aiqbench analysis
This paper applies reinforcement learning to quadruped locomotion, training one PPO policy that takes raw LiDAR point clouds as input and skips the intermediate terrain map. The reported results are 90.1% mean success in simulation and zero-shot real-world tests on a Deep Robotics X30, including a 50-kg payload and a 2.5-km route. Readers should weigh the simulation figures against the limited hardware evidence, and note that the sim-to-real claims rest on the authors' own dual-channel simulation of see-through gratings.
PointWalker: Mapless Quadruped Locomotion Directly From Raw LiDAR Point Clouds
arXiv:2610.15548v1 Announce Type: new Abstract: Quadruped robots deployed for inspection and field missions must cross heterogeneous terrain, from stairs, slopes, and gaps to high platforms, stepping stones, and balance beams, where safe support is sparse or see-through. Existing methods interpose a terrain map or a depth image between sensing and control. Maps are costly to maintain, corrupted by odometry drift and outliers, and misled by see-through gratings, while depth cameras degrade with lighting and surface material. In this paper, we present PointWalker, a mapless framework whose single reinforcement learning policy directly consumes raw point clouds from four light detection and ranging (LiDAR) sensors. Voxelized points are encoded by linear self-attention and fused with proprioception through cross-attention, so terrain geometry reaches the policy without maps or foothold estimation. One proximal policy optimization (PPO) run, with terrain-aware rewards and dual-channel simulation of see-through gratings, yields one policy for all seven terrain classes with a 90.1% mean success rate in simulation. Deployed zero-shot on a Deep Robotics X30 at 50 Hz onboard, it crosses 0.8-m gaps, climbs 0.8-m platforms and 45{\deg} steel-grating stairs, carries a 50-kg payload, and completes a 2.5-km mountain route.
Abstract:Quadruped robots deployed for inspection and field missions must cross heterogeneous terrain, from stairs, slopes, and gaps to high platforms, stepping stones, and balance beams, where safe support is sparse or see-through. Existing methods interpose a terrain map or a depth image between sensing and control. Maps are costly to maintain, corrupted by odometry drift and outliers, and misled by see-through gratings, while depth cameras degrade with lighting and surface material. In this paper, we present PointWalker, a mapless framework whose single reinforcement learning policy directly consumes raw point clouds from four light detection and ranging (LiDAR) sensors. Voxelized points are encoded by linear self-attention and fused with proprioception through cross-attention, so terrain geometry reaches the policy without maps or foothold estimation. One proximal policy optimization (PPO) run, with terrain-aware rewards and dual-channel simulation of see-through gratings, yields one policy for all seven terrain classes with a 90.1% mean success rate in simulation. Deployed zero-shot on a Deep Robotics X30 at 50 Hz onboard, it crosses 0.8-m gaps, climbs 0.8-m platforms and 45° steel-grating stairs, carries a 50-kg payload, and completes a 2.5-km mountain route.
8 pages, 15 figures, 5 tables. Submitted to IEEE Robotics and Automation Letters (RA-L)
Robotics (cs.RO)
arXiv:2610.15548 [cs.RO]
(or arXiv:2610.15548v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.15548 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.
Full text under the publisher's open licence · 292 words
Related coverage
DR-MPC: Fast and Feasible Dynamics-Relaxed Model-Predictive Control for Legged Locomotion
arXiv · Sep 18, 2026
Scale Robot Reinforcement Learning with NVIDIA Isaac Lab on Amazon SageMaker AI
AWS · Jun 9, 2026
Affordance-Based Hierarchical Reinforcement Learning for Quadruped Pedipulation
arXiv · Jun 8, 2026
Quadruped Obstacle Avoidance and Footstep Planning with Distributed Low-cost Time-of-Flight Sensors
arXiv · Sep 28, 2026