.
Together, they will implement an approach called “full-stack robotics” that will take new and existing technologies and create an open-source perception and movement system that can be applied to robots outside the project.
“What full-stack robotics means is that we’re attacking every layer of the problem at once and integrating them together,” Grizzle said in a statement. “Up to now, a lot of roboticists have been solving very specific individual problems. With this project, we aim to integrate what has already been done into a cohesive system, then identify its weak points and develop new technology where necessary to fill in the gaps.”
The team will be testing the technology on Mini Cheetah and Digit robots.
One of the project’s goals is to help robots create multidimensional maps based on information they receive in real time that will impact its decision-making when covering an unexplored area.
“When we humans go hiking, it’s easy for us to recognize areas that are too difficult or dangerous and stay away,” Ghaffari said in a statement. “We want a robot to be able to do something similar by using its perception tools to build a real-time map that looks several steps ahead and includes a measure of walkability. So it will know to stay away from dangerous areas, and it will be able to plan a route that uses its energy efficiently.”
Grizzle said he predicts the robots will be able to adapt by using math, and plans to develop more sophisticated tools to help robots gather data and perceive what their limbs are doing to generate new data points.
Ghaffari explained that rich maps will be essential one day to humans operating robots in remote search-and-rescue operations or in other circumstances.
“A shared understanding of the environment between humans and robots is essential, because the more a human team can see, the better they can interpret what the robot team is trying to accomplish,” Ghaffari said. “And that can help humans to make better decisions about what other resources need to be brought in or how the mission should proceed.”
In addition to technology development, the project will work with the U-M School of Education on outreach efforts with a high school in Detroit to develop interest in robotics.