Dual-Population Ant Colony Optimization for Solving the 75-City Chinese TSP Problem

This program implements a dual-population ant colony optimization algorithm to solve the shortest path problem for 75 Chinese cities, a classic Traveling Salesman Problem (TSP) instance. The solution package contains MATLAB code that can be executed by extracting the files and running the main.m script, which initializes the algorithm configuration and starts the optimization process.

The dual-population ant colony algorithm mimics ant foraging behavior through pheromone trail deposition and evaporation mechanisms. The implementation features two ant populations that explore solution spaces simultaneously, with inter-population information exchange enhancing global search capability. Key algorithmic components include: pheromone matrix updates, probabilistic path selection using roulette wheel selection, and elitist strategy implementation for solution preservation.

The main.m file serves as the central controller where users can modify parameters such as ant population size, evaporation rate, and iteration count. The algorithm executes multiple optimization cycles, with each iteration refining the path solution through collaborative exploration between the dual populations. This design enables efficient convergence toward near-optimal solutions for large-scale TSP instances.