Go-Explore Complex 3D Game Environments for Automated Reachability Testing

TL;DR

This paper introduces an exploration algorithm, Go-Explore, tailored for automated testing of large 3D game environments, effectively uncovering reachability bugs and achieving comprehensive coverage efficiently.

Contribution

It adapts Go-Explore for 3D game testing, enabling exhaustive exploration without human input and outperforming existing methods in bug detection and coverage.

Findings

Outperforms reinforcement learning baselines in coverage.

Finds challenging bugs in complex environments.

Covers 1.5km x 1.5km map within 10 hours on a single machine.

Abstract

Modern AAA video games feature huge game levels and maps which are increasingly hard for level testers to cover exhaustively. As a result, games often ship with catastrophic bugs such as the player falling through the floor or being stuck in walls. We propose an approach specifically targeted at reachability bugs in simulated 3D environments based on the powerful exploration algorithm, Go-Explore, which saves unique checkpoints across the map and then identifies promising ones to explore from. We show that when coupled with simple heuristics derived from the game's navigation mesh, Go-Explore finds challenging bugs and comprehensively explores complex environments without the need for human demonstration or knowledge of the game dynamics. Go-Explore vastly outperforms more complicated baselines including reinforcement learning with intrinsic curiosity in both covering the navigation mesh and number of unique positions across the map discovered. Finally, due to our use of parallel agents, our algorithm can fully cover a vast 1.5km x 1.5km game world within 10 hours on a single machine making it extremely promising for continuous testing suites.