solver README

A simple puzzle solver coded in Java used to find the shortest path to a given solution, implementing BFS in tandem with backtracking.

Further documentation is detailed in the code for both the solver algorithm and demo configurations.

Usage guidelines

• The code, namely the core solver algorithm, belonging to this repo are entirely free to use!
• Crediting the source, however, is preferred.

What are configurations?

• This algorithm can't just be copied and then instantly deployed; it requires the implementation of a configuration that represents the problem you wish to solve.
• A configuration is an attempted solution at any stage of a given problem.
  • It is considered invalid if it breaks any of the constraints/rules imposed by the problem, thus resulting in an impossible solution.
  • Conversely, a configuration is the goal if it represents a valid solution to the problem.
• A successor configuration is the state that immediately follows the current step towards solving a problem.
  • This successor is a new configuration that includes all the choices made by its parent, plus one additional choice, usually included as an array of other possible solutions.

Example cases

• Provided a configuration, the solver algorithm recurses on each possible move made from that configuration, which represents some step in solving the puzzle.
• You can base this configuration (whose required methods are also included in the Configuration interface) on the two puzzles used as examples here: src/demo/watch and src/demo/crossing.
• Each puzzle takes in arguments from the command line and displays the shortest path to the solution in standard output.

---

This algorithm, as well as the output files for each puzzle, were programmed under the instruction and guidance of RIT's CS department as part of assigned coursework.