CMoE: Contrastive Mixture of Experts for Motion Control and Terrain Adaptation of Humanoid Robots

• URL: http://arxiv.org/abs/2603.03067v1
• Date: Tue, 03 Mar 2026 15:06:30 GMT
• Title: CMoE: Contrastive Mixture of Experts for Motion Control and Terrain Adaptation of Humanoid Robots
• Abstract summary: Vanilla mixture of experts (MoE) framework is theoretically capable of modeling diverse terrain features.<n>We introduce CMoE, a novel single-stage reinforcement learning framework that integrates contrastive learning to refine expert activation distributions.<n>Results demonstrate that CMoE enables the robot to traverse continuous steps up to 20 cm high and gaps up to 80 cm wide.
• Score: 15.062634107888982
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: For effective deployment in real-world environments, humanoid robots must autonomously navigate a diverse range of complex terrains with abrupt transitions. While the Vanilla mixture of experts (MoE) framework is theoretically capable of modeling diverse terrain features, in practice, the gating network exhibits nearly uniform expert activations across different terrains, weakening the expert specialization and limiting the model's expressive power. To address this limitation, we introduce CMoE, a novel single-stage reinforcement learning framework that integrates contrastive learning to refine expert activation distributions. By imposing contrastive constraints, CMoE maximizes the consistency of expert activations within the same terrain while minimizing their similarity across different terrains, thereby encouraging experts to specialize in distinct terrain types. We validated our approach on the Unitree G1 humanoid robot through a series of challenging experiments. Results demonstrate that CMoE enables the robot to traverse continuous steps up to 20 cm high and gaps up to 80 cm wide, while achieving robust and natural gait across diverse mixed terrains, surpassing the limits of existing methods. To support further research and foster community development, we release our code publicly.

Related papers

• CoRe-MoE: Contrastive Reweighted Mixture of Experts for Multi-Terrain Humanoid Locomotion with Gait Adaptation [18.803751724563448]
  We propose CoRe-MoE, a two-stage reinforcement learning framework that decouples gait generation from terrain adaptation.<n>In the first stage, a stable locomotion policy is learned to produce natural walking and running behaviors with smooth transitions.<n>In the second stage, a terrain-aware MoE branch is introduced, and the gating network is trained with a contrastive objective to learn structured terrain representations.
  arXiv  Detail & Related papers  (2026-06-03T10:51:46Z)
• Perceptive Humanoid Parkour: Chaining Dynamic Human Skills via Motion Matching [77.28042137892943]
  We present Perceptive Humanoid Parkour (PHP), a modular framework that enables humanoid robots to autonomously perform long-horizon, vision-based parkour.<n>We train motion-tracking reinforcement learning expert policies for these composed motions, and distill them into a single depth-based, multi-skill student policy.<n>We validate our framework with extensive real-world experiments on a Unitree G1 humanoid robot.
  arXiv  Detail & Related papers  (2026-02-17T18:59:11Z)
• PILOT: A Perceptive Integrated Low-level Controller for Loco-manipulation over Unstructured Scenes [18.916978556216097]
  PILOT is a single-stage reinforcement learning framework tailored for perceptive loco-manipulation.<n>It synergizes perceptive locomotion and expansive whole-body control within a single policy.<n> PILOT demonstrates superior stability, command tracking precision, and terrain traversability compared to existing baselines.
  arXiv  Detail & Related papers  (2026-01-24T12:25:20Z)
• Heterogeneous Robot Collaboration in Unstructured Environments with Grounded Generative Intelligence [54.91177026001217]
  Large language model (LLM)-enabled teaming methods typically assume well-structured and known environments.<n>We present SPINE-HT, a framework that addresses these limitations by grounding the reasoning abilities of LLMs in the context of a heterogeneous robot team.<n>Our framework achieves nearly twice the success rate compared to prior LLM-enabled heterogeneous teaming approaches.
  arXiv  Detail & Related papers  (2025-10-30T18:24:38Z)
• Recognizing Actions from Robotic View for Natural Human-Robot Interaction [52.00935005918032]
  Natural Human-Robot Interaction (N-HRI) requires robots to recognize human actions at varying distances and states, regardless of whether the robot itself is in motion or stationary.<n>Existing benchmarks for N-HRI fail to address the unique complexities in N-HRI due to limited data, modalities, task categories, and diversity of subjects and environments.<n>We introduce (Action from Robotic View) a large-scale dataset for perception-centric robotic views prevalent in mobile service robots.
  arXiv  Detail & Related papers  (2025-07-30T09:48:34Z)
• MoE-Loco: Mixture of Experts for Multitask Locomotion [52.04025933292957]
  We present MoE-Loco, a framework for multitask locomotion for legged robots.<n>Our method enables a single policy to handle diverse terrains, while supporting quadrupedal and bipedal gaits.
  arXiv  Detail & Related papers  (2025-03-11T15:53:54Z)
• Simultaneous Multi-Robot Motion Planning with Projected Diffusion Models [57.45019514036948]
  Simultaneous MRMP Diffusion (SMD) is a novel approach integrating constrained optimization into the diffusion sampling process to produce collision-free, kinematically feasible trajectories.<n>The paper introduces a comprehensive MRMP benchmark to evaluate trajectory planning algorithms across scenarios with varying robot densities, obstacle complexities, and motion constraints.
  arXiv  Detail & Related papers  (2025-02-05T20:51:28Z)
• Leveraging Hyperbolic Embeddings for Coarse-to-Fine Robot Design [40.01142267374432]
  Multi-cellular robot design aims to create robots comprised of numerous cells that can be efficiently controlled to perform diverse tasks. Previous research has demonstrated the ability to generate robots for various tasks, but these approaches often optimize robots directly in the vast design space. This paper presents a novel coarse-to-fine method for designing multi-cellular robots.
  arXiv  Detail & Related papers  (2023-11-01T11:56:32Z)
• Transferring Foundation Models for Generalizable Robotic Manipulation [82.12754319808197]
  We propose a novel paradigm that effectively leverages language-reasoning segmentation mask generated by internet-scale foundation models.<n>Our approach can effectively and robustly perceive object pose and enable sample-efficient generalization learning.<n>Demos can be found in our submitted video, and more comprehensive ones can be found in link1 or link2.
  arXiv  Detail & Related papers  (2023-06-09T07:22:12Z)
• Deep Multi-Agent Reinforcement Learning with Hybrid Action Spaces based on Maximum Entropy [0.0]
  We propose Deep Multi-Agent Hybrid Soft Actor-Critic (MAHSAC) to handle multi-agent problems with hybrid action spaces. This algorithm follows the centralized training but decentralized execution (CTDE) paradigm, and extend the Soft Actor-Critic algorithm (SAC) to handle hybrid action space problems. Our experiences are running on an easy multi-agent particle world with a continuous observation and discrete action space, along with some basic simulated physics.
  arXiv  Detail & Related papers  (2022-06-10T13:52:59Z)
• Learning to Move with Affordance Maps [57.198806691838364]
  The ability to autonomously explore and navigate a physical space is a fundamental requirement for virtually any mobile autonomous agent. Traditional SLAM-based approaches for exploration and navigation largely focus on leveraging scene geometry. We show that learned affordance maps can be used to augment traditional approaches for both exploration and navigation, providing significant improvements in performance.
  arXiv  Detail & Related papers  (2020-01-08T04:05:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.