论文部分内容阅读
自适应蚁群算法是蚁群算法的衍生,并成功有效地解决了许多路径优化问题,但随着问题规模增大,自适应蚁群算法已不能有效快速地得到我们期盼的结果。对于大规模的路径优化问题,根据分布式的思想,引入关联强化学习和后悔值以最大限度降低每一步决策对全局解的坏影响,实现多蚁之间的协作来提高解的质量,并缩短寻优时间,并加入2-opt算法避免陷入局部最优。以TSP为例,对比了不同规模的TSP的仿真结果,并加以分析得出结论。
Adaptive ant colony algorithm is a derivative of ant colony algorithm and successfully solves many path optimization problems. However, as the scale of the problem increases, adaptive ant colony algorithm can not get the result we expect quickly and effectively. For large-scale path optimization problems, according to the distributed thought, the introduction of association reinforcement learning and regret values to minimize the impact of each step of the decision-making on the global solution to achieve collaboration between multiple ants to improve the quality of the solution and shorten Find the best time, and add 2-opt algorithm to avoid falling into local optimum. Taking TSP as an example, the simulation results of TSP with different scales are compared and analyzed.