Tackling Existence Probabilities of Objects with Motion Planning for Automated Urban Driving

TL;DR

This paper introduces a probabilistic motion planning approach for urban driving that accounts for uncertain object existence, enabling more flexible and reliable navigation despite noisy perception data.

Contribution

It presents a novel planning method that integrates object existence probabilities into motion decisions, improving robustness against perception errors.

Findings

Reduces overly cautious behavior caused by false positive detections

Maintains collision-free operation with uncertain object information

Enhances operational reliability in urban driving scenarios

Abstract

Motion planners take uncertain information about the environment as an input. The environment information is often quite noisy and has a tendency to contain false positive object detection. State-of-the-art motion planners consider all objects alike, thus producing overcautious behavior. In this paper we present a planning approach that considers alternative maneuvers in a combined fashion and plans a motion that is formed by the probabilities of those alternatives. The proposed planner can smoothly react to objects with low existence probability while remaining collision-free in case their existence substantiates. In this way, it tolerates the faults arising from perception and prediction, thus reducing their impact on operational reliability.