Intelligent Intersection Management Systems Considering Autonomous Vehicles: A Systematic Literature Review

Abstract

Over the past several decades, the development of technologies and the production of autonomous vehicles have enhanced the need for intelligent intersection management systems. Subsequently, growing interest in studying the traffic management of autonomous vehicles at intersections has been evident, which indicates a critical need to conduct a systematic literature review on this topic. This paper offers a systematic review of the proposed methodologies for intelligent intersection management systems and presents the remaining research gaps and possible future research approaches. We consider both pure autonomous vehicle traffic and mixed traffic at four-way signalized and unsignalized intersection(s). We searched for articles published from 2008 to 2019, and identified 105 primary studies. We applied the thematic analysis method to analyze the extracted data, which led to the identification of four main classes of methodologies, namely rule-based, optimization, hybrid, and machine learning methods. We also compared how well the methods satisfy their goals, namely efficiency, safety, ecology, and passenger comfort. This analysis allowed us to determine the primary challenges of the presented methodologies and propose new approaches in this area.INDEX TERMS Autonomous vehicle, intelligent intersection management system, mixed traffic, vehicleto-infrastructure (V2I) communication, vehicle-to-vehicle (V2V) communication.

Connectivity and Automation as Enablers for Energy-Efficient Driving and Road Traffic Management

“…Currently, few works are based on machine learning approaches (4%). Namazi et al (2019) also explain that the vast majority (93%) of the articles reviewed consider a CAVs penetration rate of 100%. To be more realistic, autonomous intersections should be compliant with mixed traffic (CAVs and conventional HVs), pedestrians, and cyclists.…”

“…8, of such an autonomous intersection, in which vehicles negotiate the "right of way" using V2V communication. More generally, a systematic literature review of intelligent intersection management systems for CAVs is proposed by Namazi et al (2019). The authors indicate that some works on this topic consider rule-based methodologies (40%), while others consider optimization methodologies (45%), and hybrid methodologies (11%).…”

Efficiently Bounding the Probabilities of Vehicle Collision at Intelligent Intersections

“…The literature on intelligent intersection management is extensive. A recent literature review on the subject [1] classifies "safety" as one of five subcategories or subgoals within this literature. In addition, within studies on "safety", there are also two main directions: "collision avoidance" [2] and "resolving conflict" [3], [4].…”

RACE: Reinforced Cooperative Autonomous Vehicle Collision Avoidance

“…However, rule-based solutions do not easily adapt to different scenarios, while optimization-based methods may rapidly converge to local minimal solutions and hence, have relatively low computational efficiency. In addition, a study by Namazi et al [34] shows that traditional machine learning-based solutions are not suitable for a complex and dynamic environment such as autonomous driving. Leveraging deep learning especially the Convolutional Neural Networks (CNNs), Lv et al [35] handled collision avoidance by predicting the traffic flow while Chen et al [10] utilized DRL with multi-agents settings to avoid collisions.…”