anthrax3 / js-aco

A visual demo of Ant Colony Optimisation applied to TSP written in Javascript

JS ACO

JS ACO is a visual demo of Ant Colony Optimisation written in Javascript (ES6). It contains basic example of how ACO works given a randomly generated TSP. The basic parameters of the Ant System are available to be tuned.

The UI will show the current optimum tour, and also a heat map of the heuristic (distance) matrix plus a pheromone matrix. This information will update as the algorithm runs.

Demo

You can see it in action here.

Setup

• Clone the repository git clone https://github.com/GordyD/js-aco
• cd js-aco
• npm install
• bower install

To run in development

• Run npm run develop
• Go to http://localhost:8080
• To increase problem size set query param nodes to an Integer
• To use a different seed set query param seed to a String or Number

Further Work

I intend to:

• add other ACO algorithms to this library
• add further support for more formats of TSP problems found in TSPLib
• provide greater interactivity with visualisations