Towards Consistent and Explainable Motion Prediction using Heterogeneous Graph Attention

TL;DR

This paper presents a novel scene encoder using a heterogeneous graph attention network for more consistent and explainable motion prediction in autonomous driving, along with a refinement module to ensure trajectory validity.

Contribution

It introduces a unified heterogeneous graph attention-based scene encoder and a refinement module to improve trajectory consistency and explainability in motion prediction.

Findings

Refined trajectories are more aligned with actual lanes.

Attention analysis provides insights into model decision-making.

The approach enhances prediction consistency and interpretability.

Abstract

In autonomous driving, accurately interpreting the movements of other road users and leveraging this knowledge to forecast future trajectories is crucial. This is typically achieved through the integration of map data and tracked trajectories of various agents. Numerous methodologies combine this information into a singular embedding for each agent, which is then utilized to predict future behavior. However, these approaches have a notable drawback in that they may lose exact location information during the encoding process. The encoding still includes general map information. However, the generation of valid and consistent trajectories is not guaranteed. This can cause the predicted trajectories to stray from the actual lanes. This paper introduces a new refinement module designed to project the predicted trajectories back onto the actual map, rectifying these discrepancies and leading towards more consistent predictions. This versatile module can be readily incorporated into a wide range of architectures. Additionally, we propose a novel scene encoder that handles all relations between agents and their environment in a single unified heterogeneous graph attention network. By analyzing the attention values on the different edges in this graph, we can gain unique insights into the neural network's inner workings leading towards a more explainable prediction.