Conditional Flow-VAE for Safety-Critical Traffic Scenario Generation
This work addresses the need for scalable and realistic safety-critical scenario generation to improve autonomous vehicle training and testing.
The authors propose a conditional latent flow matching method to generate realistic safety-critical traffic scenarios for autonomous vehicles, demonstrating more consistent and diverse scenario generation compared to existing approaches.
Safety-critical scenarios are essential for the development of autonomous vehicles (AVs) but are rare in real-world driving data. While simulation offers a way to generate such scenarios, manually designed test cases lack scalability, and adversarial optimization often produces unrealistic behaviors. In this work, we introduce a conditional latent flow matching approach for scalable and realistic safety-critical scenario generation. Our method uses distribution matching to transform nominal scenes into safety-critical rollouts. Furthermore, we demonstrate that incorporating both simulation and real-world data enables our framework to efficiently generate diverse, data-driven scenarios. Experimental results highlight that our approach is able to more consistently and realistically generate novel safety-critical scenarios, making it a valuable tool for training and benchmarking AV systems.