基于自适应遗传算法的连续时空最优搜索路径规划研究

针对连续时空最优搜索者路径问题，利用随机微分方程描述Markov运动目标，建立了同时优化搜索者方向和速度的规划模型，并考虑了搜索速度对探测能力的影响。设计了一种新颖的自适应变异遗传算法，算法采用较高的变异概率作用于父代精英个体组，通过引入3种控制因子对变异方向和幅度进行自适应控制，动态调节局部搜索和全局搜索的平衡。在对方向未知的逃离目标搜索算例中，得到了近似对数螺旋曲线的搜索路径；在直升机搜索多目标的路径规划中，提供了合理有效的搜索方案。算法对比表明所给出的算法在全局优化能力和稳定性上有明显的优势，适用于求解连续搜索路径规划问题。

Research on Optimal Search Path Programming in Continuous Time and Space Based on an Adaptive Genetic Algorithm

A Markovian-target model based on stochastic differential equations and a path programming model with both searcher’s direction and velocity treated as decision variables are presented for optimal searcher path problem in continuous time and space, and the effect of searcher’s velocity on the detection ability is considered. A genetic algorithm with adaptive mutation is designed by introducing three kinds of control factors, which fulfills the adaptive control of the direction and range of mutation and dynamically regulates the balance between local search and global search. In an example of searching a target with a random escaping direction, an approximate logarithmic spiral path is found. Moreover, the algorithm provides a reasonable and effective search scheme in a path programming problem for a helicopter searching multiple targets. The results indicate that the proposed algorithm has the significant advantages of stability and global optimizing ability in comparison with other methods, and is well suitable for the search path programming problem in continuous time and space.