Advanced search
1 file | 2.00 MB
Author
Organization
Abstract
Deep reinforcement learning is becoming increasingly popular for robot control algorithms, with the aim for a robot to self-learn useful feature representations from unstructured sensory input leading to the optimal actuation policy. In addition to sensors mounted on the robot, sensors might also be deployed in the environment, although these might need to be accessed via an unreliable wireless connection. In this paper, we demonstrate deep neural network architectures that are able to fuse information generated by multiple sensors and are robust to sensor failures at runtime. We evaluate our method on a search and pick task for a robot both in simulation and the real world.

Downloads

    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 2.00 MB

Citation

Please use this url to cite or link to this publication:

Chicago
Bohez, Steven, Tim Verbelen, Elias De Coninck, Bert Vankeirsbilck, Pieter Simoens, and Bart Dhoedt. 2017. “Sensor Fusion for Robot Control Through Deep Reinforcement Learning.” In 2017 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS), 2365–2370. New york: Ieee.
APA
Bohez, S., Verbelen, T., De Coninck, E., Vankeirsbilck, B., Simoens, P., & Dhoedt, B. (2017). Sensor fusion for robot control through deep reinforcement learning. 2017 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS) (pp. 2365–2370). Presented at the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), New york: Ieee.
Vancouver
1.
Bohez S, Verbelen T, De Coninck E, Vankeirsbilck B, Simoens P, Dhoedt B. Sensor fusion for robot control through deep reinforcement learning. 2017 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS). New york: Ieee; 2017. p. 2365–70.
MLA
Bohez, Steven, Tim Verbelen, Elias De Coninck, et al. “Sensor Fusion for Robot Control Through Deep Reinforcement Learning.” 2017 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS). New york: Ieee, 2017. 2365–2370. Print.
@inproceedings{8558393,
  abstract     = {Deep reinforcement learning is becoming increasingly popular for robot control algorithms, with the aim for a robot to self-learn useful feature representations from unstructured sensory input leading to the optimal actuation policy. In addition to sensors mounted on the robot, sensors might also be deployed in the environment, although these might need to be accessed via an unreliable wireless connection. In this paper, we demonstrate deep neural network architectures that are able to fuse information generated by multiple sensors and are robust to sensor failures at runtime. We evaluate our method on a search and pick task for a robot both in simulation and the real world.},
  author       = {Bohez, Steven and Verbelen, Tim and De Coninck, Elias and Vankeirsbilck, Bert and Simoens, Pieter and Dhoedt, Bart},
  booktitle    = {2017 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS)},
  isbn         = {978-1-5386-2682-5},
  issn         = {2153-0858},
  language     = {eng},
  location     = {Vancouver, CANADA},
  pages        = {2365--2370},
  publisher    = {Ieee},
  title        = {Sensor fusion for robot control through deep reinforcement learning},
  year         = {2017},
}

Web of Science
Times cited: