WcDT: World-centric Diffusion Transformer for Traffic Scene Generation
This work addresses the need for high-quality trajectory generation in autonomous driving simulations, representing an incremental improvement by integrating existing methods like diffusion models and transformers.
The paper tackles the problem of generating realistic and diverse traffic trajectories for autonomous driving simulation by proposing the World-Centric Diffusion Transformer (WcDT), which combines diffusion models and transformers to optimize trajectory generation from feature extraction to inference, achieving superior performance in realism and diversity.
In this paper, we introduce a novel approach for autonomous driving trajectory generation by harnessing the complementary strengths of diffusion probabilistic models (a.k.a., diffusion models) and transformers. Our proposed framework, termed the "World-Centric Diffusion Transformer"(WcDT), optimizes the entire trajectory generation process, from feature extraction to model inference. To enhance the scene diversity and stochasticity, the historical trajectory data is first preprocessed into "Agent Move Statement" and encoded into latent space using Denoising Diffusion Probabilistic Models (DDPM) enhanced with Diffusion with Transformer (DiT) blocks. Then, the latent features, historical trajectories, HD map features, and historical traffic signal information are fused with various transformer-based encoders that are used to enhance the interaction of agents with other elements in the traffic scene. The encoded traffic scenes are then decoded by a trajectory decoder to generate multimodal future trajectories. Comprehensive experimental results show that the proposed approach exhibits superior performance in generating both realistic and diverse trajectories, showing its potential for integration into automatic driving simulation systems. Our code is available at \url{https://github.com/yangchen1997/WcDT}.