Airborne Computer System Based Collision-Free Flight Path Finding Strategy Design for Drone Model.

An airborne computer system is avionics-based model that consists of an electronic system and a specifically designed equipment. In this paper, we propose a collision-free path finding flight strategy that enhances the output performance of a drone system for free-navigation between buildings. This strategy is based on a reactive method that requires metaheuristic algorithms to process environmental information during drone flying. In addition, a robust controller is designed to achieve drone flight and reach the target point. The goal of this paper is to discuss the determination of the shortest flight path with the minimum cost function evaluation and this flight path should avoid the collision by using four heuristic algorithms including: Chaotic Particle Swarm Optimization (CPSO) algorithm, Fire-Fly (FF) algorithm, Bees (B) algorithm and a hybrid optimization algorithm that combines the CPSO, FF and B algorithms. To fly the drone and to follow the generated flightpath, six adaptive PID controllers are designed in order to control the highly nonlinear model and under-actuated drone system using an on-line CPSO algorithm to learn and tune the eighteen control gain parameters. The purpose of this control design is to precisely and quickly obtain robust thrust forces control actions to control the attitude and altitude of the drone model. The numerical results of the proposed flightpath-finding algorithms and robust control strategy confirm that the hybrid (CPSOFFB) flightpath algorithm has the minimum number of iterations and evaluation functions as well as free navigation and the shortest flight path length generated. Moreover, the proposed six adaptive PID controller results show that the four thrust forces control actions are smooth and accurately generated making the drone take off and follow the desired flight path quickly with the minimum number of cost function and around ±10 cm minimum error tracking translation location. The maximum overshoot of the altitude did not exceed 10 cm in the transient state and the orientation error of the drone is approximately zero in the steady-state. To demonstrate the effectiveness of the proposed flight path finding and control strategy, a comparison is made with the results other types of algorithms. [ABSTRACT FROM AUTHOR]