Say "Sul Sul!" to SimSim, A Sims-Inspired Platform for Sandbox Game AI

TL;DR

This paper introduces SimSim, a Sims-inspired platform for testing divergent search algorithms in environment design, demonstrating how novelty-based evolutionary algorithms can generate viable, diverse house designs satisfying agent needs.

Contribution

It presents a new sandbox environment for AI testing in simulation games and evaluates the effectiveness of novelty-based evolutionary algorithms in this context.

Findings

Novel environment design platform for sandbox games.

Effective use of novelty-based algorithms for diverse solutions.

Demonstrated viability of AI-generated house designs.

Abstract

This paper proposes environment design in the life simulation game The Sims as a novel platform and challenge for testing divergent search algorithms. In this domain, which includes a minimal viability criterion, the goal is to furnish a house with objects that satisfy the physical needs of a simulated agent. Importantly, the large number of objects available to the player (whether human or automated) affords a wide variety of solutions to the underlying design problem. Empirical studies in a novel open source simulator called SimSim investigate the ability of novelty-based evolutionary algorithms to effectively generate viable environment designs.