mlsanityData to 14 Oct 2026

Multi-expert learning of adaptive legged locomotion

arXiv 2012.05810 ↗First posted 10 Dec 2020cs.RO cs.AI cs.LGcited by 67

Not on the map. Why?

Abstract (arXiv)

Achieving versatile robot locomotion requires motor skills which can adapt to previously unseen situations. We propose a Multi-Expert Learning Architecture (MELA) that learns to generate adaptive skills from a group of representative expert skills. During training, MELA is first initialised by a distinct set of pre-trained experts, each in a separate deep neural network (DNN). Then by learning the combination of these DNNs using a Gating Neural Network (GNN), MELA can acquire more specialised experts and transitional skills across various locomotion modes. During runtime, MELA constantly blends multiple DNNs and dynamically synthesises a new DNN to produce adaptive behaviours in response to changing situations. This approach leverages the advantages of trained expert skills and the fast online synthesis of adaptive policies to generate responsive motor skills during the changing tasks. Using a unified MELA framework, we demonstrated successful multi-skill locomotion on a real quadruped robot that performed coherent trotting, steering, and fall recovery autonomously, and showed the merit of multi-expert learning generating behaviours which can adapt to unseen scenarios.

Key facts. Cited by 67 papers in the map (data through 14 Oct 2026). First posted on arXiv 10 Dec 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 7; 2022: 13; 2023: 14; 2024: 14; 2025: 10; 2026: 9.

67 papers.

Sort:

By area:

  1. 2023Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewardscited by 243
  2. 2024Deep Reinforcement Learning for Robotics: A Survey of Real-World Successescited by 176
  3. 2023Robot Parkour Learningcited by 149
  4. 2022Concurrent Training of a Control Policy and a State Estimator for Dynamic and Robust Legged Locomotioncited by 88
  5. 2024Agile But Safe: Learning Collision-Free High-Speed Legged Locomotioncited by 73
  6. 2022GenLoco: Generalized Locomotion Controllers for Quadrupedal Robotscited by 48
  7. 2024Towards Generalist Robot Learning from Internet Video: A Surveycited by 40
  8. 2023Hybrid Internal Model: Learning Agile Legged Locomotion with Simulated Robot Responsecited by 39
  9. 2023Lifelike Agility and Play in Quadrupedal Robots using Reinforcement Learning and Generative Pre-trained Modelscited by 37
  10. 2022Imitate and Repurpose: Learning Reusable Robot Movement Skills From Human and Animal Behaviorscited by 30
  11. 2022Learning Free Gait Transition for Quadruped Robots via Phase-Guided Controllercited by 29
  12. 2022Learning Torque Control for Quadrupedal Locomotioncited by 29
38 more
  1. 2021GLiDE: Generalizable Quadrupedal Locomotion in Diverse Environments with a Centroidal Modelcited by 25
  2. 2025MoE-Loco: Mixture of Experts for Multitask Locomotioncited by 23
  3. 2023Learning Bipedal Walking for Humanoids with Current Feedbackcited by 18
  4. 2022Learning and Deploying Robust Locomotion Policies with Minimal Dynamics Randomizationcited by 17
  5. 2021Reinforcement Learning with Evolutionary Trajectory Generator: A General Approach for Quadrupedal Locomotioncited by 14
  6. 2024Rethinking Robustness Assessment: Adversarial Attacks on Learning-based Quadrupedal Locomotion Controllerscited by 14
  7. 2024SoloParkour: Constrained Reinforcement Learning for Visual Locomotion from Privileged Experiencecited by 14
  8. 2022Learning Low-Frequency Motion Control for Robust and Dynamic Robot Locomotioncited by 12
  9. 2023Active Policy Improvement from Multiple Black-box Oraclescited by 11
  10. 2024AddBiomechanics Dataset: Capturing the Physics of Human Motion at Scalecited by 11
  11. 2024Generalizing Consistency Policy to Visual RL with Prioritized Proximal Experience Regularizationcited by 11
  12. 2022Cascaded Compositional Residual Learning for Complex Interactive Behaviorscited by 10
  13. 2023Hierarchical generative modelling for autonomous robotscited by 9
  14. 2024One Filter to Deploy Them All: Robust Safety for Quadrupedal Navigation in Unknown Environmentscited by 9
  15. 2025SkillBlender: Towards Versatile Humanoid Whole-Body Loco-Manipulation via Skill Blendingcited by 9
  16. 2023Intrinsic Language-Guided Exploration for Complex Long-Horizon Robotic Manipulation Taskscited by 8
  17. 2024PA-LOCO: Learning Perturbation-Adaptive Locomotion for Quadruped Robotscited by 8
  18. 2025Dynamic Mixture of Progressive Parameter-Efficient Expert Library for Lifelong Robot Learningcited by 8
  19. 2021Versatile modular neural locomotion control with fast learningcited by 6
  20. 2023Identifying Important Sensory Feedback for Learning Locomotion Skillscited by 6
  21. 2024Constrained Skill Discovery: Quadruped Locomotion with Unsupervised Reinforcement Learningcited by 6
  22. 2021Learning Perceptual Locomotion on Uneven Terrains using Sparse Visual Observationscited by 5
  23. 2024Distilling Reinforcement Learning Policies for Interpretable Robot Locomotion: Gradient Boosting Machines and Symbolic Regressioncited by 5
  24. 2025Flexible Multitask Learning with Factorized Diffusion Policycited by 5
  25. 2022Imitation and Adaptation Based on Consistency: A Quadruped Robot Imitates Animals from Videos Using Deep Reinforcement Learningcited by 4
  26. 2022Generating a Terrain-Robustness Benchmark for Legged Locomotion: A Prototype via Terrain Authoring and Active Learningcited by 3
  27. 2023Roll-Drop: accounting for observation noise with a single parametercited by 3
  28. 2021Learning Setup Policies: Reliable Transition Between Locomotion Behaviourscited by 2
  29. 2021What Robot do I Need? Fast Co-Adaptation of Morphology and Control using Graph Neural Networkscited by 2
  30. 2022Exploring with Sticky Mittens: Reinforcement Learning with Expert Interventions via Option Templatescited by 2
  31. 2022A Transferable Legged Mobile Manipulation Framework Based on Disturbance Predictive Controlcited by 2
  32. 2024An Efficient Model-Based Approach on Learning Agile Motor Skills without Reinforcementcited by 2
  33. 2026Closing the Reality Gap: Zero-Shot Sim-to-Real Deployment for Dexterous Force-Based Grasping and Manipulationcited by 2
  34. 2026Toward Reliable Sim-to-Real Predictability for MoE-based Robust Quadrupedal Locomotioncited by 2
  35. 2023Boosting Feedback Efficiency of Interactive Reinforcement Learning by Adaptive Learning from Scorescited by 1
  36. 2024Behavior evolution-inspired approach to walking gait reinforcement training for quadruped robotscited by 1
  37. 2024QuadWBG: Generalizable Quadrupedal Whole-Body Graspingcited by 1
  38. 2025Bio-Inspired Plastic Neural Networks for Zero-Shot Out-of-Distribution Generalization in Complex Animal-Inspired Robotscited by 1

50 most cited of 67 All 67 (JSON)