Method of using reuse strategy-based intelligent swarm algorithm to optimize dynamic traveling salesman problem (TSP)

Abstract

The invention discloses a method of using a reuse strategy-based intelligent swarm algorithm to optimize a dynamic traveling salesman problem (TSP). A traditional traveling salesman problem needs finding of a Hamiltonian circuit with a least cost in a static searching space, but actually, certain application which is in a real world and can use the traveling salesman problem as models is not all static, and city sets and weight matrices in the problem models thereof dynamically change. In dynamic environments, a searching result in a last environment can be reused by a swarm under a new environment and is thus learned, thus a searching space of the problem can be reduced, and thus the algorithm is enabled to obtain a better path by searching in shorter time. The invention provides a dynamic travel-salesman modeling method with more realistic meaning and a strategy of reusing historical searching results. In experiment, dynamic performance of the optimization method is tested through setting different-degree changes of environments, and it is proved that the optimization method is reasonable and effective under all the different dynamic environments.

Description

technical field [0001] The invention relates to the technical field of dynamic traveling salesman modeling and intelligent computing, in particular to a method for optimizing a dynamic traveling salesman problem by using an intelligent group algorithm based on a reuse strategy. Background technique [0002] The Traveling Salesman Problem (TSP) has been considered as a classic combinatorial optimization problem since it was proposed. The salesperson needs to find a route from a certain city node in a fixed undirected graph G={V,A,W} Start from the point and traverse all other city nodes once, and finally return to the shortest Hamiltonian circuit at the starting point, where V represents a set of given nodes (in different problem models, the representative objects of the nodes are different), and A represents the graph The set of all undirected edges in , W is the weight matrix about the undirected graph G. The value of each element in the matrix represents the distance from...

AI Technical Summary

Problems solved by technology

Most of these problems are NP-hard combinatorial optimization problems, some traditional algorithms are difficult or impossible to solve, and the ant colony algorithm provides an effective means for solving these problems

Images