Robot Crash Course: Learning Soft and Stylized Falling
This addresses safety and damage reduction for bipedal robots in real-world applications, representing an incremental approach by shifting focus from fall prevention to controlled falling.
The paper tackles the problem of bipedal robots falling by focusing on reducing physical damage and allowing user control over the end pose, demonstrating through experiments that robots can perform controlled, soft falls.
Despite recent advances in robust locomotion, bipedal robots operating in the real world remain at risk of falling. While most research focuses on preventing such events, we instead concentrate on the phenomenon of falling itself. Specifically, we aim to reduce physical damage to the robot while providing users with control over a robot's end pose. To this end, we propose a robot agnostic reward function that balances the achievement of a desired end pose with impact minimization and the protection of critical robot parts during reinforcement learning. To make the policy robust to a broad range of initial falling conditions and to enable the specification of an arbitrary and unseen end pose at inference time, we introduce a simulation-based sampling strategy of initial and end poses. Through simulated and real-world experiments, our work demonstrates that even bipedal robots can perform controlled, soft falls.