Scenario Generation in Roundabouts with Adjustable Interaction Intensity

2026-05-18

AI Summary

Existing methods struggle to efficiently generate roundabout near-critical scenarios with controllable interaction intensity from naturalistic driving data, hindering systematic safety testing of autonomous driving systems. This work proposes a novel roundabout scenario generation approach that enables continuous and controllable modulation of interaction intensity by decoupling geometric trajectories from temporal dynamics and mapping them into a latent space. Leveraging a conditional Wasserstein Generative Adversarial Network (WGAN) combined with a scalable yielding code, the method supports fine-grained control over interaction intensity for the first time in roundabout settings. A yielding-timing intervention mechanism further calibrates scenario criticality. Experiments demonstrate that the generated scenarios significantly outperform baseline methods in terms of spatio-temporal–latent consistency and interaction plausibility. Moreover, adjusting the scaling factor λ flexibly modulates safety margins, enabling a controllable and scalable safety validation framework.

Abstract

Roundabouts, characterized by frequent merging and yielding interactions, remain a safety-critical corner case for the development and testing of intelligent driving functions. However, extracting sufficient near-critical scenarios from naturalistic data is inefficient. Most existing scenario generation methods provide limited controllability over interaction intensity and criticality, making systematic safety testing and detailed analysis difficult. This paper presents an interaction-aware roundabout scenario generator with continuously adjustable interaction intensity. Geometric routes and temporal progress profiles are first decoupled and mapped to latent codes using pretrained autoencoders. Conditional latent generation is then performed with Wasserstein Generative Adversarial Networks (WGAN) to generate scenarios. Yielding is modeled as a controllable timing intervention via a compact yield code during the approach-to-entry segment, where interaction intensity is modulated by scaling the code with a factor $λ$. Results demonstrate enhanced timing-latent fidelity and plausible interaction responses compared to a baseline model. Under criticality-calibrated scaling, increasing $λ$ expands the safety margin, providing a scalable and controlled testing mechanism.

Problem

• roundabouts
• scenario generation
• interaction intensity
• safety testing
• critical scenarios

Innovation

• interaction-aware scenario generation
• adjustable interaction intensity
• yield code
• Wasserstein GAN
• latent space modeling