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

TL;DR

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 that dynamically captures the uncertainty of surrounding vehicles by online estimation of their acceleration bounds.

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.

Figures (9)

• Figure 1: The forced merging scenario on the highway. Lane 1 is the current lane of the EV, and lane 2 is the target lane, $p_{x, {\rm ter}}$ is the longitudinal terminal position of lane 1, and $w_{\rm lane}$ is the lane width. Note that more surrounding vehicles can be involved in the scenario.
• Figure 2: The predicted reachable set and occupancy of the SV.
• Figure 3: The merging process of the EV planned by the proposed method. The minimum distance between EV and SV0 in the same lane is $\textbf{3.97} \ {\rm m}$.
• Figure 4: The merging process of the EV planned by RMPC. The minimum distance between EV and SV1 in the same lane is $\textbf{34.3} \ {\rm m}$.
• Figure 5: The merging process of the EV planned by DMPC. The minimum distance between EV and SV0 in the same lane is $\textbf{0.17} \ {\rm m}$.

Theorems & Definitions (1)

• Remark IV.1