ALLSTEPS: Curriculum-driven Learning of Stepping Stone Skills
This addresses a fundamental problem in animation and robotics for creating realistic bipedal movements in constrained environments, though it appears incremental as it focuses on curriculum improvements.
The paper tackled the challenge of generating robust locomotion for stepping-stone scenarios with fully constrained foot placements using reinforcement learning, achieving plausible motions across simulated human, robot, and monster characters.
Humans are highly adept at walking in environments with foot placement constraints, including stepping-stone scenarios where the footstep locations are fully constrained. Finding good solutions to stepping-stone locomotion is a longstanding and fundamental challenge for animation and robotics. We present fully learned solutions to this difficult problem using reinforcement learning. We demonstrate the importance of a curriculum for efficient learning and evaluate four possible curriculum choices compared to a non-curriculum baseline. Results are presented for a simulated human character, a realistic bipedal robot simulation and a monster character, in each case producing robust, plausible motions for challenging stepping stone sequences and terrains.