A Reinforcement Learning-Based Framework for Robot Manipulation Skill Acquisition
This paper studies robot manipulation skill acquisition based on a proposed reinforcement learning framework. Robot can learn policy autonomously by interacting with environment with a better learning efficiency. Aiming at the manipulator operation task, a reward function design method based on obje...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9112186/ |
| id |
doaj-37b561cf8f4d4c53acb4096a4e930909 |
|---|---|
| record_format |
Article |
| spelling |
doaj-37b561cf8f4d4c53acb4096a4e9309092021-03-30T03:02:16ZengIEEEIEEE Access2169-35362020-01-01810842910843710.1109/ACCESS.2020.30011309112186A Reinforcement Learning-Based Framework for Robot Manipulation Skill AcquisitionDong Liu0https://orcid.org/0000-0003-3988-7879Zitu Wang1Binpeng Lu2Ming Cong3Honghua Yu4Qiang Zou5School of Mechanical Engineering, Dalian University of Technology, Dalian, ChinaSchool of Mechanical Engineering, Dalian University of Technology, Dalian, ChinaSchool of Mechanical Engineering, Dalian University of Technology, Dalian, ChinaSchool of Mechanical Engineering, Dalian University of Technology, Dalian, ChinaSchool of Mechanical Engineering, Dalian University of Technology, Dalian, ChinaSchool of Mechanical Engineering, Dalian University of Technology, Dalian, ChinaThis paper studies robot manipulation skill acquisition based on a proposed reinforcement learning framework. Robot can learn policy autonomously by interacting with environment with a better learning efficiency. Aiming at the manipulator operation task, a reward function design method based on objects configuration matching (OCM) is proposed. It is simple and suitable for most Pick and Place skills learning. Integrating robot and object state, high-level action set and the designed reward function, the Markov model of robot manipulator is built. An improved Proximal Policy Optimize algorithm with manipulation set as the output of Actor (MAPPO) is proposed as the main structure to construct the robot reinforcement learning framework. The framework combines with the Markov model to learn and optimize the skill policy. A same simulation environment as the real robot is set up, and three robot manipulation tasks are designed to verify the effectiveness and feasibility of the reinforcement learning framework for skill acquisition.https://ieeexplore.ieee.org/document/9112186/Robot skill acquisitionreinforcement learningreword functionMAPPO |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Dong Liu Zitu Wang Binpeng Lu Ming Cong Honghua Yu Qiang Zou |
| spellingShingle |
Dong Liu Zitu Wang Binpeng Lu Ming Cong Honghua Yu Qiang Zou A Reinforcement Learning-Based Framework for Robot Manipulation Skill Acquisition IEEE Access Robot skill acquisition reinforcement learning reword function MAPPO |
| author_facet |
Dong Liu Zitu Wang Binpeng Lu Ming Cong Honghua Yu Qiang Zou |
| author_sort |
Dong Liu |
| title |
A Reinforcement Learning-Based Framework for Robot Manipulation Skill Acquisition |
| title_short |
A Reinforcement Learning-Based Framework for Robot Manipulation Skill Acquisition |
| title_full |
A Reinforcement Learning-Based Framework for Robot Manipulation Skill Acquisition |
| title_fullStr |
A Reinforcement Learning-Based Framework for Robot Manipulation Skill Acquisition |
| title_full_unstemmed |
A Reinforcement Learning-Based Framework for Robot Manipulation Skill Acquisition |
| title_sort |
reinforcement learning-based framework for robot manipulation skill acquisition |
| publisher |
IEEE |
| series |
IEEE Access |
| issn |
2169-3536 |
| publishDate |
2020-01-01 |
| description |
This paper studies robot manipulation skill acquisition based on a proposed reinforcement learning framework. Robot can learn policy autonomously by interacting with environment with a better learning efficiency. Aiming at the manipulator operation task, a reward function design method based on objects configuration matching (OCM) is proposed. It is simple and suitable for most Pick and Place skills learning. Integrating robot and object state, high-level action set and the designed reward function, the Markov model of robot manipulator is built. An improved Proximal Policy Optimize algorithm with manipulation set as the output of Actor (MAPPO) is proposed as the main structure to construct the robot reinforcement learning framework. The framework combines with the Markov model to learn and optimize the skill policy. A same simulation environment as the real robot is set up, and three robot manipulation tasks are designed to verify the effectiveness and feasibility of the reinforcement learning framework for skill acquisition. |
| topic |
Robot skill acquisition reinforcement learning reword function MAPPO |
| url |
https://ieeexplore.ieee.org/document/9112186/ |
| work_keys_str_mv |
AT dongliu areinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT zituwang areinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT binpenglu areinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT mingcong areinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT honghuayu areinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT qiangzou areinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT dongliu reinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT zituwang reinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT binpenglu reinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT mingcong reinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT honghuayu reinforcementlearningbasedframeworkforrobotmanipulationskillacquisition AT qiangzou reinforcementlearningbasedframeworkforrobotmanipulationskillacquisition |
| _version_ |
1724184173515636736 |