Open-set Intersection Intention Prediction for Autonomous Driving

TL;DR

This paper introduces an open-set intention prediction model for autonomous driving at intersections, leveraging map-centric features and attention mechanisms to handle diverse and unseen intersection configurations effectively.

Contribution

It formulates intersection intention prediction as an open-set problem using a novel spatial-temporal graph approach with attention modules, enabling direct transfer from simulation to real-world data.

Findings

Model trained on simulated data performs well on real-world intersections.

Attention scores effectively estimate intention probabilities.

The approach handles diverse intersection layouts without fine-tuning.

Abstract

Intention prediction is a crucial task for Autonomous Driving (AD). Due to the variety of size and layout of intersections, it is challenging to predict intention of human driver at different intersections, especially unseen and irregular intersections. In this paper, we formulate the prediction of intention at intersections as an open-set prediction problem that requires context specific matching of the target vehicle state and the diverse intersection configurations that are in principle unbounded. We capture map-centric features that correspond to intersection structures under a spatial-temporal graph representation, and use two MAAMs (mutually auxiliary attention module) that cover respectively lane-level and exitlevel intentions to predict a target that best matches intersection elements in map-centric feature space. Under our model, attention scores estimate the probability distribution of the openset intentions that are contextually defined by the structure of the current intersection. The proposed model is trained and evaluated on simulated dataset. Furthermore, the model, trained on simulated dataset and without any fine tuning, is directly validated on in-house real-world dataset collected at 98 realworld intersections and exhibits satisfactory performance,demonstrating the practical viability of our approach.