ROOF: Learning Humanoid Fall Recovery under Overhead Constraints
This paper introduces ROOF, a reinforcement learning method that enables humanoids to recover from falls beneath overhead obstacles using contact-supported geometric escape guidance. The approach learns when to clear the obstacle, establish support, and complete the rise using only proprioception. In simulation, it achieves 88.5% recovery under lift-only plates (vs. 18.9% without guidance) and 98.6% under free plates with randomized properties. On a Unitree G1 robot, the policy successfully recovered in 95 of 150 attempts across seven real-world obstacle scenes, including furniture and multi-object arrangements not present during training.