RAIST: Learning Risk Aware Traffic Interactions via Spatio-Temporal Graph Convolutional Networks

TL;DR

This paper introduces RAIST, a novel framework using spatio-temporal graph convolutional networks to model traffic interactions and predict risk-aware behaviors in autonomous driving scenarios.

Contribution

It proposes a new traffic graph-based approach that models spatial and temporal interactions and intentions of road users for risk-aware decision making.

Findings

Improves risk object identification, especially for pedestrians and cyclists.

Learns risk-aware representations effectively.

Enhances understanding of traffic interactions in autonomous driving.

Abstract

A key aspect of driving a road vehicle is to interact with other road users, assess their intentions and make risk-aware tactical decisions. An intuitive approach to enabling an intelligent automated driving system would be incorporating some aspects of human driving behavior. To this end, we propose a novel driving framework for egocentric views based on spatio-temporal traffic graphs. The traffic graphs model not only the spatial interactions amongst the road users but also their individual intentions through temporally associated message passing. We leverage a spatio-temporal graph convolutional network (ST-GCN) to train the graph edges. These edges are formulated using parameterized functions of 3D positions and scene-aware appearance features of road agents. Along with tactical behavior prediction, it is crucial to evaluate the risk-assessing ability of the proposed framework. We claim that our framework learns risk-aware representations by improving on the task of risk object identification, especially in identifying objects with vulnerable interactions like pedestrians and cyclists.