改进麻雀搜索算法的无人车路径规划

Path planning of unmanned vehicle based on improved sparrow search algorithm

针对麻雀搜索算法后期容易早熟,易陷入局部最优后搜索能力下降的不足,提出了一种结合伯努利(Bernoulli)映射和禁忌搜索算法的改进麻雀搜索算法。改进麻雀搜索算法在初始化阶段,使用Bernoulli映射初始化种群位置,取代传统算法初始化阶段采用随机数的方式,优化了种群分布不均,搜索范围不足的问题,同时当超过一定迭代次数全局最优值仍没有更新时再次使用Bernoulli映射对种群进行扰动增强全局搜索能力,在算法进入后期寻优阶段时使用禁忌搜索算法,应用其禁忌准则和特赦准则的结构实现良好的全局搜索能力。然后将该算法应用到无人车全局路径规划问题中,并进行了实验验证其有效性。实验结果表明改进后的麻雀搜索算法拥有更好的全局搜索能力和更高的精度。

An improved sparrow search algorithm based on Bernoulli mapping and tabu search algorithm is proposed to solve the problem that sparrow search algorithm is easy to be premature and fall into local optimization. In the initialization phase of the improved sparrow search algorithm, Bernoulli map is used to initialize the population position, instead of the random number method used in the initialization phase of the traditional algorithm, which optimizes the problems of uneven populatio n distribution and insufficient search range. At the same time, when the global optimal value is not updated after a certain number of iterations, Bernoulli map is used again to disturb the population to enhance the global search ability. The tabu search algorithm is used when the algorithm enters the later optimization stage, and its structure of tabu criteria and Amnesty criteria is applied to achieve good global search ability. Then the algorithm is applied to the global path planning problem of unmanned vehicles, a nd experiments are carried out to verify its effectiveness. The experimental results show that the improved sparrow search algorithm has better global search ability and higher accuracy.