Robust Quadruped Locomotion via Reinforcement Learning with Deep Generalized-Momentum-Based Kalman Filter - 2026-06-08 - Research Publication - Lite3 hardware tests benchmark a reinforcement-learning controller that explicitly estimates base velocity and external disturbance forces. - source: mdpi.com
Learning Stable Bipedal Locomotion Skills for Quadrupedal Robots on Challenging Terrains with Automatic Fall Recovery - 2025-08-01 - Research Publication - Lite3 provides a second real-robot validation of a morphology-adaptive controller for bipedal quadruped locomotion and recovery. - source: nature.com
GRoQ-LoCO: Generalist and Robot-agnostic Quadruped Locomotion Control using Offline Datasets - 2025-05-16 - Research Publication - A single offline policy is tested across tracked Unitree and DEEP Robotics quadrupeds in simulation. - source: arxiv.org
Walking with Terrain Reconstruction: Learning to Traverse Risky Sparse Footholds - 2024-09-24 - Research Publication - A Lite3 uses reconstructed local terrain, proprioception and depth imagery to place its feet on highly sparse and narrow supports. - source: arxiv.org
PIE: Parkour with Implicit-Explicit Learning Framework for Legged Robots - 2024-08-25 - Research Publication - A Lite3 deploys a single-stage vision-and-proprioception policy for zero-shot parkour across large obstacles and an outdoor mountain route. - source: arxiv.org
fan-ziqi/robot_lab - 2024-07-06 - Research Publication - The source registers exact model-specific Isaac Lab environments for tracked quadruped, wheeled-quadruped and humanoid platforms. - source: github.com