REAL achieves extreme quadruped parkour agility that is robust even to a 1-meter visual blind zone.
arXiv · March 19, 2026 · 2603.17653
The Takeaway
It combines FiLM-modulated Mamba backbones for terrain memory with physics-guided Bayesian state estimation. This allows robots to maintain high-speed maneuvers through sensory corruption or perception lag, solving a major bottleneck in real-world robotics deployment.
From the abstract
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 percepti