基于混合变异策略差分进化算法的无人机航路规划

Path Planning for UAV Based on the Hybrid Strategy Differential Evolution Algorithm

提出在一种三维场景参数部分已知的有威胁空域中,采用差分进化算法(DE)规划无人机(UAV)航路,为了提高差分进化算法性能,以生成最优的路径,而采取多种变异策略的差分进化算法。算法根据种群个体的适应度值,将初始化的种群分为3个子种群,然后对于不同子种群采取不同变异策略。重点构建了环境模型和生存模型。数值实验结果表明,相较于单一变异策略的DE算法,混合变异策略的DE算法在无人机航路规划问题求解中,具有更好的求解效果和环境适应能力。

This paper uses a three-dimensional differential evolution algorithm for unmanned aerial vehicle（UAV） flying in partially known hostile environment. In order to improve the performance of differential evolution algorithm to generate a smooth and optimal fight route,this paper proposes an effective differential evolution algorithm with hybrid strategy. This algorithm divides an initial population into three sub-populations,and each sub-population select a mutation strategy according to their fitness Values. The model of environment and survival is also built in this paper.According to the results obtained,the proposed improved algorithm shows a superior performance in comparison with pure mutation strategy in terms of the solution quality,and adaptive capacity to environment.