genetic_algorithm_TSP

A genetic algorithm in a header file for finding a good approximate solution to a travelling salesman style problem. The header file also include facilities for somewhat preparing the data ready to be passed into the algorithm.

0. Proposed works

• Convert to templated format

1. Folder structure

• GA_TSP.h:
• main.cpp:
• test_main-GA_TSP.cpp:

2. Dependencies

• GA_TSP.h: utilizes C++11 features and the standard template library (STL).
• tests/test_main-GA_TSP.cpp: utilizes the Boost Unit Test Framework.

3. Setup

3.1 Configure

Examples can be seen inside "example/main.cpp" that show the following configuration steps. Inside a relevant C++ file carry out the following:

• Include the header file

#include "GA_TSP.h"

• seed the pseudo random number generator using srand() and call rand() once.

srand(static_cast(time(0))); rand();

• create a vector to hold the points (vertice).

std::vector points;

• initialize the points with data from either a txt file.

initVetices(points, short dimensions, short numPoints, std::string fileName);

• OR initialize the points with user input.

initVetices(points, short dimensions, short numPoints);

• run the the genetic algorithm to obtain the best approximate solution Genome

Genome bestGenome = GA_TSP(points, short populationSize, float survivingPopulationPercent, short averageNumMutations, short numGenerations);