ETRI-Knowledge Sharing Plaform



논문 검색
구분 SCI
연도 ~ 키워드


학술대회 Learning Control Policy with Previous Experiences from Robot Simulator
Cited 2 time in scopus Download 0 time Share share facebook twitter linkedin kakaostory
이동훈, 김현석, 김성현, 박찬원, 박준희
International Conference on Information and Communication Technology Convergence (ICTC) 2020, pp.863-865
20ZR1100, 자율적으로 연결·제어·진화하는 초연결 지능화 기술 연구, 박준희
Advances in deep reinforcement learning enabled cost-efficient training of control policy of physical robot actions from robot simulators. Learning control policy in a simulated environment is cost-efficient over learning in a real environment. Reward engineering is one of the key components to train efficient control policy. For tasks with long horizons such as navigation and manipulation, a sparse reward is providing limited information. The robot simulator for a physical engine of physical robot manipulation has made it easy for researchers in the field of deep reinforcement learning to simulate complicated robot manipulation environments. In this paper, A robot manipulation simulator and a deep RL framework are utilized for implement a training control policy by utilizing previous experiences. For implementation, Recent innovation Hindsight Experience Replay (HER) algorithms with previous experiences to calculate dense rewards from a sparse reward is leveraged. Proposed implementation showed an approach to investigate the reward engineering method to formulate dense reward in robot manipulator tasks.
KSP 제안 키워드
Control policy, Cost-efficient, Deep reinforcement learning, Experience replay, Key Components, Learning control, Limited information, Real environment, Reinforcement Learning(RL), Robot manipulation, Robot manipulator