A Driver-in-the Loop Fuel Economic Control Strategy for Connected Vehicles in Urban Roads

TL;DR

This paper presents a decentralized, driver-in-the-loop control strategy for connected vehicles that improves fuel economy and traffic flow in urban environments by integrating V2V communication, traffic light info, and driver error modeling.

Contribution

It introduces a novel stochastic model predictive control framework that accounts for driver errors and uses local information for each vehicle to enhance fuel efficiency and traffic mobility.

Findings

Driver error modeling impacts fuel economy strategies.

Decentralized control improves urban traffic flow.

Simulation confirms effectiveness of the proposed method.

Abstract

In this paper, we focus on developing driver-in-the loop fuel economic control strategy for multiple connected vehicles. The control strategy is considered to work in a driver assistance framework where the controller gives command to a driver to follow while considering the ability of the driver in following control commands. Our proposed method uses vehicle-to-vehicle (V2V) communication, exploits traffic lights' Signal Phase and Timing (SPAT) information, models driver error injection with Markov chain, and employs scenario tree based stochastic model predictive control to improve vehicle fuel economy and traffic mobility. The proposed strategy is decentralized in nature as every vehicle evaluates its own strategy using only local information. Simulation results show the effect of consideration of driver error injection when synthesizing fuel economic controllers in a driver assistance fashion.