# Parkour in the Wild: Learning a General and Extensible Agile Locomotion Policy Using Multi-expert Distillation and RL Fine-tuning

arXiv: https://arxiv.org/abs/2505.11164 · HTML: https://mlsanity.com/paper/2505.11164/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 May 2025.

## Built on by (most cited first, 26 of 26)

- [RSL-RL: A Learning Library for Robotics Research](https://mlsanity.com/paper/2509.10771/) (2025)
- [Perceptive Humanoid Parkour: Chaining Dynamic Human Skills via Motion Matching](https://mlsanity.com/paper/2602.15827/) (2026)
- [Hiking in the Wild: A Scalable Perceptive Parkour Framework for Humanoids](https://mlsanity.com/paper/2601.07718/) (2026)
- [ZEST: Zero-shot Embodied Skill Transfer for Athletic Robot Control](https://mlsanity.com/paper/2602.00401/) (2026)
- [Olaf: Bringing an Animated Character to Life in the Physical World](https://mlsanity.com/paper/2512.16705/) (2025)
- [Humanoid Occupancy: Enabling A Generalized Multimodal Occupancy Perception System on Humanoid Robots](https://mlsanity.com/paper/2507.20217/) (2025)
- [Enhancing Tactile-based Reinforcement Learning for Robotic Control](https://mlsanity.com/paper/2510.21609/) (2025)
- [TTT-Parkour: Rapid Test-Time Training for Perceptive Robot Parkour](https://mlsanity.com/paper/2602.02331/) (2026)
- [DeFM: Learning Foundation Representations from Depth for Robotics](https://mlsanity.com/paper/2601.18923/) (2026)
- [Towards Quadrupedal Jumping and Walking for Dynamic Locomotion using Reinforcement Learning](https://mlsanity.com/paper/2510.24584/) (2025)
- [APEX: Learning Adaptive High-Platform Traversal for Humanoid Robots](https://mlsanity.com/paper/2602.11143/) (2026)
- [roto 2.0: The Robot Tactile Olympiad](https://mlsanity.com/paper/2605.21429/) (2026)
- [Efficient On-policy Visual-RL via Stochastic Decoupled Policy Gradient](https://mlsanity.com/paper/2605.26478/) (2026)
- [GeCCo -- a Generalist Contact-Conditioned Policy for Loco-Manipulation Skills on Legged Robots](https://mlsanity.com/paper/2509.17582/) (2025)
- [PGTT: Phase-Guided Terrain Traversal for Perceptive Legged Locomotion](https://mlsanity.com/paper/2510.18348/) (2025)
- [Learning Terrain Aware Bipedal Locomotion via Reduced Dimensional Perceptual Representations](https://mlsanity.com/paper/2512.12993/) (2025)
- [Online World Modeling Enables Real-World Inverse Reinforcement Learning from Observation](https://mlsanity.com/paper/2602.24121/) (2026)
- [CoRe-MoE: Contrastive Reweighted Mixture of Experts for Multi-Terrain Humanoid Locomotion with Gait Adaptation](https://mlsanity.com/paper/2606.04718/) (2026)
- [LadderMan: Learning Humanoid Perceptive Ladder Climbing](https://mlsanity.com/paper/2606.05873/) (2026)
- [SWAP: Symmetric Equivariant World-Model for Agile Robot Parkour](https://mlsanity.com/paper/2606.19928/) (2026)
- [EgoHTR: Egocentric 4D Demonstrations of Human Terrain Traversal](https://mlsanity.com/paper/2607.13472/) (2026)
- [Continual-RL for Generalization in Autonomous Racing on the RoboRacer Platform](https://mlsanity.com/paper/2607.24320/) (2026)
- [Actuator Dynamics Curricula for Narrow-Viability Tasks in Legged Robot Learning](https://mlsanity.com/paper/2609.09492/) (2026)
- [Teaching a Robot Dog New Tricks: Diverse Quadruped Skills via Combined Reinforcement and Imitation Learning with Adversarial Task Selection](https://mlsanity.com/paper/2610.10601/) (2026)
- [A Balanced Data Diet: Addressing the Exploration Bottleneck in Mega-Scale RL for Robot Control](https://mlsanity.com/paper/2610.12465/) (2026)
- [GLST: Global-Local Spatio-Temporal Memory Decomposition Framework for Online Chained-Goal Navigation](https://mlsanity.com/paper/2610.17927/) (2026)

## Abstract (arXiv)

Legged robots are well-suited for navigating terrains inaccessible to wheeled robots, making them ideal for applications in search and rescue or space exploration. However, current control methods often struggle to generalize across diverse, unstructured environments. This paper introduces a novel framework for agile locomotion of legged robots by combining multi-expert distillation with reinforcement learning (RL) fine-tuning to achieve robust generalization. Initially, terrain-specific expert policies are trained to develop specialized locomotion skills. These policies are then distilled into a unified foundation policy via the DAgger algorithm. The distilled policy is subsequently fine-tuned using RL on a broader terrain set, including real-world 3D scans. The framework allows further adaptation to new terrains through repeated fine-tuning. The proposed policy leverages depth images as exteroceptive inputs, enabling robust navigation across diverse, unstructured terrains. Experimental results demonstrate significant performance improvements over existing methods in synthesizing multi-terrain skills into a single controller. Deployment on the ANYmal D robot validates the policy's ability to navigate complex environments with agility and robustness, setting a new benchmark for legged robot locomotion.