Four-legged robot learns to jump through tight gaps without human help
SMRTR summary
Researchers at the University of Hong Kong and the Oxford Robotics Institute have developed a system that lets a four-legged robot jump through tight gaps on its own. Using a two-level AI framework called ConsJump, the robot decides when to sprint, leap, and land without pre-programmed instructions. Tested on a 48-pound Unitree Aliengo robot, it reached speeds of 2.5 meters per second and cleared openings comparable to its own body size.
SMRTR provides this summary for quick context. The original article belongs to Interesting Engineering.
Read the original article