A neural network based visuosteering control algorithm for autonomous land vehicles
SCOPUS
- Title
- A neural network based visuosteering control algorithm for autonomous land vehicles
- Authors
- Choi, D.-H.; Oh, S.-Y.; KIM, KWANG IK
- Date Issued
- 2021-09
- Publisher
- Taylor and Francis
- Abstract
- A neural network based navigation algorithm has been developed for the Postech Road Vehicle I (PRV I) to drive on outdoor roads with intersections on campus. The neural net essentially watches a human to drive and remembers driving signals for different road conditions and later on, generalizes this knowledge to similar road conditions. Four neural net modules are used for outdoor driving. The first one detects intersections just by being shown typical intersections. If an intersection is detected, a higher level command selects the proper one among three network outputs, namely the left turn net, the straight-ahead net, and the right turn net. For verification, the whole algorithm has been tested on campus roads and also in simulation of various driving conditions. © 1994 by Taylor & Francis. All rights reserved.
- URI
- https://oasis.postech.ac.kr/handle/2014.oak/113248
- ISSN
- 0000-0000
- Article Type
- Article
- Citation
- World Congress on Neural Networks, vol. 2, page. II.23 - II.28, 2021-09
- Files in This Item:
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.