GraphSCENE: On-Demand Critical Scenario Generation for Autonomous Vehicles in Simulation

TL;DR

This paper presents GraphSCENE, a method for on-demand generation of diverse, critical traffic scenarios in simulation for autonomous vehicle testing, using a temporal graph neural network guided by real-world interaction patterns.

Contribution

It introduces a novel approach that generates dynamic scene graphs tailored to user preferences, improving scenario diversity and relevance for AV validation.

Findings

Outperforms baselines in link prediction accuracy

Generates realistic and diverse traffic scenarios

Effective for AV testing in simulation environments

Abstract

Testing and validating Autonomous Vehicle (AV) performance in safety-critical and diverse scenarios is crucial before real-world deployment. However, manually creating such scenarios in simulation remains a significant and time-consuming challenge. This work introduces a novel method that generates dynamic temporal scene graphs corresponding to diverse traffic scenarios, on-demand, tailored to user-defined preferences, such as AV actions, sets of dynamic agents, and criticality levels. A temporal Graph Neural Network (GNN) model learns to predict relationships between ego-vehicle, agents, and static structures, guided by real-world spatiotemporal interaction patterns and constrained by an ontology that restricts predictions to semantically valid links. Our model consistently outperforms the baselines in accurately generating links corresponding to the requested scenarios. We render the predicted scenarios in simulation to further demonstrate their effectiveness as testing environments for AV agents.