Advanced search
1 file | 2.70 MB Add to list

Active vision for robot manipulators using the free energy principle

Toon Van de Maele (UGent) , Tim Verbelen (UGent) , Ozan Catal (UGent) , Cedric De Boom (UGent) and Bart Dhoedt (UGent)
Author
Organization
Project
Abstract
Occlusions, restricted field of view and limited resolution all constrain a robot's ability to sense its environment from a single observation. In these cases, the robot first needs to actively query multiple observations and accumulate information before it can complete a task. In this paper, we cast this problem of active vision as active inference, which states that an intelligent agent maintains a generative model of its environment and acts in order to minimize its surprise, or expected free energy according to this model. We apply this to an object-reaching task for a 7-DOF robotic manipulator with an in-hand camera to scan the workspace. A novel generative model using deep neural networks is proposed that is able to fuse multiple views into an abstract representation and is trained from data by minimizing variational free energy. We validate our approach experimentally for a reaching task in simulation in which a robotic agent starts without any knowledge about its workspace. Each step, the next view pose is chosen by evaluating the expected free energy. We find that by minimizing the expected free energy, exploratory behavior emerges when the target object to reach is not in view, and the end effector is moved to the correct reach position once the target is located. Similar to an owl scavenging for prey, the robot naturally prefers higher ground for exploring, approaching its target once located.
Keywords
active vision, active inference, deep learning, generative modeling, robotics

Downloads

  • 7868.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 2.70 MB

Citation

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

MLA
Van de Maele, Toon, et al. “Active Vision for Robot Manipulators Using the Free Energy Principle.” FRONTIERS IN NEUROROBOTICS, vol. 15, 2021, doi:10.3389/fnbot.2021.642780.
APA
Van de Maele, T., Verbelen, T., Catal, O., De Boom, C., & Dhoedt, B. (2021). Active vision for robot manipulators using the free energy principle. FRONTIERS IN NEUROROBOTICS, 15. https://doi.org/10.3389/fnbot.2021.642780
Chicago author-date
Van de Maele, Toon, Tim Verbelen, Ozan Catal, Cedric De Boom, and Bart Dhoedt. 2021. “Active Vision for Robot Manipulators Using the Free Energy Principle.” FRONTIERS IN NEUROROBOTICS 15. https://doi.org/10.3389/fnbot.2021.642780.
Chicago author-date (all authors)
Van de Maele, Toon, Tim Verbelen, Ozan Catal, Cedric De Boom, and Bart Dhoedt. 2021. “Active Vision for Robot Manipulators Using the Free Energy Principle.” FRONTIERS IN NEUROROBOTICS 15. doi:10.3389/fnbot.2021.642780.
Vancouver
1.
Van de Maele T, Verbelen T, Catal O, De Boom C, Dhoedt B. Active vision for robot manipulators using the free energy principle. FRONTIERS IN NEUROROBOTICS. 2021;15.
IEEE
[1]
T. Van de Maele, T. Verbelen, O. Catal, C. De Boom, and B. Dhoedt, “Active vision for robot manipulators using the free energy principle,” FRONTIERS IN NEUROROBOTICS, vol. 15, 2021.
@article{8704545,
  abstract     = {{Occlusions, restricted field of view and limited resolution all constrain a robot's ability to sense its environment from a single observation. In these cases, the robot first needs to actively query multiple observations and accumulate information before it can complete a task. In this paper, we cast this problem of active vision as active inference, which states that an intelligent agent maintains a generative model of its environment and acts in order to minimize its surprise, or expected free energy according to this model. We apply this to an object-reaching task for a 7-DOF robotic manipulator with an in-hand camera to scan the workspace. A novel generative model using deep neural networks is proposed that is able to fuse multiple views into an abstract representation and is trained from data by minimizing variational free energy. We validate our approach experimentally for a reaching task in simulation in which a robotic agent starts without any knowledge about its workspace. Each step, the next view pose is chosen by evaluating the expected free energy. We find that by minimizing the expected free energy, exploratory behavior emerges when the target object to reach is not in view, and the end effector is moved to the correct reach position once the target is located. Similar to an owl scavenging for prey, the robot naturally prefers higher ground for exploring, approaching its target once located.}},
  articleno    = {{642780}},
  author       = {{Van de Maele, Toon and Verbelen, Tim and Catal, Ozan and De Boom, Cedric and Dhoedt, Bart}},
  issn         = {{1662-5218}},
  journal      = {{FRONTIERS IN NEUROROBOTICS}},
  keywords     = {{active vision,active inference,deep learning,generative modeling,robotics}},
  language     = {{eng}},
  pages        = {{18}},
  title        = {{Active vision for robot manipulators using the free energy principle}},
  url          = {{http://dx.doi.org/10.3389/fnbot.2021.642780}},
  volume       = {{15}},
  year         = {{2021}},
}

Altmetric
View in Altmetric
Web of Science
Times cited: