RangXuan / MAPF

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

MAPF

Implementation of some of the most famous search-based algorithms for solving the Multi-Agent Path Finding problem, with some variants of the problem and a graphic user interface for the visualization of the results.

Implemented Algorithms

  1. Hierarchical Cooperative A* (HCA*) [David Silver, 2005]
  2. A*-Based Search
  3. Standley's enhancement: Operator Decomposition (OD) [Trevor Standley, 2010]
  4. Standley's enhancement: Independence Detection (ID) [Trevor Standley, 2010]
  5. Increasing Cost Tree Search (ICTS) [Guni Sharon et al., 2013]
  6. Conflict-Based Search (CBS) [Guni Sharon et al., 2015]
  7. M* [Glenn Wagner and Howie Choset, 2011]

Implemented Variants of the problem

Agents' behaviour at goal:

  • Stay at goal;
  • Disappear at goal after a choosable number of time steps;

Conflicts:

  • Vertex conflicts: always checked;
  • Edge conflicts: up to the user;

Objective functions:

  • Sum of costs (SOC);
  • Makespan

About


Languages

Language:Python 100.0%