arXiv:2606. 08253v1 Announce Type: cross Abstract: Enabling humanoid robots to operate in complex, dynamic environments remains a critical challenge, fundamentally limited by the ability to navigate robustly, safely, and accurately.
The paper proposes a lightweight foothold prediction model (FPM) that anticipates future foot placements using only proprioception and velocity commands, allowing the robot to sample elevation maps only around predicted footholds. This preprocessing reduces redundant sensory input, enabling the locomotion policy to learn faster—reaching a given terrain difficulty in about one‑third the training iterations compared to using the full elevation map. The resulting policy outperforms both full‑map and attention‑based encoders on sparse foothold scenarios, and it zero‑shot transfers to a Unitree G1 humanoid across diverse indoor and outdoor obstacles.
REACH is a reinforcement‑learning based hierarchical framework that allows a Unitree G1 humanoid robot to navigate confined spaces by switching between variable‑height bipedal walking and crawling. The system learns policies that adjust body height from 1.27 m down to 0.20 m, uses a history‑conditioned latent representation to handle contact‑rich crawling dynamics, and incorporates a bidirectional transition policy to connect the two locomotion modes. A LiDAR‑based planner generates collision‑free 3D paths and proactive height targets, and real‑world tests confirm accurate height and velocity tracking, robust terrain traversal, and reliable mode transitions.
arXiv:2601. 15995v2 Announce Type: replace-cross Abstract: Parkour tasks for quadrupeds have emerged as a promising benchmark for agile locomotion.
arXiv:2602.02331v2 Announce Type: replace-cross Abstract: Achieving highly dynamic parkour with humanoid robots on complex terrains unseen during the pre-training stage remains a challenge in robotic...
arXiv:2606. 15896v1 Announce Type: cross Abstract: Learning-based quadrupedal locomotion typically relies on complex reward formulations that entangle task specification, operational limits, gait preference, and terrain adaptation within a single optimization objective.
Humanoid locomotion in confined spaces requires continuous height adaptation across multiple locomotion modes, yet most existing approaches remain limited to height-adjustable bipedal walking. We pres...
arXiv:2510. 18348v2 Announce Type: replace-cross Abstract: State-of-the-art perceptive Reinforcement Learning controllers for legged robots typically either (i) impose oscillator-or IK-based gait priors that constrain the action space, bias policy optimization, and limit adaptability across robot morphologies, or (ii) operate "blind," making them unable to anticipate hind-leg terrain and brittle to observation noise.
FootQuery is a perceptive locomotion framework that retrieves depth information from a robot’s own history by querying each foot’s predicted next touchdown. The policy uses proprioceptive predictions of touchdown locations and uncertainties to sample relevant historical depth frames, fuses these per‑foot features with global visual memory, and generates control actions. In simulation and on a real Unitree G1 robot, FootQuery enables continuous traversal of complex outdoor stairs, indoor routes, platforms, and gaps, outperforming component ablations.
arXiv:2606. 04718v1 Announce Type: cross Abstract: Humans primarily rely on walking and running to traverse complex terrains, without resorting to unnecessarily complex motion patterns.
arXiv:2607. 13579v1 Announce Type: cross Abstract: Enabling quadrupedal robots to traverse complex terrains-from rugged outdoor environments to urban landscapes-requires seamless integration of multiple motor skills, smooth transitions between gaits, and high-speed perceptive locomotion using only onboard sensors.
arXiv:2509. 26633v3 Announce Type: replace-cross Abstract: A dominant paradigm for teaching humanoid robots complex skills is to retarget human motions as kinematic references to train reinforcement learning (RL) policies.