Agile Locomotion

Advancing the accessibility and efficacy of neuromotor rehabilitation requires new technology that enables robust and accurate measurements of human behavior during everyday activities in real-world conditions. 

In dynamic-legged locomotion, the size of the valid stepping region is crucial as it determines the maximum CoM velocity that can be safely regulated. Previous work on dynamic legged locomotion conservatively restricted the stepping region in the lateral direction so that the robot's legs do not cross one another to mitigate the risk of self-collision. However, this restriction prevents the robots from taking aggressive but stabilizing steps, thus reducing their balance stability and robustness to external disturbances. In this work, we aim to address this undue trade-off by deploying an RMPflow-based reactive collision-avoidance swing leg controller. The controller steers the swing foot toward the planned step location while avoiding collisions between the robot's links. This approach enables full utilization of the kinematically reachable region as a valid stepping area, thereby widening the robot's stability margin and improving its robustness to external disturbances.