Cyber-physical models for distributed CAV data intelligence in support of self-organized adaptive traffic signal coordination control

https://doi.org/10.1016/j.eswa.2023.120035Get rights and content

Abstract

While more studies have been focused on adaptive traffic signal control (ATSC) algorithms to learn the control policy from interactions with the traffic environment by using connected and automated vehicle (CAV) data with no human intervention, less mechanism on streamlining the interactions between signal control regulations and CAV-data based traffic features has been integrated into the ATSC algorithms. Such underlying mechanism is of essence to make the CAV-data intelligent functions scalable to varied scales of ATSC system. The presented cyber-physical modeling methodology is attempted to fill in the gap through developing intelligent cell models embedded within the proposed roadside units data processors (RSU-DPs) (i.e., a V2X hub-signalized intersection), which can communicate with on-board units (OBU) mounted on vehicles through C-V2X technologies. The RSU-DPs can directly update the controller for dynamic update of vehicle status and arrivals at the concerned intersection approaches. The intelligent cell model is developed to connect the traffic flow status of the upstream and downstream intersections so as to dynamically adjust or ensure reasonable parameters for signal coordination control (such as offset and/or bandwidth). In this way, the naturalized CAV mobility data can be modeled as a “floating sensing mobility” data source from moving CAVs that will be potentially available ubiquitously on a cycle-to-cycle basis. The simulation-based experiments are designed to test the developed mechanism and models for producing core parameters for implementing the CAV-data-driven self-organized adaptive traffic signal coordination control and testing the functions for improving the efficiency of the corridor mobility, compared with traditional ATSC schemes.

Introduction

Traffic signal coordination control is designed to lead to grouping vehicles as a platoon fleet that can pass through a series of consecutive intersections along a corridor with minimal stops. Handling several consecutive intersections along a corridor has been proved a more efficient and tactic control scheme than conventional isolated signal control schemes in last decades. Since the implementation of infrastructure-based coordinated signal control highly relies on reliability of fixed-point sensors, the performance of the coordinated signal control is often limited against our desired level of effectiveness due to the natures of fixed-point sensors. Some problems may be associated with the limited coverage of the detection areas, high cost of installation and maintenance, and low accuracy of the detected traffic. All of those limitations greatly degrade the performance of quickly capturing the spatiotemporal traffic patterns along the corridor. According to the Traffic Detector Handbook published by Federal Highway Administration (FHWA), loop failures may be resulted from individual loop malfunctions, pavement cracks, and electricity or electronic components issues (Klein, Mills, & Gibson, 2006). Poor sealants, inadequate sealant application, or inadequate splices may cause loop malfunctions. Pavement deteriorates (i.e., pavement cracking and moving) or problems caused by other external factors (i.e., constructions, crashes) are another reason for reliability issues to the loop performance. In addition, breakdown of wire insulation and electronic units, electrical surges, or improper electronics unit tuning often incur the loop failure. Therefore, the high risk of the loop malfunctions or failures is a barrier to reliable signal coordination for a period of time. Traditional data sources are not sufficient to enable the signal coordination to achieve the desired level of effective mobility or congestion mitigation.

With the introduction of emerging connected vehicle (CV) and/or autonomous vehicle (AV) technologies (or CAV technologies), especially, vehicle-to-everything (V2X)-dependent CAV technologies enable vehicles a “floating sensor” data source to seamlessly cover concerned highways over continuous time horizon for traffic monitoring while making the vehicular trajectories trackable (Lin et al., 2022b, Wei et al., 2018). In this circumstance, the CAV-generated data source has a great potential to overcome the weakness of the traditional fixed-point sensor data sources for coordinated signal coordination. The feature of the naturalized “floating sensing mobility” data source that will be potentially available ubiquitously provides the opportunity of capturing critical traffic patterns on a cycle-to-cycle basis. Moreover, cooperative driving with the support of CAV technologies will make it possible to execute self-organized or self-configured adaptive signal coordination control by using the CAV-generated data source.

Despite the optimistic anticipation, little in-depth research progress has been reported on this topic in literature. Most of related literatures concentrate on traditional fixed-time coordinated control and actuated or adaptive coordinated control plans wherein all signal timing parameters are predetermined and fixed, or cycle length is fixed while the green time is varying on a cycle-to-cycle basis in responding to detected traffic demand. While a few researchers have proposed self-organized adaptive signal coordination control strategies by using the CAV-generated data, there is still lack of underlying mechanism and associated intelligent CAV-generated data models that can be applied at scalable scales of the distributed adaptive signal control systems. Supportive data process and analytics technique is still yet mature in self-organized adaptive traffic signal coordination control. To address the challenges, the study presented in this paper aims to systemize an innovated mechanism and data process models to enable adaptive signal coordination with the aid of CAV-enabled cooperative driving functions. The data process models are developed to generate CAV-enabled parameters that can be used as core inputs to adaptive signal controllers for the self-organized adaptive traffic signal coordination control. An arterial corridor in Uptown Cincinnati, OH is targeted as a case study corridor for testing of the developed models.

The remainder of this paper is organized as follows. Section 2 presents a literature review. Section 3 provides methodological framework for distributed intelligent CAV-generated data fusion. Section 4 presents the proof-of-concept study with testing of the developed models. Lastly, Section 5 includes the discussions and conclusions.

Section snippets

Traffic signal coordination control

Effinger (2023) gives a good summary of signal control schemes. More valuably, he pointed out obvious different features of the existing signal controls, namely, pretimed traffic signal, actuated traffic signal, responsive traffic signal and adaptive traffic signal, as summarized in Table 1. This summary is beneficial to better understanding of the features of various types of signal control schemes under varied traffic and roadway conditions.

As early as the 1960s, Morgan and Little (1964)

Overall description of the proposed methodology

Fig. 1 shows the conceptual architecture and data flowchart to configure the CV/AV-generated data settings that can be used to describe traffic characteristics for modeling self-organized adaptive traffic signal coordination control. The C-V2X technologies enable wireless communication among CVs/AVs and CVs/AVs and roadside units data processors (RSU-DPs) with the support of IoT under standard protocols. Among intersections along a concerned corridor, each individual one is referred to as a V2X

Experiment scenario design

To test the developed models, a proof-of-concept study of a real corridor, Martin Luther King Drive East (simply named MLK) in Uptown Cincinnati, Ohio was conducted within the microsimulation VISSIM platform. As shown in Fig. 8, the tested MLK corridor consists of three intersections at Highland Avenue, Burnet Avenue, and Harvey Avenue. To facilitate the discussion, those intersections from left to right are simply named intersection 1, 2 and 3, respectively. According to the Average Annual

Conclusions

The research presented in the paper aims to develop cyber-physical models for making the CV/AV-generated data interoperable with the signal controllers to enhance the self-organized adaptive traffic signal coordination control. The accountable and quality of inputs involved in the phase timing determine the efficiency of the adaptive traffic signal coordination. The traditional fixed-point detection systems have been approved with some inherent limitations such as low coverage and costly

CRediT authorship contribution statement

Wei Lin: Conceptualization, Methodology, Validation, Formal analysis, Writing – original draft, Writing – review & editing, Visualization. Heng Wei: Conceptualization, Methodology, Writing – original draft, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The data used for the paper was partially adopted from the Kleingers group. Any opinions expressed in this paper are those of the authors and do not necessarily reflect the views of the Agency; therefore, no official endorsement should be inferred.

References (77)

  • D. Kim et al.

    Cooperative Traffic Signal Control with Traffic Flow Prediction in Multi-Intersection

    Sensors

    (2020)
  • S.K. Kumaran et al.

    Computer vision-guided intelligent traffic signaling for isolated intersections

    Expert Systems with Applications

    (2019)
  • S. Lee et al.

    Intelligent traffic control for autonomous vehicle systems based on machine learning

    Expert Systems with Applications

    (2020)
  • L. Li et al.

    Adaptive coordinated traffic control for stochastic demand. Transportation research

    Part C, Emerging technologies

    (2018)
  • X. Liang et al.

    An equitable traffic signal control scheme at isolated signalized intersections using Connected Vehicle technology. Transportation research

    Part C, Emerging technologies

    (2020)
  • X. Lin et al.

    A driving-style-oriented adaptive control strategy based PSO-fuzzy expert algorithm for a plug-in hybrid electric vehicle

    Expert Systems with Applications

    (2022)
  • W. Ma et al.

    Multi-stage stochastic program to optimize signal timings under coordinated adaptive control. Transportation research

    Part C, Emerging technologies

    (2016)
  • A. Mirheli et al.

    A consensus-based distributed trajectory control in a signal-free intersection

    Transportation Research Part C Emerging Technologies

    (2019)
  • M. Noaeen et al.

    Reinforcement learning in urban network traffic signal control: A systematic literature review

    Expert Systems with Applications

    (2022)
  • C. Ozan et al.

    A modified reinforcement learning algorithm for solving coordinated signalized networks. Transportation research

    Part C, Emerging technologies

    (2015)
  • P. Serafini et al.

    A mathematical model for the fixed-time traffic control problem

    European journal of operational research

    (1989)
  • M. Wang et al.

    Meta-learning based spatial-temporal graph attention network for traffic signal control

    Knowledge-based systems

    (2022)
  • Y. Wu et al.

    DCL-AIM: Decentralized coordination learning of autonomous intersection management for connected and automated vehicles. Transportation research

    Part C, Emerging technologies

    (2019)
  • K. Yang et al.

    Isolated intersection control for various levels of vehicle technology: Conventional, connected, and automated vehicles

    Transportation Research Part C Emerging Technologies

    (2016)
  • S. Yang et al.

    An inductive heterogeneous graph attention-based multi-agent deep graph infomax algorithm for adaptive traffic signal control

    Information fusion

    (2022)
  • S. Yang et al.

    IHG-MA: Inductive heterogeneous graph multi-agent reinforcement learning for multi-intersection traffic signal control

    Neural networks

    (2021)
  • Y. Zou et al.

    Signal adaptive cooperative control of two adjacent traffic intersections using a two-stage algorithm

    Expert Systems with Applications

    (2021)
  • M. Abbas et al.

    Real-time offset transitioning algorithm for coordinating traffic signals

    Transportation research record

    (2001)
  • G. Abu-Lebdeh et al.

    Signal Coordination and Arterial Capacity in Oversaturated Conditions

    Transportation research record

    (2000)
  • P. Athol

    Headways Groupings

    Transportation research record

    (1965)
  • B. Cesme et al.

    Self-Organizing Control Logic for Oversaturated Arterials

    Transportation research record

    (2013)
  • C.M. Day et al.

    Investigation of Self-Organizing Traffic Signal Control with Graphical Signal Performance Measures

    Transportation Research Record

    (2017)
  • R.W. Denney et al.

    Long Green Times and Cycles at Congested Traffic Signals

    Transportation research record

    (2009)
  • Effinger, J. (2023). Why Are Adaptive Traffic Signals Not Used More in The United States? Retrieved from:...
  • M. Fayyazi et al.

    Real-Time Self-Adaptive Q-learning controller for Energy Management of Conventional Autonomous Vehicles

    Expert Systems with Applications

    (2023)
  • Florida Department of Transportation (FDOT) (2016). Advanced Signal Control...
  • N.H. Gartner et al.

    Optimization of Traffic Signal Settings by Mixed-Integer Linear Programming - 1

    The Network Coordination Problem. Transportation science

    (1975)
  • D.C. Gazis

    Optimum Control of a System of Oversaturated Intersections

    Operations research

    (1964)
  • Cited by (3)

    • Dedicated Bus Arterial Coordination Control Based on Particle Swarm Optimization

      2023, 2023 International Annual Conference on Complex Systems and Intelligent Science, CSIS-IAC 2023
    View full text