kumarganduri / ai-mapp

MAPP for Warehouse Butlers using Decoupled Coordination Technique and A-star variant, with Simulator

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

ai-mapp

This project is on Multi-Agent Path Planning(MAPP) for Warehouse Butlers using Decoupled Coordination Technique and a variant of A-star algorithm. It has a simulator that makes analysis of algorithm very easy. The project has been published for the students or anyone who pursue Artificial Intelligence, and is licensed under Apache License 2.0.

This project has been made using NetBeans and JDK8.

This is just the basic version (see demo) which assigns the bots with targets in a grid with numerous obstacles, finds their paths (independent of each other; decoupling) using variant of A-star alogorithm which we call as A-Plus algorithm (only horizontal and vertical motions are allowed), finds path-intersections and manages their motion dynamically (decoupling).

It has a lot of scope for improvement like: bots-assignment, velocity management, increasing problem complexity etc. Simulator makes it easy to have a better vision of such improvements - this is the second important goal of this project. If anyone uses it, they can let me know and I'll help improve the simulator and also make the project more manageable.

It is advised that you try using simulator (you'll need Java 8 to run the JAR file) along with reading the source code and find out what improvements are possible. Then add/improve algorithms. You can post your algorithm or improvements and let someone, including me, write the code for you. Then you can test your algorithm using the simulator.

About

MAPP for Warehouse Butlers using Decoupled Coordination Technique and A-star variant, with Simulator


Languages

Language:Java 100.0%