UAV path planning using segmented cellular evolutionary algorithm

Abstract

Path planning is one of the most important issues for sustainable development of unmanned aerial vehicles (UAVs). The aim of the path planning is to provide the shortest and safest route. In this study, an algorithm has been proposed that achieves the solution diversity of genetic algorithms quickly. A novel cellular evolutionary algorithm with fixed initial population and segmented chromosome structure (SCEA) has obtained a high convergence speed to the best solution specifically about the path planning problems for UAVs. In comparison of proposed algorithm and derivatives with native evolutionary algorithm, the SCEA, has achieved almost all best results in each title. After 50 iteration of 2D simulation, the SCEA is two times faster than traditional native evolutionary algorithm, the speed difference reaches 11 times better about the first valid solution. 3D simulation results are not much different. Although, the success of the algorithm is reduced because of the reasons described in the article, the SCEA finishes simulation 1.56 times faster and get first valid results 2.35 times faster.