Neural network based vision for precise control of a walking robot

Dean Pomerleau
Machine Learning, Vol. 15, 1994, pp. 125-135.


Download
  • Adobe portable document format (pdf) (178KB)
Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.

Abstract
This article describes a connectionist vision system for the precise control of a robot designed to walk on the exterior of the space station. The network learns to use video camera input to determine the displacement of the robot's gripper relative to a hole in which the gripper must be inserted. Once trained, the network's output is used to control the robot, with a resulting factor of five fewer missed gripper insertions than occur when the robot walks without sensor feedback. The neural network visual feedback techniques described could also be applied in domains such as manufacturing, where precise robot positioning is required in an uncertain environment.

Notes
Associated Center(s) / Consortia: Vision and Autonomous Systems Center

Text Reference
Dean Pomerleau, "Neural network based vision for precise control of a walking robot," Machine Learning, Vol. 15, 1994, pp. 125-135.

BibTeX Reference
@article{Pomerleau_1994_1730,
   author = "Dean Pomerleau",
   title = "Neural network based vision for precise control of a walking robot",
   journal = "Machine Learning",
   pages = "125-135",
   year = "1994",
   volume = "15",
}