论文部分内容阅读
蚁群优化算法(Ant Colony Optimization,ACO)通过模拟蚁群觅食的行为提出的一种启发式防生算法,被应用于很多组合优化问题中。针对蚁群算法在求解过程中容易陷入局部最优出现早熟停滞,对基本的蚁群算法进行了改进,提出了基于混沌扰动的蚁群算法(CD-ACO),在算法的概率转移中引入混沌扰动因子,以保持蚁群群的多样性,使算法易于跳出局部极值区间,加快收敛速度,从而提高全局搜索能力。在蚁群选取下一跳节点的时候,通过欧式距离对邻节点的选取进行了筛选,使蚁群朝着目标节点的方向进行移动,提高搜索效率;最后在MATLAB中建立道路交通网络模型,并进行了路径导航仿真实验,实验表明,CD-ACO的稳定性强,收敛速度快,并且能够得到全局较优的路径。
Ant Colony Optimization (ACO) A heuristic anti-spion algorithm proposed by simulating foraging behavior of ant colony is applied to many combinatorial optimization problems. Aiming at the precarious stagnation of ant colony algorithm which is easy to fall into the local optimum in the process of solving, the basic ant colony algorithm is improved. The ant colony algorithm (CD-ACO) based on chaotic perturbation is proposed. Chaos is introduced in the probability transfer of the algorithm The perturbation factor, in order to keep the diversity of ant colony, makes the algorithm easy to jump out of the local extremal interval, speed up the convergence, and improve the global search ability. When the ant colony chooses the next-hop node, the selection of neighbor nodes by the European distance is screened so that the ant colony moves towards the target node to improve the search efficiency. Finally, a road traffic network model is established in MATLAB The experiments of path navigation simulation show that the CD-ACO has strong stability and fast convergence, and can obtain a globally optimal path.