基于遗传退火算法的无人机航路规划

Simulation of Genetic Annealing Algorithm for Route Planning of Unmanned Aerial Vehicle

文章研究无人机航路规划问题;无人机航路规划问题约束条件较多,且对计算实时性要求较高,传统的优化方法不能很好满足实时性的要求,遗传算法计算速度较快,但局部搜索能力不强,在求解具有复杂约束条件的航路规划问题时容易陷入局部最优;为此提出一种求解航路规划问题的改进遗传算法,算法将遗传算法和模拟退火算法相结合,利用模拟退火算法增强了算法的局部搜索能力,改善了遗传算法易早熟的缺点;最后利用改进算法对无人机航路规划进行仿真,仿真结果表明该算法能避免陷入局部最优,具有较快收敛速度,航路规划质量较高。

This paper studied on route planning of Unmanned Aerial Vehicle （UAV）. UAV Route Planning included many constraints, and required a higher real time computing, traditional optimization methods can＇t meet the realmtime requirements, the genetic algorithm （GA） was faster, but the local search ability wasn＇t strong, so that GA was easy to fall into local optimal when solved to route planning with complex constraints. So this paper proposed an improved genetic algorithm for route planning problem. By combining GA and simulated an nealing algorithm （SA）, the algorithm used SA to enhance the ability of local search algorithm to improve GA precocious shortcomings. Fi- nally, the improved algorithm for route planning is simulated, and the results show that this algorithm can avoid being trapped in local opti- mum, convergence speed and route quality are improved.