analiza aiqbench
This research advances humanoid locomotion by deploying a reinforcement learning system that processes raw solid-state lidar data to navigate sparse 3D structures like monkey bars. The approach integrates attention-based encoders with recurrent memory to handle noisy sensor inputs and physical constraints such as battery sag, achieving successful traversal in hardware trials. This work represents a significant step toward robust, perception-driven whole-body control for next-generation autonomous robots.
Learning Agile Perceptive Traversal of Sparse 3D Structures for Humanoids
arXiv:2608.29769v1 Announce Type: new Abstract: Traversing sparse 3D structures requires humanoid robots to perceive thin, overhanging geometry while executing agile, accurate whole-body motions. We study this problem through monkey-bar traversal, where the robot must jump to the structure, traverse it through sparse bar interactions, and land safely. For this task, we present a reinforcement-learning-based perceptive control system that operates directly on observations from a head-mounted solid-state lidar. To extract task-relevant geometry from the sparse returns, the policy consumes the raw lidar scan through an attention-based encoder with recurrent memory. This policy is obtained by a phase-scheduled teacher- student pipeline that combines privileged experts for jumping up, brachiating, and jumping down. For transfer to hardware, we model lidar noise, battery-voltage sag, and actuator thermal limits, and equip the humanoid with passive hook end-effectors for robust bar interaction. On hardware, the resulting policy completes the full jump-up->brachiation->jump-down sequence in 14 of 15 trials across three bar configurations and reaches brachiation speeds up to 0.5 m/s. Beyond brachiation, the same perception backbone supports a separately trained policy that ducks beneath thin overhead obstacles with 2 cm cross-sections.
Abstract:Traversing sparse 3D structures requires humanoid robots to perceive thin, overhanging geometry while executing agile, accurate whole-body motions. We study this problem through monkey-bar traversal, where the robot must jump to the structure, traverse it through sparse bar interactions, and land safely. For this task, we present a reinforcement-learning-based perceptive control system that operates directly on observations from a head-mounted solid-state lidar. To extract task-relevant geometry from the sparse returns, the policy consumes the raw lidar scan through an attention-based encoder with recurrent memory. This policy is obtained by a phase-scheduled teacher- student pipeline that combines privileged experts for jumping up, brachiating, and jumping down. For transfer to hardware, we model lidar noise, battery-voltage sag, and actuator thermal limits, and equip the humanoid with passive hook end-effectors for robust bar interaction. On hardware, the resulting policy completes the full jump-up->brachiation->jump-down sequence in 14 of 15 trials across three bar configurations and reaches brachiation speeds up to 0.5 m/s. Beyond brachiation, the same perception backbone supports a separately trained policy that ducks beneath thin overhead obstacles with 2 cm cross-sections.
Robotics (cs.RO)
arXiv:2608.29769 [cs.RO]
(or arXiv:2608.29769v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2608.29769 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 272 cuvinte
Subiecte conexe
Echo in the Steps: Learning Perceptive Humanoid Parkour with Gated Memory
arXiv · 25 sept. 2026
TAGA: Terrain-aware Active Gaze Learning for Generalizable Agile Humanoid Locomotion
arXiv · 28 sept. 2026
TAGA: Terrain-aware Active Gaze Learning for Generalizable Agile Humanoid Locomotion
arXiv · 5 iun. 2026
Physics-Guided Residual Reinforcement Learning for Humanoid Narrow-Path Traversal
arXiv · 25 sept. 2026