MIT’s Robotic Mini Cheetah can jump across uneven terrain in real-time
A novel control technique, demonstrated by’s robotic mini cheetah, allows four-legged robots to hop across rough terrain in real time.
A loping cheetah bounds across a rolling meadow, leaping over abrupt gaps in the rocky terrain. The movement appears to be effortless, but getting a robot to move in this manner is a very different situation.
Four-legged robots inspired by the movement of cheetahs and other animals have made significant advances in recent years, but they still lag behind their mammalian counterparts when traversing through a landscape with quick elevation changes.
“In those settings, you need to use vision in order to avoid failure. For example, stepping in a gap is difficult to avoid if you can’t see it. Although there are some existing methods for incorporating vision into legged locomotion, most of them aren’t really suitable for use with emerging agile robotic systems,” says Gabriel Margolis, a PhD student in Pulkit Agrawal’s lab at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL).
Margolis and his colleagues have now devised a technique that improves the speed and agility of legged robots as they jump across terrain gaps. The unique control system is divided into two parts: one that analyses real-time input from a video camera attached on the robot’s front, and another that interprets that information into instructions for how the robot should move its body. The researchers put their system through its paces on the MIT micro cheetah, a powerful, nimble robot developed in the lab of mechanical engineering professor Sangbae Kim.
Unlike other ways for directing a four-legged robot, this two-part approach does not require the terrain to be pre-mapped, allowing the robot to move wherever it wants. This may allow robots to charge into the woods on an emergency response mission or to climb a flight of stairs to deliver medication to an elderly shut-in.
Margolis collaborated on the paper with senior author Pulkit Agrawal, who heads the Improbable AI lab at MIT and is the Steven G. and Renee Finn Career Development Assistant Professor in the Department of Electrical Engineering and Computer Science; Professor Sangbae Kim in the Department of Mechanical Engineering at MIT; and MIT graduate students Tao Chen and Xiang Fu. Kartik Paigwar, a graduate student at Arizona State University, and Donghyun Kim, an assistant professor at the University of Massachusetts in Amherst, are also co-authors…. Summary of Brinkwire News.