Salvato in:
Dettagli Bibliografici
Autori principali: Qiu, Yu, Lin, Xin, Wang, Jingbo, Li, Xiangtai, Qi, Lu, Yang, Ming-Hsuan
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2502.03035
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910814892457984
author Qiu, Yu
Lin, Xin
Wang, Jingbo
Li, Xiangtai
Qi, Lu
Yang, Ming-Hsuan
author_facet Qiu, Yu
Lin, Xin
Wang, Jingbo
Li, Xiangtai
Qi, Lu
Yang, Ming-Hsuan
contents Adaptation to unpredictable damages is crucial for autonomous legged robots, yet existing methods based on multi-policy or meta-learning frameworks face challenges like limited generalization and complex maintenance. To address this issue, we first analyze and summarize eight types of damage scenarios, including sensor failures and joint malfunctions. Then, we propose a novel, model-free, two-stage training framework, Unified Malfunction Controller (UMC), incorporating a masking mechanism to enhance damage resilience. Specifically, the model is initially trained with normal environments to ensure robust performance under standard conditions. In the second stage, we use masks to prevent the legged robot from relying on malfunctioning limbs, enabling adaptive gait and movement adjustments upon malfunction. Experimental results demonstrate that our approach improves the task completion capability by an average of 36% for the transformer and 39% for the MLP across three locomotion tasks. The source code and trained models will be made available to the public.
format Preprint
id arxiv_https___arxiv_org_abs_2502_03035
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle UMC: Unified Resilient Controller for Legged Robots with Joint Malfunctions
Qiu, Yu
Lin, Xin
Wang, Jingbo
Li, Xiangtai
Qi, Lu
Yang, Ming-Hsuan
Robotics
Adaptation to unpredictable damages is crucial for autonomous legged robots, yet existing methods based on multi-policy or meta-learning frameworks face challenges like limited generalization and complex maintenance. To address this issue, we first analyze and summarize eight types of damage scenarios, including sensor failures and joint malfunctions. Then, we propose a novel, model-free, two-stage training framework, Unified Malfunction Controller (UMC), incorporating a masking mechanism to enhance damage resilience. Specifically, the model is initially trained with normal environments to ensure robust performance under standard conditions. In the second stage, we use masks to prevent the legged robot from relying on malfunctioning limbs, enabling adaptive gait and movement adjustments upon malfunction. Experimental results demonstrate that our approach improves the task completion capability by an average of 36% for the transformer and 39% for the MLP across three locomotion tasks. The source code and trained models will be made available to the public.
title UMC: Unified Resilient Controller for Legged Robots with Joint Malfunctions
topic Robotics
url https://arxiv.org/abs/2502.03035