A Study on Path Planning of Unmanned Aerial Vehicle Based on Improved Genetic Algorithm

Genetic algorithm is a kind of way to solve complex problems effectively, for it is not bound by the restrictive assumptions of the search space, and doesn't require the assumption conditions such as continuity and derivatives. So this algorithm has its advantage that the traditional algorithm can not compared. Genetic algorithm uses multi-point search. In each iteration, the new individuals are generated by mating and mutation, so the searching range can be expanded, and the local optimal solution can be effectively prevented. In this paper, a real number encoding method based the change of course is proposed. According to the change of course, the algorithm constructs the individuality, and constructs the temporary path by the individual coding vector. And on this basis, the related operators shall be designed through the new encoding method and a series of genetic operations to carry out the path planning. The simulation results show that this method can improve the global search ability of genetic algorithm, and also improves the quality of Unmanned Aerial Vehicle flight path. The Unmanned Aerial Vehicle could get a better path in terms of the performance cost.

I. Introduction

Path planning of Unmanned Aerial Vehicle (UAV) refers to that under the comprehensive consideration of the UAV maneuver performance, penetration probability, hit probabil-ity, flight time within bounds, it helps to look for a optimal or feasible path from the initial point to the target point flight. To solve the problem of genetic algorithms in path planning, many works have been carried out in the last decades. For example. Shima presents a unified approach to the management of multiple UAVs in military scenarios [1]. And many other improved genetic algorithms have been proposed to find the optimal route for UAVs effectively [2]–[4]. Genetic algorithm is a probability search technology. We convert problem into a chromosome data string by the some encoding techniques.