MIT Cheetah is an open-source quadropedal robot designed using low-inertia actuators. This robot is developed in the lab of Professor Sangbae Kim at MIT[1]. The MIT Cheetah is known for its agility and ability to adapt to varying terrain conditions without requiring a terrain map in advance
MIT Cheetah | |
---|---|
Name | MIT Cheetah |
Organization | MIT |
Weight | 20 pounds |
Speed | Twice the speed of average human walking speed[2],[3] |
Status | Active |
Design and Development
The MIT Cheetah displays remarkable adaptability in its functions. Despite being only 20 pounds, the robot can bend and swing its legs wide, allowing it to either walk right side up or upside down[4]. This flexibility is a result of its design that prioritizes a wide range of motion.
The robot is further equipped to deal with challenges posed by rough, uneven terrain. It can traverse such landscapes at a pace twice as fast as an average person's walking speed[5]. The MIT Cheetah’s design, particularly the implementation of a control system that enables agile running, has been largely driven by the "learn-by-experience model". This approach, in contrast to previous designs reliant primarily on human analytical insights, allows the robot to respond quickly to changes in the environment[6].
Chips
References
- ↑ https://www.csail.mit.edu/news/one-giant-leap-mini-cheetah
- ↑ https://news.mit.edu/2019/mit-mini-cheetah-first-four-legged-robot-to-backflip-0304
- ↑ http://robotics.mit.edu/mini-cheetah-first-four-legged-robot-do-backflip
- ↑ https://news.mit.edu/2019/mit-mini-cheetah-first-four-legged-robot-to-backflip-0304
- ↑ http://robotics.mit.edu/mini-cheetah-first-four-legged-robot-do-backflip
- ↑ https://news.mit.edu/2022/3-questions-how-mit-mini-cheetah-learns-run-fast-0317