Verifiable Goal Recognition for Autonomous Driving with Occlusions

TL;DR

This paper introduces OGRIT, a decision tree-based goal recognition method for autonomous driving that effectively handles occlusions, providing accurate, interpretable, and verifiable predictions across various scenarios.

Contribution

The paper presents a novel goal recognition approach using interpretable decision trees that manage occlusions and missing data in autonomous driving environments.

Findings

OGRIT is computationally fast and accurate.

It effectively handles occlusions and missing data.

The method generalizes across multiple scenarios.

Abstract

Goal recognition (GR) involves inferring the goals of other vehicles, such as a certain junction exit, which can enable more accurate prediction of their future behaviour. In autonomous driving, vehicles can encounter many different scenarios and the environment may be partially observable due to occlusions. We present a novel GR method named Goal Recognition with Interpretable Trees under Occlusion (OGRIT). OGRIT uses decision trees learned from vehicle trajectory data to infer the probabilities of a set of generated goals. We demonstrate that OGRIT can handle missing data due to occlusions and make inferences across multiple scenarios using the same learned decision trees, while being computationally fast, accurate, interpretable and verifiable. We also release the inDO, rounDO and OpenDDO datasets of occluded regions used to evaluate OGRIT.