Using Restart Heuristics to Improve Agent Performance in Angry Birds

TL;DR

This paper introduces a restart heuristic framework for Angry Birds agents, demonstrating that strategic restarting can significantly enhance performance in solving levels.

Contribution

It proposes a novel restart strategy framework for Angry Birds agents, addressing a key human tactic previously unimplemented in AI agents.

Findings

Restart strategies improve agent success rates.

Restarting helps in unsolvable or unproductive level states.

The framework effectively guides when to restart levels.

Abstract

Over the past few years the Angry Birds AI competition has been held in an attempt to develop intelligent agents that can successfully and efficiently solve levels for the video game Angry Birds. Many different agents and strategies have been developed to solve the complex and challenging physical reasoning problems associated with such a game. However none of these agents attempt one of the key strategies which humans employ to solve Angry Birds levels, which is restarting levels. Restarting is important in Angry Birds because sometimes the level is no longer solvable or some given shot made has little to no benefit towards the ultimate goal of the game. This paper proposes a framework and experimental evaluation for when to restart levels in Angry Birds. We demonstrate that restarting is a viable strategy to improve agent performance in many cases.