Angry-HEX

Project Description

Angry-HEX is an artificial player for the popular video game Angry Birds; it participated in the 2013, 2014 and 2015 Angry Birds AI Competition.

The agent is based on declarative knowledge bases, and features a combination of traditional imperative programming and declarative programming that allows to achieve high flexibility in strategy design and knowledge modelling. In particular, it relies on the use of Answer Set Programming (ASP) and HEX programs.

Logic programming is used for decision tasks such as which target to hit at each shot (tactic gameplay) or which level to play at each turn, and which way (strategic gameplay).

A lot of work was carried out during these years, as shown by the Competition Results (2013, 2014, 2015) and Benchmarks (2013, 2014). However, there are still a lot of room for improvements to can be designed and implemented.

Goal

The aim of this thesis work is to improve the Angry-HEX agent using one of the most used physical engines for 2D games (Box2D).

The main focus will be on the development of specific functionality related to the physic simulation and the modelling of the current state of the game environment.

Requirements

The candidate must know the C++ programming language and the basics of Object Oriented Programming. She should also have a basic background of Answer Set Programming.