ALLSTEPS: Curriculum-driven Learning of Stepping Stone Skills

Abstract

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.

References in corpus (6)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Emergence of Locomotion Behaviours in Rich Environments
• Intrinsically Motivated Goal Exploration Processes with Automatic Curriculum Learning
• Reverse Curriculum Generation for Reinforcement Learning
• Paired Open-Ended Trailblazer (POET): Endlessly Generating Increasingly Complex and Diverse Learning Environments and Their Solutions
• Synthesis of Biologically Realistic Human Motion Using Joint Torque Actuation