Reinforcement learning based on movement primitives for contact tasks

Young Loul Kim, Kuk Hyun Ahn, Jae Bok Song

    Research output: Contribution to journalArticlepeer-review

    43 Citations (Scopus)

    Abstract

    Recently, robot learning through deep reinforcement learning has incorporated various robot tasks through deep neural networks, without using specific control or recognition algorithms. However, this learning method is difficult to apply to the contact tasks of a robot, due to the exertion of excessive force from the random search process of reinforcement learning. Therefore, when applying reinforcement learning to contact tasks, solving the contact problem using an existing force controller is necessary. A neural-network-based movement primitive (NNMP) that generates a continuous trajectory which can be transmitted to the force controller and learned through a deep deterministic policy gradient (DDPG) algorithm is proposed for this study. In addition, an imitation learning algorithm suitable for NNMP is proposed such that the trajectories similar to the demonstration trajectory are stably generated. The performance of the proposed algorithms was verified using a square peg-in-hole assembly task with a tolerance of 0.1 mm. The results confirm that the complicated assembly trajectory can be learned stably through NNMP by the proposed imitation learning algorithm, and that the assembly trajectory is improved by learning the proposed NNMP through the DDPG algorithm.

    Original languageEnglish
    Article number101863
    JournalRobotics and Computer-Integrated Manufacturing
    Volume62
    DOIs
    Publication statusPublished - 2020 Apr

    Bibliographical note

    Funding Information:
    This work was supported by an IITP grant funded by the Korea Government MSIT (grant no. 2018-0-00622 ).

    Publisher Copyright:
    © 2019 Elsevier Ltd

    Keywords

    • AI-based methods
    • Deep Learning in robotics and automation
    • Force control

    ASJC Scopus subject areas

    • Control and Systems Engineering
    • Software
    • General Mathematics
    • Computer Science Applications
    • Industrial and Manufacturing Engineering

    Fingerprint

    Dive into the research topics of 'Reinforcement learning based on movement primitives for contact tasks'. Together they form a unique fingerprint.

    Cite this