This project was done as a part of the AI introduction course at TEK, Southern University of Denmark.
The problem is two-fold; An algorithm should to be built to solve Sokoban maps, and a physical robot should be designed and programmed to execute the solution.
This repository contains both parts; An A* algorithm to find a solution to a given Sokoban map, and the software implementation for our robot design.
The two parts represent traditional, logic-based AI and embodied AI, respectively, and are glued together with an instruction string of basic moves like fr for go forward and turn right.
The programs are run as python modules from the project root directory.
The following command runs the 2018 competition map with the can-closest (d_closest) heuristic.
$ python3 -m solver.solve solver/maps/2018-compitition-map.txt d_closestThe solution can be shown graphically with this command:
$ python3 -m solver.util.exploreSolution solver/solutions/2018{...}.txtThe solution string for the robot can be printed by:
$ python3 -m solver.util.createPath solver/solutions/2018{...}.txt