Deep Reinforcement Learning for Robotic Control in High-Dexterity Assembly Tasks — A Reward Curriculum Approach

International Journal of Semantic Computing, vol. 16, pp. 381–402

Abstract

For years, the fully-automated robotic assembly has been a highly sought-after technology in large-scale manufacturing. Yet it still struggles to find widespread implementation in industrial environments. Traditional programming has so far proven to be insufficient in providing the required flexibility and dexterity to solve complex assembly tasks. Research in robotic control using deep reinforcement learning (DRL) advances quickly, however, the transfer to real-world applications in industrial settings is lagging behind. In this study, we apply DRL for robotic motion control to a multi-body contact automotive assembly task. Our focus lies on optimizing the final performance on the real-world setup. We propose a reward-curriculum learning approach in combination with domain randomization to obtain both force-sensitivity and generalizability of the controller’s performance. We train the agent exclusively in simulation and successfully perform the Sim-to-Real transfer. Finally, we evaluate the controller’s performance and robustness on an industrial setup and reflect its adherence to the high standards of automotive production.

Authors 5

1. Lars Leyendecker Aachen Fraunhofer Institute for Production Technology IPT

   Fraunhofer Institute for Production Technology IPT

   Affiliation as printed

   Fraunhofer Institute for Production Technology IPT, Steinbachstraße 17, 52074 Aachen, Germany

2. Hans Aoyang Zhou Aachen Institute of Information Management in Mechanical Engineering

   RWTH Aachen University

   Affiliation as printed

   Institute of Information Management in Mechanical Engineering, RWTH Aachen University, Dennewartstraße 27, 52068 Aachen, Germany

3. Владимир Самсонов Aachen Institute of Information Management in Mechanical Engineering

   RWTH Aachen University

   Affiliation as printed

   Institute of Information Management in Mechanical Engineering, RWTH Aachen University, Dennewartstraße 27, 52068 Aachen, Germany

4. Marius Rittstieg

   BMW (Germany)

   Affiliation as printed

   Department of Innovation and Digitalization/Data Analytics at BMW AG, Petuelring 130, 80807 Munich, Germany

5. Daniel Lütticke Aachen Institute of Information Management in Mechanical Engineering

   RWTH Aachen University

   Affiliation as printed

   Institute of Information Management in Mechanical Engineering, RWTH Aachen University, Dennewartstraße 27, 52068 Aachen, Germany