CONSTRUCTING CONTROL RULES FOR A DYNAMIC SYSTEM - PROBABILISTIC QUALITATIVE MODELS, LOOKAHEAD AND EXAGGERATION

Dynamic systems have sometimes to be controlled in circumstances that are very inconvenient for classical control methods. In such cases, us e of qualitative controllers seems to be a promising alternative, as t hey can be designed without having an exact mathematical model. A meth od for synthesizing control rules is presented, in which a probabilist ic qualitative model is used together with a decision strategy that ta kes into account the results of qualitative simulation. As a case stud y, the problem of the inverted pendulum control is considered. By comb ining the model simulation lookahead and a decision criterion, an inte resting phenomenon of 'exaggerated simulation' has emerged. Experiment al results show that in certain cases, using an exaggerated force in t he process of simulation considerably improves the performance of gene rated control rules. The exaggeration helps to detect some trends of t he system's behaviour that would otherwise have remained hidden. Exper iments and mathematical analysis show that the improvement due to the exaggeration is comparable to that achieved by a deeper lookahead in t he state transition graph.