Intelligent Coordination of Multiple Systems with Neural Networks

Many control applications require cooperation of two or more independently designed, separately located, but mutually affecting, subsystems. In addition to the good behavior of each subsystem, effective coordination of these subsystems is very important to achieve the desired overall system peformance. However, such coordination is very difficult to accomplish due mainly to the lack of precise system models and/or dynamic parameters as well as the lack of efficient tools for system analysis, design, and real-time computation of optimal solution. In this paper, we propose a new multiple-system coordinator which combines the techniques of intelligent control and neural networks, and forms the high-level coordinator in a hierarchical structure. The basic idea is to estimate the effects of the control commands to subsystems using a predictor and modify these commands using a knowledge-based coordinator so as to achieve tie desired performance. The predictor is designed for multiple-input, multiple-output systems using neural networks. The knowledge-based coordinator is responsible for knowledge acquisition and representation, inference process, and the overall system stability. Because the internal structure and parameters of the low level are not affected by using the proposed method, some commercially-designed servo controller for single systems can be coordinated to perform more sophisticated tasks for multiple systems than originally intended.