Online Trajectory Generation for Step Recovery in Bipedal Robot Locomotion

We present a computationally efficient technique aimed at planning bipedal walking trajectories for push recovery. By modifying a layered walking controller, in case of pushes, we are able to plan a new position and timing for the forthcoming step. Tests have been performed in the Gazebo simulation environment using the one-meter-tall iCub humanoid robot model. Results show the effectiveness of the approach in stabilizing walking motions in case of strong external disturbances.https://youtu.be/BOPfzk917b