analiza aiqbench
TAGA introduces a reinforcement learning framework that enables humanoids to actively focus on critical terrain features, significantly improving training efficiency and robustness under computational constraints. The approach achieves state-of-the-art performance in real-world hardware tests, including a 1.2m gap traversal distance, by naturally emerging gaze behaviors without explicit supervision. This advancement marks a key step toward scalable, perception-driven locomotion for agile robots operating in complex environments.
TAGA: Terrain-aware Active Gaze Learning for Generalizable Agile Humanoid Locomotion
arXiv:2606.05880v1 Announce Type: new Abstract: Agile humanoid locomotion across diverse challenging terrain demands both wide perceptual coverage and precise local geometry understanding. Motivated by the way humans selectively look at relevant terrain during locomotion, we introduce TAGA, a Terrain-aware Active Gaze learning framework for Attention-based humanoid control. By fusing vision, proprioception, and motion commands, our framework guides the model to learn anticipatory cues and actively attend to specific areas of the height scan, selectively using these informative regions for the downstream network. This adaptively increases the information density of observations under tight onboard computational constraints, thus enabling fine-grained perceptive locomotion over larger-scale terrains. We find that such gaze behaviors can naturally emerge through reinforcement learning alone, without requiring additional supervision or explicit guidance, significantly improve training efficiency. As a result, the trained policy demonstrates robust and generalizable locomotion in simulation and on hardware, including reliable terrain-aware foothold selection, elevated-platform traversal, competitive sparse-foothold traversal, and the largest reported real-world gap traversal distance of 1.2m among perceptive humanoid locomotion systems, while maintaining stability under severe perceptual disturbances and environmental interference.
Abstract:Agile humanoid locomotion across diverse challenging terrain demands both wide perceptual coverage and precise local geometry understanding. Motivated by the way humans selectively look at relevant terrain during locomotion, we introduce TAGA, a Terrain-aware Active Gaze learning framework for Attention-based humanoid control. By fusing vision, proprioception, and motion commands, our framework guides the model to learn anticipatory cues and actively attend to specific areas of the height scan, selectively using these informative regions for the downstream network. This adaptively increases the information density of observations under tight onboard computational constraints, thus enabling fine-grained perceptive locomotion over larger-scale terrains. We find that such gaze behaviors can naturally emerge through reinforcement learning alone, without requiring additional supervision or explicit guidance, significantly improve training efficiency. As a result, the trained policy demonstrates robust and generalizable locomotion in simulation and on hardware, including reliable terrain-aware foothold selection, elevated-platform traversal, competitive sparse-foothold traversal, and the largest reported real-world gap traversal distance of 1.2m among perceptive humanoid locomotion systems, while maintaining stability under severe perceptual disturbances and environmental interference.
Robotics (cs.RO)
arXiv:2606.05880 [cs.RO]
(or arXiv:2606.05880v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.05880 arXiv-issued DOI via DataCite (pending registration)
Text integral sub licența deschisă a publicației · 190 cuvinte
Subiecte conexe
TAGA: Terrain-aware Active Gaze Learning for Generalizable Agile Humanoid Locomotion
arXiv · 28 sept. 2026
StairMaster: Learning to Conquer Risky Hollow Stairs for Agile Quadrupedal Robots
arXiv · 30 iul. 2026
Vertical Advantage: Transforming Industries with Lakebase and Agentic AI
Databricks · 27 aug. 2026
News
Quantum Computing Report · 25 iul. 2026