WcDT: World-centric Diffusion Transformer for Traffic Scene Generation

TL;DR

This paper presents WcDT, a novel traffic scene generation framework combining diffusion models and transformers to produce diverse, realistic autonomous driving trajectories for simulation systems.

Contribution

The paper introduces a new world-centric diffusion transformer framework that integrates diffusion models with transformer-based encoding for improved traffic scene trajectory generation.

Findings

Superior performance in generating realistic trajectories

Enhanced scene diversity and stochasticity

Effective integration of diffusion models with transformers

Abstract

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}.