TY - GEN
T1 - Sim-to-Real Transfer of Robotic Assembly with Visual Inputs Using CycleGAN and Force Control
AU - Yuan, Chengjie
AU - Shi, Yunlei
AU - Feng, Qian
AU - Chang, Chunyang
AU - Liu, Michael
AU - Chen, Zhaopeng
AU - Knoll, Alois Christian
AU - Zhang, Jianwei
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Recently, deep reinforcement learning (RL) has shown some impressive successes in robotic manipulation applications. However, training robots in the real world is nontrivial owing to sample efficiency and safety concerns. Sim-to-real transfer is proposed to address the aforementioned concerns but introduces a new issue called the reality gap. In this work, we introduce a sim-to-real learning framework for vision-based assembly tasks and perform training in a simulated environment by employing inputs from a single camera to address the aforementioned issues. We present a domain adaptation method based on cycle-consistent generative adversarial networks (CycleGAN) and a force control transfer approach to bridge the reality gap. We demonstrate that the proposed framework trained in a simulated environment can be successfully transferred to a real peg-in-hole setup.
AB - Recently, deep reinforcement learning (RL) has shown some impressive successes in robotic manipulation applications. However, training robots in the real world is nontrivial owing to sample efficiency and safety concerns. Sim-to-real transfer is proposed to address the aforementioned concerns but introduces a new issue called the reality gap. In this work, we introduce a sim-to-real learning framework for vision-based assembly tasks and perform training in a simulated environment by employing inputs from a single camera to address the aforementioned issues. We present a domain adaptation method based on cycle-consistent generative adversarial networks (CycleGAN) and a force control transfer approach to bridge the reality gap. We demonstrate that the proposed framework trained in a simulated environment can be successfully transferred to a real peg-in-hole setup.
UR - http://www.scopus.com/inward/record.url?scp=85147325971&partnerID=8YFLogxK
U2 - 10.1109/ROBIO55434.2022.10011878
DO - 10.1109/ROBIO55434.2022.10011878
M3 - Conference contribution
AN - SCOPUS:85147325971
T3 - 2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022
SP - 1426
EP - 1432
BT - 2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 IEEE International Conference on Robotics and Biomimetics, ROBIO 2022
Y2 - 5 December 2022 through 9 December 2022
ER -