An Artificial Neural System for Autonomous Undersea Vehicles
Abstract
This paper introduces a new approach to the problem of information processing in autonomous undersea vehicles (AUVs). The approach involves the emulation of biological mechanisms of intelligence, principally through the evolutionary design of neural networks and systems of interacting networks, for implementation in software, parallel hardware, and analog/digital very-large- scale integration (VLSI). The objectives of artificial neural systems (ANS), in common with other AI approaches, are to provide real-time, operator-independent, pattern recognition, task planning, and adaptive AUV control for many of the underwater tasks currently being performed by man. The ANS approach emphasizes the development of autonomous intelligence capabilities through the progressive expansion of processing centers that define, orchestrate, and control basic reflexes. Keywords: Artificial neural systems; Autonomous undersea vehicles; Neural networks.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1988
- Accession Number
- ADA199725
Entities
People
- H. G. Nguyen
- M. R. Blackburn