GeneticAlgorithm For TSP (Java)

This is a project to solve TSP by GeneticAlgorithm. If you look the source code, please click here.

Demo

I use processing to visual this algorithm and test it.

If you processing ide, download floder named GeneticP5Visaul. Then open GeneticP5Visaul.pde and run it, you will visual view. But you don't processing ide, you can load processing library or you run src/GeneticTest.java, you will watch it in console.

Processing Demo Java Demo

Example

Method 1

If you use this method, you are better in a thread. Because it will return result untill MaxGeneration.

// get a GeneticAlgorithm instance
GeneticAlgorithm ga = GeneticAlgorithm.getInstance();
ga.setMaxGeneration(1000);
ga.setAutoNextGeneration(true);
// points is a array of all point, the function of getdist() is get adjacency matrix
best = ga.tsp(getDist(points));
System.out.print("best path:");
for (int i = 0; i < best.length; i++) {
    System.out.print(best[i] + " ");
}
System.out.println();

Method 2

If you use this method, you will control it by yourselt and using nextGeneration() function.

GeneticAlgorithm ga = new GeneticAlgorithm();
best = ga.tsp(getDist(points));

int n = 0;
while (n++ < 100) {
    best = ga.nextGeneration();

    System.out.println("best distance:" + ga.getBestDist() + " current generation:" + ga.getCurrentGeneration() + " mutation times:" + ga.getMutationTimes());
    System.out.print("best path:");
    for (int i = 0; i < best.length; i++) {
        System.out.print(best[i] + " ");
    }
    System.out.println();
}

About me

Welcome to pull requests.

If you have any new idea about this project, feel free to contact me. 😃