Delving into Mapping Uncertainty for Mapless Trajectory Prediction

TL;DR

This paper introduces a novel approach to incorporate map uncertainty into mapless trajectory prediction for autonomous driving, leveraging vehicle kinematics and map geometry to improve prediction accuracy and interpretability.

Contribution

It proposes a Proprioceptive Scenario Gating method and a Covariance-based Map Uncertainty approach, enhancing the integration of map uncertainty into trajectory prediction.

Findings

Achieves up to 23.6% performance improvement over state-of-the-art.

Effectively identifies scenarios where map uncertainty benefits prediction.

Provides interpretable insights into the role of uncertainty in driving scenarios.

Abstract

Recent advances in autonomous driving are moving towards mapless approaches, where High-Definition (HD) maps are generated online directly from sensor data, reducing the need for expensive labeling and maintenance. However, the reliability of these online-generated maps remains uncertain. While incorporating map uncertainty into downstream trajectory prediction tasks has shown potential for performance improvements, current strategies provide limited insights into the specific scenarios where this uncertainty is beneficial. In this work, we first analyze the driving scenarios in which mapping uncertainty has the greatest positive impact on trajectory prediction and identify a critical, previously overlooked factor: the agent's kinematic state. Building on these insights, we propose a novel Proprioceptive Scenario Gating that adaptively integrates map uncertainty into trajectory prediction based on forecasts of the ego vehicle's future kinematics. This lightweight, self-supervised approach enhances the synergy between online mapping and trajectory prediction, providing interpretability around where uncertainty is advantageous and outperforming previous integration methods. Additionally, we introduce a Covariance-based Map Uncertainty approach that better aligns with map geometry, further improving trajectory prediction. Extensive ablation studies confirm the effectiveness of our approach, achieving up to 23.6% improvement in mapless trajectory prediction performance over the state-of-the-art method using the real-world nuScenes driving dataset. Our code, data, and models are publicly available at https://github.com/Ethan-Zheng136/Map-Uncertainty-for-Trajectory-Prediction.