An Intersection Platoon Speed Control Model Considering Traffic Efficiency and Energy Consumption in CVIS

Wenhao Ren,Qian Zhou,Li Li,Junyou Zhang,Yunfeng Hu

doi:10.1155/2021/2891247

Abstract

This paper proposed an intersection platoon speed control model considering traffic efficiency and energy consumption in the cooperative vehicle-infrastructure systems (CVIS) environment. This model divides the control situation in detail according to the different state of signal lights at the intersection and splits the platoon that cannot pass the intersection completely. The optimization model is established by taking the traffic delay and energy consumption of the platoon as the control objectives, and the model is solved by using a genetic algorithm (GA). Finally, the simulation platform is built by SUMO traffic simulation software, MATLAB, and Python to verify the model. The simulation results show that the total number of queued vehicles, the maximum number of queued vehicles, and the mean travel time of vehicles decreased by 77.81%, 33.33%, and 10.95%, respectively. Besides, the total fuel consumption is reduced by 19.95%, the total emissions of CO2, CO, HC, NOx, and PMx decreased by 19.96%, 58.55%, 51.33%, 23.81%, and 37.51%, respectively. It indicates that the proposed platoon speed control model can effectively improve traffic efficiency while reducing energy consumption and pollutant emissions.

Highlights

With the rapid growth of car ownership, the existing urban traffic infrastructure construction level cannot meet the growing demand for traffic travel, and urban traffic problems are increasingly serious [1, 2]
In the traditional “go-stop” type transport operation mode, the collaboration between the vehicles is less, and the vehicle just passively accept intersection signal control, which leads to the vehicle in the process of the intersection of frequent deceleration or startstop, reducing the efficiency of intersection, increasing energy consumption and pollutant emissions of the vehicle at the same time, and causing traffic congestion, environmental pollution, and a series of problems [4, 5]
Wu et al [10] proposed a novel multiagent recurrent deep deterministic policy gradient (MARDDPG) algorithm based on deep deterministic policy gradient (DDPG) algorithm for traffic light control (TLC) in vehicle networks

Summary

Introduction

With the rapid growth of car ownership, the existing urban traffic infrastructure construction level cannot meet the growing demand for traffic travel, and urban traffic problems are increasingly serious [1, 2]. In the traditional “go-stop” type transport operation mode, the collaboration between the vehicles is less, and the vehicle just passively accept intersection signal control, which leads to the vehicle in the process of the intersection of frequent deceleration or startstop, reducing the efficiency of intersection, increasing energy consumption and pollutant emissions of the vehicle at the same time, and causing traffic congestion, environmental pollution, and a series of problems [4, 5]. Rafter et al [9] proposed a new traffic signal control algorithm, multimode adaptive traffic signals (MATS), which combines information from existing fixed-time plans and loop detectors, and position data from connected vehicles to perform decentralized control on signalized intersections. Wu et al [10] proposed a novel multiagent recurrent deep deterministic policy gradient (MARDDPG) algorithm based on deep deterministic policy gradient (DDPG) algorithm for traffic light control (TLC) in vehicle networks

Methods

Results

Conclusion