Paper Image

Modelling a puzzle video game with constraint programming and AI planning

Published on:

2 October 2023

Primary Category:

Artificial Intelligence

Paper Authors:

Joan Espasa,

Ian P. Gent,

Ian Miguel,

Peter Nightingale,

András Z. Salamon,

Mateu Villaret

Bullets

Key Details

The video game Puzznic is modelled with constraint programming and AI planning tools

PDDL planning model outperforms initial constraint models on benchmark levels

Constraint models use Essence Prime and Essence languages, translated to SAT

Potential improvements to constraint models are identified, like symmetry breaking

Additional benchmark instances will allow more thorough comparison of approaches

AI generated summary

Modelling a puzzle video game with constraint programming and AI planning

This paper explores modelling the puzzle video game Puzznic using constraint programming and AI planning tools. Both approaches are applied to benchmark game levels without moving blocks. The planning approach using PDDL currently outperforms the constraint models, but the paper discusses ways to improve the constraint formulations.

Answers from this paper

Comments

No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up