Uncertainty-Aware Diffusion Model for Multimodal Highway Trajectory Prediction via DDIM Sampling

2026-02-24

AI Summary

This work addresses the challenge of high-precision, uncertainty-aware trajectory prediction in autonomous driving, where complex multi-agent interactions, scene dynamics, and motion stochasticity pose significant difficulties. To this end, the authors propose the cVMDx framework, which leverages a diffusion generative model enhanced with DDIM sampling to achieve over two orders of magnitude faster inference. The framework incorporates a Gaussian mixture model to fit generated trajectories, yielding interpretable multimodal outputs, and explores a CVQ-VAE variant for efficient encoding of scene context. Experiments on the highD dataset demonstrate that cVMDx substantially outperforms its predecessor cVMD, achieving higher prediction accuracy, diversity, and robustness while maintaining real-time capability for generating multiple trajectory samples. The approach enables fully randomized multimodal trajectory prediction and principled uncertainty quantification.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Accurate and uncertainty-aware trajectory prediction remains a core challenge for autonomous driving, driven by complex multi-agent interactions, diverse scene contexts and the inherently stochastic nature of future motion. Diffusion-based generative models have recently shown strong potential for capturing multimodal futures, yet existing approaches such as cVMD suffer from slow sampling, limited exploitation of generative diversity and brittle scenario encodings. This work introduces cVMDx, an enhanced diffusion-based trajectory prediction framework that improves efficiency, robustness and multimodal predictive capability. Through DDIM sampling, cVMDx achieves up to a 100x reduction in inference time, enabling practical multi-sample generation for uncertainty estimation. A fitted Gaussian Mixture Model further provides tractable multimodal predictions from the generated trajectories. In addition, a CVQ-VAE variant is evaluated for scenario encoding. Experiments on the publicly available highD dataset show that cVMDx achieves higher accuracy and significantly improved efficiency over cVMD, enabling fully stochastic, multimodal trajectory prediction.

Problem

• trajectory prediction
• uncertainty-aware
• multimodal
• autonomous driving
• diffusion model

Innovation

• DDIM sampling
• multimodal trajectory prediction
• uncertainty-aware diffusion model
• Gaussian Mixture Model
• CVQ-VAE