The ISC is part of the Johns Hopkins Applied Physics Laboratory and will follow all current policies. Please visit the JHU/APL page for more information on the Lab's visitor guidance.

2018

DRL Based Intelligent Joint Manipulator and Viewing Camera Control for Reaching Tasks and Environments with Obstacles and Occluders


Abstract

This work studies joint camera and robotic manipulator control for reaching tasks in complex environments with obstacles and occluders. We obviate the conventional challenges involved in complex perception, planning, and control modules and careful calibration for sensing and actuation and seek a solution leveraging deep reinforcement learning (DRL). Our method using DRL and deep Q-learning learns a policy for robot actuation and perception control, mapping directly raw image pixels inputs into camera motion and manipulator joint control actions outputs. We show results comparing different training approaches, and demonstrating competency for increasingly complex situations and degrees of freedom. These preliminary experiments suggest the effectiveness and robustness of the proposed approach.

Citation

@inproceedingsStaley_2018 doi: 10.1109/ijcnn.2018.8489273 url: https://doi.org/10.1109/ijcnn.2018.8489273 year: 2018 month: jul publisher: IEEE author: Staley Edward W. and Katyal Kapil D. and Burlina Philippe title: DRL Based Intelligent Joint Manipulator and Viewing Camera Control for Reaching Tasks and Environments with Obstacles and Occluders booktitle: 2018 International Joint Conference on Neural Networks (IJCNN)

Citation

@inproceedingsStaley_2018 doi: 10.1109/ijcnn.2018.8489273 url: https://doi.org/10.1109/ijcnn.2018.8489273 year: 2018 month: jul publisher: IEEE author: Staley Edward W. and Katyal Kapil D. and Burlina Philippe title: DRL Based Intelligent Joint Manipulator and Viewing Camera Control for Reaching Tasks and Environments with Obstacles and Occluders booktitle: 2018 International Joint Conference on Neural Networks (IJCNN)