STRAP: Spatial-Temporal Risk-Attentive Vehicle Trajectory Prediction for Autonomous Driving

TL;DR

This paper introduces a risk-aware vehicle trajectory prediction framework that incorporates a risk potential field and a risk-scaled loss to improve accuracy and safety in autonomous driving, especially in high-risk scenarios.

Contribution

It proposes a novel spatial-temporal risk-attentive framework with a risk potential field and a risk-scaled loss, enhancing prediction accuracy and interpretability in autonomous driving.

Findings

Reduces average prediction errors by 4.8% on NGSIM dataset

Reduces average prediction errors by 31.2% on HighD dataset

Improves prediction accuracy in high-risk scenarios

Abstract

Accurate vehicle trajectory prediction is essential for ensuring safety and efficiency in fully autonomous driving systems. While existing methods primarily focus on modeling observed motion patterns and interactions with other vehicles, they often neglect the potential risks posed by the uncertain or aggressive behaviors of surrounding vehicles. In this paper, we propose a novel spatial-temporal risk-attentive trajectory prediction framework that incorporates a risk potential field to assess perceived risks arising from behaviors of nearby vehicles. The framework leverages a spatial-temporal encoder and a risk-attentive feature fusion decoder to embed the risk potential field into the extracted spatial-temporal feature representations for trajectory prediction. A risk-scaled loss function is further designed to improve the prediction accuracy of high-risk scenarios, such as short relative spacing. Experiments on the widely used NGSIM and HighD datasets demonstrate that our method reduces average prediction errors by 4.8% and 31.2% respectively compared to state-of-the-art approaches, especially in high-risk scenarios. The proposed framework provides interpretable, risk-aware predictions, contributing to more robust decision-making for autonomous driving systems.