Uncertainty-Aware Decision-Making and Planning for Autonomous Forced Merging

TL;DR

This paper presents an uncertainty-aware decision-making and planning method for autonomous vehicles in forced merging scenarios, dynamically estimating surrounding vehicle behaviors to improve safety and resilience.

Contribution

It introduces a novel online estimation of surrounding vehicles' acceleration bounds and integrates this uncertainty into decision-making and planning for autonomous merging.

Findings

Successfully handles challenging forced merging scenarios.

Outperforms several alternative approaches in safety and resilience.

Enhances autonomous vehicle decision-making with real-time uncertainty estimation.

Abstract

In this paper, we develop an uncertainty-aware decision-making and motion-planning method for an autonomous ego vehicle in forced merging scenarios, considering the motion uncertainty of surrounding vehicles. The method dynamically captures the uncertainty of surrounding vehicles by online estimation of their acceleration bounds, enabling a reactive but rapid understanding of the uncertainty characteristics of the surrounding vehicles. By leveraging these estimated bounds, a non-conservative forward occupancy of surrounding vehicles is predicted over a horizon, which is incorporated in both the decision-making process and the motion-planning strategy, to enhance the resilience and safety of the planned reference trajectory. The method successfully fulfills the tasks in challenging forced merging scenarios, and the properties are illustrated by comparison with several alternative approaches.