mlsanityData to 14 Oct 2026

Learning Terrain-Specialized Policies for Adaptive Locomotion in Challenging Environments

arXiv 2509.20635 ↗First posted 25 Sep 2025cs.RO cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

Legged robots must exhibit robust and agile locomotion across diverse, unstructured terrains, a challenge exacerbated under blind locomotion settings where terrain information is unavailable. This work introduces a hierarchical reinforcement learning framework that leverages terrain-specialized policies and curriculum learning to enhance agility and tracking performance in complex environments. We validated our method on simulation, where our approach outperforms a generalist policy by up to 16% in success rate and achieves lower tracking errors as the velocity target increases, particularly on low-friction and discontinuous terrains, demonstrating superior adaptability and robustness across mixed-terrain scenarios.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 19 of its 23 references that are in the map. First posted on arXiv 25 Sep 2025.

Builds on

19 of 23 references.

References · 19

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2020Learning Quadrupedal Locomotion over Challenging Terraincited by 604
  3. 2021Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learningcited by 512
  4. 2021RMA: Rapid Motor Adaptation for Legged Robotscited by 508
  5. 2019Learning agile and dynamic motor skills for legged robotscited by 504
  6. 2022Learning robust perceptive locomotion for quadrupedal robots in the wildcited by 382
  7. 2023Orbit: A Unified Simulation Framework for Interactive Robot Learning Environmentscited by 295
  8. 2022Rapid Locomotion via Reinforcement Learningcited by 168
  9. 2018Learning to Walk via Deep Reinforcement Learningcited by 158
  10. 2023ANYmal Parkour: Learning Agile Navigation for Quadrupedal Robotscited by 152
  11. 2022Walk These Ways: Tuning Robot Control for Generalization with Multiplicity of Behaviorcited by 138
  12. 2019Data Efficient Reinforcement Learning for Legged Robotscited by 63
7 more
  1. 2022Perceptive Locomotion through Nonlinear Model Predictive Controlcited by 38
  2. 2023Lifelike Agility and Play in Quadrupedal Robots using Reinforcement Learning and Generative Pre-trained Modelscited by 37
  3. 2024Learning Quadruped Locomotion Using Differentiable Simulationcited by 32
  4. 2019Learning Generalizable Locomotion Skills with Hierarchical Reinforcement Learningcited by 13
  5. 2023Learning Agility and Adaptive Legged Locomotion via Curricular Hindsight Reinforcement Learningcited by 6
  6. 2021Learning Perceptual Locomotion on Uneven Terrains using Sparse Visual Observationscited by 5
  7. 2025Learning to Walk With Less: A Dyna-Style Approach to Quadrupedal Locomotioncited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026CTS-MoE: Implicit Terrain Adaptation via Mixture-of-Experts for Perceptive Locomotioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Motion Priors Reimagined: Adapting Flat-Terrain Skills for Complex Quadruped Mobility
  2. 2023Learning Robust, Agile, Natural Legged Locomotion Skills in the Wild
  3. 2025Parkour in the Wild: Learning a General and Extensible Agile Locomotion Policy Using Multi-expert Distillation and RL Fine-tuning
  4. 2024Learning to walk in confined spaces using 3D representation
  5. 2026Feasibility-Guided Planning over Multi-Specialized Locomotion Policies ↗
  6. 2025Dynamic Legged Ball Manipulation on Rugged Terrains with Hierarchical Reinforcement Learning
  7. 2025Whole-Body Constrained Learning for Legged Locomotion via Hierarchical Optimization
  8. 2025Attention-Based Map Encoding for Learning Generalized Legged Locomotion