Importance is in your attention: agent importance prediction for autonomous driving

TL;DR

This paper demonstrates that attention mechanisms in trajectory prediction models can be repurposed to measure and rank the importance of surrounding agents for autonomous vehicle planning.

Contribution

The authors propose a novel method to extract agent importance from attention weights in trajectory prediction models, enhancing understanding of agent influence.

Findings

Effective ranking of agents by importance on nuPlans dataset

Attention-based importance correlates with actual impact on ego trajectory

Method improves interpretability of autonomous driving models

Abstract

Trajectory prediction is an important task in autonomous driving. State-of-the-art trajectory prediction models often use attention mechanisms to model the interaction between agents. In this paper, we show that the attention information from such models can also be used to measure the importance of each agent with respect to the ego vehicle's future planned trajectory. Our experiment results on the nuPlans dataset show that our method can effectively find and rank surrounding agents by their impact on the ego's plan.