PUMA: Perception-driven Unified Foothold Prior for Mobility Augmented Quadruped Parkour
This work addresses the problem of real-time adaptability in quadruped parkour for robotics, representing a novel method rather than an incremental improvement.
The paper tackles the challenge of enabling legged robots to perceive and select footholds for agile parkour tasks by introducing PUMA, an end-to-end learning framework that integrates visual perception and foothold priors, resulting in exceptional agility and robustness in complex terrains as demonstrated in simulation and real-world experiments.
Parkour tasks for quadrupeds have emerged as a promising benchmark for agile locomotion. While human athletes can effectively perceive environmental characteristics to select appropriate footholds for obstacle traversal, endowing legged robots with similar perceptual reasoning remains a significant challenge. Existing methods often rely on hierarchical controllers that follow pre-computed footholds, thereby constraining the robot's real-time adaptability and the exploratory potential of reinforcement learning. To overcome these challenges, we present PUMA, an end-to-end learning framework that integrates visual perception and foothold priors into a single-stage training process. This approach leverages terrain features to estimate egocentric polar foothold priors, composed of relative distance and heading, guiding the robot in active posture adaptation for parkour tasks. Extensive experiments conducted in simulation and real-world environments across various discrete complex terrains, demonstrate PUMA's exceptional agility and robustness in challenging scenarios.