An intelligent system architecture for urban traffic control applications

This paper describes an intelligent system architecture for urban traffic control which integrates a neural network and an expert system on silicon. The intelligent decision making system consists of a backpropagation based neural network for adaptive learning and a rule-based fuzzy expert system for decision making. Both the neural network and the expert system are implemented as linear systolic arrays. Thus, the entire system can be realized in VLSI with a few basic cells. The architecture exploits the principles of pipelining and parallelism to the maximum possible extent in order to achieve high speed and throughput. The effectiveness of the proposed system is illustrated by mapping two application problems: (i) adaptive traffic light control and (ii) congestion detection and avoidance in urban traffic control. The proposed hardware can yield a real-time decision every 5 ns based on a 200 MHz clock.