Fuzzy controller design of micro-unmanned helicopter relying on improved genetic optimization algorithm

In this paper, the dynamic model of unmanned helicopter is modeled by frequency domain identification method. An adaptive fuzzy Proportion Integration Differentiation (PID) controller is established, with its design carried out from the aspects of the input and output fuzzy subset, the membership function selection, the fuzzy rule generation, as well as the fuzzy reasoning. An improved genetic algorithm is employed to optimize the initial expert empirical fuzzy rules, which avoids the traditional method from falling into the local optimal solution in the process of optimization. Specifically, adaptive crossover and mutation probability are adopted to accelerate the convergence speed of the genetic algorithm. We then apply the proposed algorithm to optimize the membership function and fuzzy control rules of the fuzzy controller. Simulation results indicate that our adaptive fuzzy PID has better control effect and anti-interference ability than the traditional PID control law. Additionally, our adaptive fuzzy PID controller is verification by flight test thought AF25B unmanned helicopter platform.