DragTraffic: Interactive and Controllable Traffic Scene Generation for Autonomous Driving

Evaluating and training autonomous driving systems require diverse and scalable corner cases. However, most existing scene generation methods lack controllability, accuracy, and versatility, resulting in unsatisfactory generation results. Inspired by DragGAN in image generation, we propose DragTraffic, a generalized, interactive, and controllable traffic scene generation framework based on conditional diffusion. DragTraffic enables non-experts to generate a variety of realistic driving scenarios for different types of traffic agents through an adaptive mixture expert architecture. We employ a regression model to provide a general initial solution and a refinement process based on the conditional diffusion model to ensure diversity. User-customized context is introduced through cross-attention to ensure high controllability. Experiments on a real-world driving dataset show that DragTraffic outperforms existing methods in terms of authenticity, diversity, and freedom. Demo videos and code are available at https://chantsss.github.io/Dragtraffic/.

I. Introduction

The safety of autonomous driving systems relies heavily on the richness of the dataset scenarios. However, due to various constraints such as safety issues, geographical environment, and weather changes, it is difficult for collected data to cover all situations. This poses challenges for training and evaluating planning and prediction modules, especially for extreme scenarios. To address this, simulators such as SUMO [1] and CARLA [2] have been used to manually set scenarios. While rule-based simulations offer interpretability and viable trajectories without extensive data, they have limitations in accuracy, generalization, and adaptability. Furthermore, they require substantial expert knowledge to establish the necessary rules.