paper-with-me

홈 › Papers

MoE-Loco: Mixture of Experts for Multitask Locomotion

2025-03-11 · Runhan Huang, Shaoting Zhu, Yilun Du, Hang Zhao

We present MoE-Loco, a Mixture of Experts (MoE) framework for multitask locomotion for legged robots. Our method enables a single policy to handle diverse terrains, including bars, pits, stairs, slopes, and baffles, while supporting quadrupedal and bipedal gaits. Using MoE, we mitigate the gradient conflicts that typically arise in multitask reinforcement learning, improving both training efficiency and performance. Our experiments demonstrate that different experts naturally specialize in distinct locomotion behaviors, which can be leveraged for task migration and skill composition. We further validate our approach in both simulation and real-world deployment, showcasing its robustness and adaptability.

📄 PDF Abstract BibTeX arXiv:2503.08564

Code (0)

등록된 구현이 없습니다.

Tasks

Mixture-of-Experts

Methods 이 논문이 사용한 방법론

MoE 설명 없음

Similar Papers 제목 키워드 기반

Mixture-of-Experts RL for Fault-Tolerant Legged Locomotion

2026-06-24 · Giulio Turrisi, Ozan Pali, Luca Oneto, Claudio Semini arxiv

Legged robots deployed in planetary exploration and other remote environments must maintain reliable locomotion despite actuator failures and challenging terrain conditions. Although reinforcement learning has achieved s…

Reinforcement Learning

CoRe-MoE: Contrastive Reweighted Mixture of Experts for Multi-Terrain Humanoid Locomotion with Gait Adaptation

2026-06-03 · Kailun Huang, Zikang Xie, Yanzhe Xie, Panpan Liao 외 arxiv

Humans primarily rely on walking and running to traverse complex terrains. Similarly, humanoid robots should be able to smoothly transition between walking and running while maintaining natural and stable locomotion. How…

Reinforcement Learning

Toward Reliable Sim-to-Real Predictability for MoE-based Robust Quadrupedal Locomotion

2026-01-31 · Tianyang Wu, Hanwei Guo, Yuhang Wang, Junshu Yang 외 arxiv

Reinforcement learning has shown strong promise for quadrupedal agile locomotion, even with proprioception-only sensing. In practice, however, sim-to-real gap and reward overfitting in complex terrains can produce polici…

Reinforcement Learning

CTS-MoE: Implicit Terrain Adaptation via Mixture-of-Experts for Perceptive Locomotion

2026-06-17 · Francisco Affonso, Matheus P. Angarola, Ana Luiza Mineiro, Aditya Potnis 외 arxiv

Perceptive legged locomotion over discontinuous terrain (e.g., stairs, gaps, and obstacles) requires adaptive behavior, as a single conservative gait cannot produce the anticipatory maneuvers needed for abrupt topology c…

Reinforcement Learning

DexDribbler: Learning Dexterous Soccer Manipulation via Dynamic Supervision

2024-03-21 · Yutong Hu, Kehan Wen, Fisher Yu

Learning dexterous locomotion policy for legged robots is becoming increasingly popular due to its ability to handle diverse terrains and resemble intelligent behaviors. However, joint manipulation of moving objects and …