Self-optimizing control

Self-optimizing control problems arise in applications where the desired task is implicitly specified as an optimization of an objective function which could be a function of the unknown plant. The control objective therefore involves the explicit determination of both the optimal task and the control action to achieve that task. The difficulty with this problem lies in the conflict between the need to identify the optimal task and to control the plant to achieve it. The proposed solution combines a reference generator and an adaptive controller. The reference generator, which provides the task to be executed by the adaptive controller, time multiplexes a training task to enable the system parameters to be identified, and an estimated optimal task. Switching is done in a manner that the training task becomes infrequently selected as the estimate for the optimal task improves.