HGNET: A Hierarchical Feature Guided Network for Occupancy Flow Field Prediction

TL;DR

HGNET is a transformer-based hierarchical network that improves occupancy flow prediction in autonomous driving by modeling agent interactions and temporal dependencies, achieving competitive results in a major challenge.

Contribution

The paper introduces HGNET, a novel hierarchical feature guided network with FGAT and Time Series Memory modules for better multi-agent motion prediction.

Findings

Achieved 3rd place in the 2024 Waymo Occupancy and Flow Prediction Challenge.

Demonstrated improved accuracy over existing methods.

Effectively models multimodal interactions and temporal dependencies.

Abstract

Predicting the motion of multiple traffic participants has always been one of the most challenging tasks in autonomous driving. The recently proposed occupancy flow field prediction method has shown to be a more effective and scalable representation compared to general trajectory prediction methods. However, in complex multi-agent traffic scenarios, it remains difficult to model the interactions among various factors and the dependencies among prediction outputs at different time steps. In view of this, we propose a transformer-based hierarchical feature guided network (HGNET), which can efficiently extract features of agents and map information from visual and vectorized inputs, modeling multimodal interaction relationships. Second, we design the Feature-Guided Attention (FGAT) module to leverage the potential guiding effects between different prediction targets, thereby improving prediction accuracy. Additionally, to enhance the temporal consistency and causal relationships of the predictions, we propose a Time Series Memory framework to learn the conditional distribution models of the prediction outputs at future time steps from multivariate time series. The results demonstrate that our model exhibits competitive performance, which ranks 3rd in the 2024 Waymo Occupancy and Flow Prediction Challenge.