Intelligent Autopilot Design Based on Adaptive Neuro -Fuzzy Technique and Genetic Algorithm

In the last decade, the Unmanned Aerial Vehicles (UAVs) industry has a rapid progress in the development and optimization of UAV’s autopilot systems. This paper proposes two flight control methods using the Aerosonde simulation model, which was modeled and simulated using Simulink/MATLAB software. The two methods are a self-tuning PID controller using genetic algorithm and Adaptive Neuro-fuzzy Inference System controller (ANFIS). In a self-tuning PID controller, a PID is genetically controlled and utilized as an autopilot to optimize the controller parameters for the proposed UAV model. However, the second method is based on fuzzy logic controller tuned using neural network. For the main navigation system Three fuzzy logic modules are designed to monitor the altitude, speed and heading angle. Simulation results for the two methods reveal high robustness and durability of ANFIS controller response especially under windy conditions.