Comparison of conventional and fuzzy predictive control

The use of fuzzy goals and fuzzy constraints in predictive control allows for a more flexible aggregation of the control objectives than the usual weighted sum of squared errors. A multistage decision making algorithm is applied to compute the optimal control action. Compared to the standard quadratic objective function, with the fuzzy decision making approach, the designer has more freedom in specifying the desired process behavior. This paper presents an experimental comparison of different cost functions, using an example of container crane control. The conventional quadratic criterion is compared with a conjunctive aggregation of fuzzy goals. The results show that a better performance can be achieved by using fuzzy goals. On the other hand, the optimization problem associated with the multistage decision making procedure has higher computational demands.