REAL: Robust Extreme Agility via Spatio-Temporal Policy Learning and Physics-Guided Filtering
This addresses the fragility of learning-based systems in robotics to perceptual degradation, enabling reliable parkour for quadrupeds in noisy environments.
The paper tackles the problem of extreme legged parkour under sensory corruption, achieving robust traversal of obstacles even with a 1-meter visual blind zone while maintaining real-time control with a bounded 13.1 ms inference time.
Extreme legged parkour demands rapid terrain assessment and precise foot placement under highly dynamic conditions. While recent learning-based systems achieve impressive agility, they remain fundamentally fragile to perceptual degradation, where even brief visual noise or latency can cause catastrophic failure. To overcome this, we propose Robust Extreme Agility Learning (REAL), an end-to-end framework for reliable parkour under sensory corruption. Instead of relying on perfectly clean perception, REAL tightly couples vision, proprioceptive history, and temporal memory. We distill a cross-modal teacher policy into a deployable student equipped with a FiLM-modulated Mamba backbone to actively filter visual noise and build short-term terrain memory actively. Furthermore, a physics-guided Bayesian state estimator enforces rigid-body consistency during high-impact maneuvers. Validated on a Unitree Go2 quadruped, REAL successfully traverses extreme obstacles even with a 1-meter visual blind zone, while strictly satisfying real-time control constraints with a bounded 13.1 ms inference time.