Uncertainty Matters: Structured Probabilistic Online Mapping for Motion Prediction in Autonomous Driving

2026-03-20

AI Summary

Existing online map generation methods are predominantly deterministic, failing to capture the spatial correlations and inherent uncertainties in road geometry, which limits the reliability of trajectory prediction for autonomous driving. This work proposes a structured probabilistic approach to online map generation, introducing uncertainty modeling into vectorized maps for the first time. By employing a low-rank plus diagonal (LRPD) covariance decomposition, the method effectively captures strong spatial dependencies among road points while preserving global structure and accounting for local noise, all without incurring the high computational cost of full covariance matrices. Built upon an end-to-end trainable probabilistic graph network, the approach significantly improves map generation quality on the nuScenes dataset and achieves a new state-of-the-art in map-based motion prediction, demonstrating the critical value of structured uncertainty modeling for downstream planning tasks.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Reasoning under Uncertainty: Relational Probabilistic Models
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Online map generation and trajectory prediction are critical components of the autonomous driving perception-prediction-planning pipeline. While modern vectorized mapping models achieve high geometric accuracy, they typically treat map estimation as a deterministic task, discarding structural uncertainty. Existing probabilistic approaches often rely on diagonal covariance matrices, which assume independence between points and fail to capture the strong spatial correlations inherent in road geometry. To address this, we propose a structured probabilistic formulation for online map generation. Our method explicitly models intra-element dependencies by predicting a dense covariance matrix, parameterized via a Low-Rank plus Diagonal (LRPD) covariance decomposition. This formulation represents uncertainty as a combination of a low-rank component, which captures global spatial structure, and a diagonal component representing independent local noise, thereby capturing geometric correlations without the prohibitive computational cost of full covariance matrices. Evaluations on the nuScenes dataset demonstrate that our uncertainty-aware framework yields consistent improvements in online map generation quality compared to deterministic baselines. Furthermore, our approach establishes new state-of-the-art performance for map-based motion prediction, highlighting the critical role of uncertainty in planning tasks. Code is published under link-available-soon.

Problem

• uncertainty
• online mapping
• spatial correlation
• autonomous driving
• probabilistic modeling

Innovation

• structured uncertainty
• probabilistic mapping
• low-rank covariance
• spatial correlation
• motion prediction