Concentric Tube Robots (CTRs) are promising for minimally invasive interventions due to their miniature diameter, high dexterity, and compliance with soft tissue. CTRs comprise individual pre-curved tubes usually composed of NiTi and are arranged concentrically. As each tube is relatively rotated and translated, the backbone elongates, twists, and bends with a dexterity that is advantageous for confined spaces. Tube interactions, unmodelled phenomena, and inaccurate tube parameter estimation make physical modeling of CTRs challenging, complicating in turn kinematics and control. Deep reinforcement learning (RL) has been investigated as a solution. However, hardware validation has remained a challenge due to differences between the simulation and hardware domains. With simulation-only data, in this work, domain randomization is proposed as a strategy for translation to hardware of a simulation policy with no additionally acquired physical training data. The differences in simulation and hardware forward kinematics accuracy and precision are characterized by errors of 14.74±8.87 mm or 26.61±17.00 % robot length. We showcase that the proposed domain randomization approach reduces errors by 56 % in mean errors as compared to no domain randomization. Furthermore, we demonstrate path following capability in hardware with a line path with resulting errors of 4.37±2.39 mm or 5.61±3.11 % robot length.


    Access

    Download


    Export, share and cite



    Title :

    Sim2Real Transfer of Reinforcement Learning for Concentric Tube Robots


    Contributors:

    Publication date :

    2023-08-09


    Remarks:

    IEEE Robotics and Automation Letters pp. 1-8. (2023) (In press).


    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English


    Classification :

    DDC:    629



    A Deep Reinforcement Learning Approach for Inverse Kinematics of Concentric Tube Robots

    Iyengar, K / Dwyer, G / Stoyanov, D | BASE | 2019

    Free access

    Deep Reinforcement Learning for Concentric Tube Robot Path Following

    Iyengar, Keshav / Spurgeon, Sarah / Stoyanov, Danail | BASE | 2023

    Free access

    Vision-based DRL Autonomous Driving Agent with Sim2Real Transfer

    Li, Dianzhao / Okhrin, Ostap | ArXiv | 2023

    Free access

    Unified tracking and shape estimation for concentric tube robots

    Vandini, A. / Bergeles, C. / Glocker, B. et al. | BASE | 2017

    Free access

    Investigating exploration for deep reinforcement learning of concentric tube robot control

    Iyengar, K / Dwyer, G / Stoyanov, D | BASE | 2020

    Free access