NEURAL NETWORK CONTROL FOR AN AUTONOMOUS GUIDED VEHICLE

In this thesis, an Autonomous Automated Guided Vehicle (AAGV) controller is developed using a hierarchical structure of neural networks. A highly forward connected, back-propagated, recurrent neural network is developed and implemented in this controller. The use of recurrent networks is shown to provide smooth, non-cyclic guidance of the AGV. The target system addressed in this work is an experimental AGV equipped with a sonar ranging system, and differential-drive propulsion.

• Supplemental Notes:
  • Publication Date: 1992 Published By: University Microfilms International, Ann Arbor, MI Remarks: Thesis (M. A.)--University of Waterloo, 1992
• Corporate Authors:

  University of Waterloo

  Department of Civil and Environmental Engineering, 200 University Avenue West
  Waterloo, Ontario  Canada  N2L 3G1
• Authors:
  • Croft, Elizabeth Anne
• Publication Date: 1992

Language

• English

Media Info

• Pagination: xv, 141 p.

Subject/Index Terms

• TRT Terms: Automated guided vehicle systems; Neural networks; Robotics
• Subject Areas: Data and Information Technology;

Filing Info

• Accession Number: 00785214
• Record Type: Publication
• Source Agency: UC Berkeley Transportation Library
• Files: PATH
• Created Date: Nov 17 2000 12:00AM