Learning Getting-Up Policies for Real-World Humanoid Robots

摘要

Automatic recovery from falls is a crucial prerequisite before humanoid robots can be reliably deployed. Hand-designing controllers for getting up is difficult because of the varied configurations a humanoid can end up in after a fall and the challenging terrains humanoid robots are expected to operate on. This paper develops a learning framework to produce controllers that enable humanoid robots to get up from varying configurations on varying terrains. Different from previous successful applications of humanoid locomotion learning, the getting-up task involves complex contact patterns, the necessity to accurately model collision geometry, and sparser rewards. We circumvent these challenges through a two-phase approach that follows a curriculum. The first stage focuses on discovering a good get up trajectory under minimal constraints on smoothness or speed / torque limits. The second stage then refines the discovered motions into deployable (