基于改进粒子群算法的三维路径规划

Improved particle swarm optimization algorithm for 3D path planning in intelligent construction

摘要:

针对粒子群算法解决建造项目中的无人机三维路径规划问题时,易陷入局部最优问题,提出了一种混合惯性牵引力的粒子群优化算法。通过在初始阶段起始点与目标点位置关系,引入自适应初始化机制,对粒子群的初始种群进行优化；采用线性递减的惯性权重方式,加强算法前期的全局搜索与后期的局部搜索性能；借助万有引力思想在速度更新中引入加速度,加强搜索的性能。采用有无自适应初始化机制的改进算法进行对比试验,结果验证了该机制更有利于提高算法的求解质量；通过IPSO算法、IHPSO算法与改进算法进行仿真实验,结果表明改进算法所的求解质量上更好,稳定性相对于IPSO较好69.75%,相对于IHPSO较好17.41%。

Abstract:

Aiming at the problem of local optimization when particle swarm optimization algorithm is used to solve the three-dimensional path planning problem of UAV in construction project, a particle swarm optimization algorithm with hybrid inertial traction is proposed. Through the position relationship between the starting point and the target point in the initial stage, the adaptive initialization mechanism is introduced to optimize the initial population of particle swarm optimization; The linear decreasing inertia weight method is adopted to strengthen the global search performance in the early stage and the local search performance in the later stage; With the help of the idea of universal gravitation, the acceleration is introduced into the velocity update to enhance the search performance. Compared with the improved algorithm with or without adaptive initialization mechanism, the results show that the mechanism is more conducive to improve the solution quality of the algorithm; Through the simulation experiments of IPSO algorithm, IHPSO algorithm and improved algorithm, the results show that the improved algorithm has better solution quality, and the stability is 69.75% better than IPSO and 17.41% better than IHPSO.