HUMAN-IMITATING INTELLIGENT CONTROL AND AN IMPROVED GENETIC ALGORITHM FOR PARAMETER-OPTIMIZATION OF HUMAN-IMITATING CONTROL LAW

摘要

In this paper we present a method of human-imitating control, which doesn't need mathematical model of the control system, and the parameter-optimization of the human-imitating control law based on an improved genetic algorithm that includes human knowledge and experience of controlling the object and uses only the nonlinear mathematical model of the object controlled. The problem of stabilizing a 2-D single-invert pendulum around its upright position with two rotational arms is researched in this paper.