Intelligent control for 2DOF helicopter via genetic algorithms

Abstract

Genetic algorithm (GA) prevails as a parallel and global search technique among current optimization approaches, such as recursive and local search techniques. This paper presents a real-time control system optimization for a 2DOF helicopter plant using the genetic algorithms. We combine two real-coded GAs in the real-time control system. One GA is used to estimate the parameters of a dynamic system and the other is used to adjust the controller parameters accordingly. As the population evolves, the best control scheme chosen by the GA is adapted to the plant in order to enhance the closed-loop system performance. Even if parameter variations or disturbances in the plant occur the population of controllers will continually adapt to its environment and maintain good performance via the online GA optimization. In the experiment PID and fuzzy logic controllers are implemented. Results of the experiment are compared and show that GA has great potential for optimization in the real-time control system.