Towards Adaptive Humanoid Control via Multi-Behavior Distillation and Reinforced Fine-Tuning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhao, Yingnan, Wang, Xinmiao, Wang, Dewei, Liu, Xinzhe, Lu, Dan, Han, Qilong, Liu, Peng, Bai, Chenjia
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911652702584832
author Zhao, Yingnan
Wang, Xinmiao
Wang, Dewei
Liu, Xinzhe
Lu, Dan
Han, Qilong
Liu, Peng
Bai, Chenjia
author_facet Zhao, Yingnan
Wang, Xinmiao
Wang, Dewei
Liu, Xinzhe
Lu, Dan
Han, Qilong
Liu, Peng
Bai, Chenjia
contents Humanoid robots are promising to learn a diverse set of human-like locomotion behaviors, including standing up, walking, running, and jumping. However, existing methods predominantly require training independent policies for each skill, yielding behavior-specific controllers that exhibit limited generalization and brittle performance when deployed on irregular terrains and in diverse situations. To address this challenge, we propose Adaptive Humanoid Control (AHC) that adopts a two-stage framework to learn an adaptive humanoid locomotion controller across different skills and terrains. Specifically, we first train several primary locomotion policies and perform a multi-behavior distillation process to obtain a basic multi-behavior controller, facilitating adaptive behavior switching based on the environment. Then, we perform reinforced fine-tuning by collecting online feedback in performing adaptive behaviors on more diverse terrains, enhancing terrain adaptability for the controller. We conduct experiments in both simulation and real-world experiments in Unitree G1 robots. The results show that our method exhibits strong adaptability across various situations and terrains. Project website: https://ahc-humanoid.github.io.
format Preprint
id arxiv_https___arxiv_org_abs_2511_06371
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Adaptive Humanoid Control via Multi-Behavior Distillation and Reinforced Fine-Tuning
Zhao, Yingnan
Wang, Xinmiao
Wang, Dewei
Liu, Xinzhe
Lu, Dan
Han, Qilong
Liu, Peng
Bai, Chenjia
Robotics
Humanoid robots are promising to learn a diverse set of human-like locomotion behaviors, including standing up, walking, running, and jumping. However, existing methods predominantly require training independent policies for each skill, yielding behavior-specific controllers that exhibit limited generalization and brittle performance when deployed on irregular terrains and in diverse situations. To address this challenge, we propose Adaptive Humanoid Control (AHC) that adopts a two-stage framework to learn an adaptive humanoid locomotion controller across different skills and terrains. Specifically, we first train several primary locomotion policies and perform a multi-behavior distillation process to obtain a basic multi-behavior controller, facilitating adaptive behavior switching based on the environment. Then, we perform reinforced fine-tuning by collecting online feedback in performing adaptive behaviors on more diverse terrains, enhancing terrain adaptability for the controller. We conduct experiments in both simulation and real-world experiments in Unitree G1 robots. The results show that our method exhibits strong adaptability across various situations and terrains. Project website: https://ahc-humanoid.github.io.
title Towards Adaptive Humanoid Control via Multi-Behavior Distillation and Reinforced Fine-Tuning
topic Robotics
url https://arxiv.org/abs/2511.06371