抄録
This paper is concerned with the stabilization control of systems with time delay. We have proposed a feedback gain adjusting method, using a cart and pendulum system as an example, in which the stability of the closed-loop system is assumed. When the time delay is longer, however, it becomes difficult to find a stabilizing feedback gain, due to various uncertainties in the controlled object. Thus, the online feedback gain adjustment is carried out in two steps. First，we search for a set of optimum solutions through offline simulation using genetic algorithms (GA). Then, using these solutions as the initial set of gains, we perform the online feedback gain tuning of the actual plant using another search method. The performance of the proposed method is examined through an experiment.
