Biologically Inspired Architectures for Learning and Control – Part II - Case Studies

Authors
  1. Cheng, D.X.P.
Corporate Authors
Defence R&D Canada - Suffield, Ralston ALTA (CAN)
Abstract
This article reviews the state of the art in biologically inspired learning architectures for autonomous systems, and discusses the major issues in three core learning techniques: reinforcement learning, neural-network-based learning, and learning using evolutionary algorithms. To illustrate implementations of these robot learning approaches, the following cases are studied: Pattern classification with a recurrent neural network (RNN); Autonomous land vehicle in neural networks (ALVINN); Obstacle avoidance using reinforcement learning (RL) Some of the limitations of cognitive learning processes in real world situations and directions for future development are also discussed.

Il y a un résumé en français ici.

Report Number
DRDC-SUFFIELD-TM-2006-269 — Technical Memorandum
Date of publication
01 Dec 2006
Number of Pages
40
DSTKIM No
CA031354
CANDIS No
530200
Format(s):
Electronic Document(PDF)

Permanent link

Document 1 of 1

Date modified: