Skip to content

Intelligent Agent that plays the Wumpus World using Prolog

License

Notifications You must be signed in to change notification settings

sherelynyap/wumpus-world-prolog

 
 

Repository files navigation

Wumpus World using Prolog

The Wumpus World (Originally Hunt the Wumpus) is a game created by Gregory Yob (Origins) and which is presented simplified in Artificial Intelligence a Modern Approach (AIMA) as an exercise for developing an Intelligent Agent that uses logic. The code presented here is one of the many possible approaches to the problem using Prolog.

Getting Started

The code uses logic programming, situational calculus, planning and heuristics to create an intelligent agent that plays the game. The original code is exaggeratedly commented (for beginners), so a less commented version was provided for those that already understand Prolog better.

Preparing the ground

To run this code you must first install a Prolog Environment such as SWI-Prolog. Once installed, double click the main.pl file to open the environment on the folder. As the main.pl file consults the worldBuilder.pl file, they must always be in the same folder, and the environment should be opened on that same folder.

Running

Once the enviroment is opened, type the following query on the Prolog environment:

run.
or
runManyMaps(1,1000).

"run." will always play the AIMA map, while runManyMaps(X,Y) will run all maps subsequently from X to Y. 2000 maps where provided in the worldBuilder.pl, which were randomly generated following AIMA rules.

Built With

Author

  • Alexander Csóka Roque - developed in April 2018

License

This project is licensed under the MIT License - see the LICENSE.md file for details

References

  • Pilar Pozos, Edgardo Yescas and Jacob Vásquez in "Planning using situation calculus, prolog and a mobile robot".
  • Markus Triska (Code).
  • Hilios (Repo).

About

Intelligent Agent that plays the Wumpus World using Prolog

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Prolog 100.0%