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

Yingnan Zhao1, Xinmiao Wang1,2, Dewei Wang2,3, Xinzhe Liu2,4, Dan Lu1†,
Qilong Han1, Peng Liu5, Chenjia Bai2†
1College of Computer Science and Technology, Harbin Engineering University 2Institute of Artificial Intelligence (TeleAI), China Telecom 3School of Information Science and Technology, University of Science and Technology of China 4School of Information Science and Technology, ShanghaiTech University 5College of Computer Science and Technology, Harbin Institute of Technology
†Corresponding Authors
arXiv Code (coming soon...)

Video

Abstract

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.

Method

图片1

BibTeX

@misc{zhao2025adaptivehumanoidcontrolmultibehavior,
  title={Towards Adaptive Humanoid Control via Multi-Behavior Distillation and Reinforced Fine-Tuning}, 
  author={Yingnan Zhao and Xinmiao Wang and Dewei Wang and Xinzhe Liu and Dan Lu and Qilong Han and Peng Liu and Chenjia Bai},
  year={2025},
  eprint={2511.06371},
  archivePrefix={arXiv},
  primaryClass={cs.RO},
  url={https://arxiv.org/abs/2511.06371}, 
}