Optimal linear quadratic tracking using genetic approach

In this paper a genetic approach is presented to the optimal linear quadratic tracking problem. In this problem the control law is designed such that a cost function is minimized. The design specifications, depends on choosing the weighting matrices Q and R. One must carry out a trial- and error process to choose the weighting matrices that can satisfy the design specifications. To overcome this difficulty we employ the genetic algorithm to find the proper weighting matrices. A computer simulation is performed to track a desired reference trajectory in a motor-generator set as a multivariable system. This method is compared with the one proposed by Bryson and Ho (1975).