A multi-task deep reinforcement learning framework based on curriculum learning and policy distillation for quadruped robot motor skill training

Deep reinforcement learning (RL) approaches are increasingly prominent in the field of robotics due to their adaptive decision-making capability. However, developing a single RL agent capable of performing multiple continuous control tasks for quadruped robots remains challenging. In this paper, a m...

Full description

Saved in:
Bibliographic Details
Main Authors: Liang Chen, Bo Shen, Jiale Hong
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:Systems Science & Control Engineering
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/21642583.2025.2498914
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850171741891985408
author Liang Chen
Bo Shen
Jiale Hong
author_facet Liang Chen
Bo Shen
Jiale Hong
author_sort Liang Chen
collection DOAJ
description Deep reinforcement learning (RL) approaches are increasingly prominent in the field of robotics due to their adaptive decision-making capability. However, developing a single RL agent capable of performing multiple continuous control tasks for quadruped robots remains challenging. In this paper, a multi-task deep RL framework based on curriculum learning and policy distillation is proposed, which aims to enhance the quadruped robot's motor performance across multiple continuous tasks by leveraging knowledge from expert skill teachers. The main novelties of the framework lie in the self-optimizing terrain curriculum learning strategy and the improved distillation loss function. The proposed self-optimizing terrain curriculum learning strategy for quadrupedal robots is designed to utilize Bayesian optimization to predict potential training terrains, thus effectively identifying the most suitable training curriculum. Additionally, the improved distillation loss function for RL weight optimization is proposed to enhance the transferability of the trained policy across diverse tasks. To validate the effectiveness of the proposed multi-task deep RL framework, the performance of the policy generated by the framework across diverse terrains is assessed. The experimental results demonstrate that the proposed multi-task deep RL framework could generate a unified policy that achieves excellent performance across multiple continuous control tasks for quadruped robots.
format Article
id doaj-art-ecb109ff8b2a454fb4af3aa5dd5aacc3
institution OA Journals
issn 2164-2583
language English
publishDate 2025-12-01
publisher Taylor & Francis Group
record_format Article
series Systems Science & Control Engineering
spelling doaj-art-ecb109ff8b2a454fb4af3aa5dd5aacc32025-08-20T02:20:13ZengTaylor & Francis GroupSystems Science & Control Engineering2164-25832025-12-0113110.1080/21642583.2025.2498914A multi-task deep reinforcement learning framework based on curriculum learning and policy distillation for quadruped robot motor skill trainingLiang Chen0Bo Shen1Jiale Hong2College of Information Science and Technology, Donghua University, Shanghai, People’s Republic of ChinaCollege of Information Science and Technology, Donghua University, Shanghai, People’s Republic of ChinaCollege of Information Science and Technology, Donghua University, Shanghai, People’s Republic of ChinaDeep reinforcement learning (RL) approaches are increasingly prominent in the field of robotics due to their adaptive decision-making capability. However, developing a single RL agent capable of performing multiple continuous control tasks for quadruped robots remains challenging. In this paper, a multi-task deep RL framework based on curriculum learning and policy distillation is proposed, which aims to enhance the quadruped robot's motor performance across multiple continuous tasks by leveraging knowledge from expert skill teachers. The main novelties of the framework lie in the self-optimizing terrain curriculum learning strategy and the improved distillation loss function. The proposed self-optimizing terrain curriculum learning strategy for quadrupedal robots is designed to utilize Bayesian optimization to predict potential training terrains, thus effectively identifying the most suitable training curriculum. Additionally, the improved distillation loss function for RL weight optimization is proposed to enhance the transferability of the trained policy across diverse tasks. To validate the effectiveness of the proposed multi-task deep RL framework, the performance of the policy generated by the framework across diverse terrains is assessed. The experimental results demonstrate that the proposed multi-task deep RL framework could generate a unified policy that achieves excellent performance across multiple continuous control tasks for quadruped robots.https://www.tandfonline.com/doi/10.1080/21642583.2025.2498914Reinforcement learningmulti-taskpolicy distillationcurriculumquadrupedal robots
spellingShingle Liang Chen
Bo Shen
Jiale Hong
A multi-task deep reinforcement learning framework based on curriculum learning and policy distillation for quadruped robot motor skill training
Systems Science & Control Engineering
Reinforcement learning
multi-task
policy distillation
curriculum
quadrupedal robots
title A multi-task deep reinforcement learning framework based on curriculum learning and policy distillation for quadruped robot motor skill training
title_full A multi-task deep reinforcement learning framework based on curriculum learning and policy distillation for quadruped robot motor skill training
title_fullStr A multi-task deep reinforcement learning framework based on curriculum learning and policy distillation for quadruped robot motor skill training
title_full_unstemmed A multi-task deep reinforcement learning framework based on curriculum learning and policy distillation for quadruped robot motor skill training
title_short A multi-task deep reinforcement learning framework based on curriculum learning and policy distillation for quadruped robot motor skill training
title_sort multi task deep reinforcement learning framework based on curriculum learning and policy distillation for quadruped robot motor skill training
topic Reinforcement learning
multi-task
policy distillation
curriculum
quadrupedal robots
url https://www.tandfonline.com/doi/10.1080/21642583.2025.2498914
work_keys_str_mv AT liangchen amultitaskdeepreinforcementlearningframeworkbasedoncurriculumlearningandpolicydistillationforquadrupedrobotmotorskilltraining
AT boshen amultitaskdeepreinforcementlearningframeworkbasedoncurriculumlearningandpolicydistillationforquadrupedrobotmotorskilltraining
AT jialehong amultitaskdeepreinforcementlearningframeworkbasedoncurriculumlearningandpolicydistillationforquadrupedrobotmotorskilltraining
AT liangchen multitaskdeepreinforcementlearningframeworkbasedoncurriculumlearningandpolicydistillationforquadrupedrobotmotorskilltraining
AT boshen multitaskdeepreinforcementlearningframeworkbasedoncurriculumlearningandpolicydistillationforquadrupedrobotmotorskilltraining
AT jialehong multitaskdeepreinforcementlearningframeworkbasedoncurriculumlearningandpolicydistillationforquadrupedrobotmotorskilltraining