GA optimal path finder

A simple genetic algorithm implementation involving a series of selection, crossover, mutation methods that minimises a fitness function to find an optimal path given any bitmap image of obstacles, from a starting point to finish.

Example obstacle image and solution

Breakdown of methods

Selection methods:

• Roulette Wheel
• Tournament
• Rank-based

Crossover methods:

• k-point
• Uniform

Mutation methods:

• Bit-flip
• Bit-swap

Running the code

You need to have at least MATLAB 2020a or later versions + few more adds-on to be install inside MATLAB as required when the program is run.

1. Clone the repo
2. Run the main.m file from the directory in MATLAB in the console
3. Install any adds-on needed, as outputed
4. The result png is overwriten to the same bmp image, and is displayed automatically once the algorithm converges