Wang, Xia (Vanderbilt University), Yang, Yuwei (Vanderbilt University), Shangguan, Yifan (Vanderbilt University), Yan, Weiyu (Vanderbilt University), An, Ziyan (Vanderbilt University), Bunting, Matt (Vanderbilt University), Nice, Matthew (Vanderbilt University), Beckers, Thomas (Vanderbilt University), Ma, Meiyi (Vanderbilt University), Work, Daniel (Vanderbilt University), Sprinkle, Jonathan (Vanderbilt University)

Interpretable Finite State Machine Controller: A Case Study on Lane Merge Yield Mode

Scheduled for presentation during the Regular Session "Driver Assistance Systems II" (FrBT1), Friday, September 27, 2024, 14:10−14:30, Salon 1

2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), September 24- 27, 2024, Edmonton, Canada

This information is tentative and subject to change. Compiled on December 26, 2024

Abstract

This paper proposes a merge yield mode for Adaptive Cruise Control (ACC) that can switch with the traditional car-following mode to meet macroscopic requirements of safety and comfort in traffic flow. In car-following mode, an Autonomous Vehicle (AV) follows a time gap with the preceding vehicle. However, the system may struggle to guarantee safety and comfort in complex and variable driving scenarios where vehicles enter and leave the lane of travel. Motivated by this, the paper proposes a multi-mode ACC based on interpretable Finite State Machine (FSM) that addresses challenges in infrastructure changes and vehicles merging into (and out of) the current lane of travel. Also, we present results from an implementation that was successfully tested on the open road, and simulation results that show dampened disturbances from the mode-switching approach, compared to single-mode use of an ACC controller in the same scenarios.