The New Jersey Institute of Technology's
Electronic Theses & Dissertations Project

Title:	Assessing the effectiveness of managed lane strategies for the rapid deployment of cooperative adaptive cruise control technology
Author:	Zhong, Zijia
View Online:	njit-etd2018-033 (xvii, 132 pages ~ 4.9 MB pdf)
Department:	Department of Civil and Environmental Engineering
Degree:	Doctor of Philosophy
Program:	Transportation
Document Type:	Dissertation
Advisory Committee:	Lee, Joyoung (Committee chair) Chien, I-Jy Steven (Committee member) Spasovic, Lazar (Committee member) Wang, Guiling (Committee member) Bhavsar, Parth (Committee member)
Date:	2018-05
Keywords:	Advanced driving assistance system Connected and automated vehicle Cooperative adaptive cruise control Managed lane strategy Microscopic traffic simulation Multiobjective optimization
Availability:	Unrestricted
Abstract:	Connected and Automated Vehicle (C/AV) technologies are fast expanding in the transportation and automotive markets. One of the highly researched examples of C/AV technologies is the Cooperative Adaptive Cruise Control (CACC) system, which exploits various vehicular sensors and vehicle-to-vehicle communication to automate vehicular longitudinal control. The operational strategies and network-level impacts of CACC have not been thoroughly discussed, especially in near-term deployment scenarios where Market Penetration Rate (MPR) is relatively low. Therefore, this study aims to assess CACC's impacts with a combination of managed lane strategies to provide insights for CACC deployment. The proposed simulation framework incorporates 1) the Enhanced Intelligent Driver Model; 2) Nakagami-based radio propagation model; and 3) a multi-objective optimization (MOOP)-based CACC control algorithm. The operational impacts of CACC are assessed under four managed lane strategies (i.e., mixed traffic (UML), HOV (High Occupancy Vehicle)-CACC lane (MML), CACC dedicated lane (DL), and CACC dedicated lane with access control (DLA)). Simulation results show that the introduction of CACC, even with 10% MPR, is able to improve the network throughput by 7% in the absence of any managed lane strategies. The segment travel times for both CACC and non-CACC vehicles are reduced. The break-even point for implementing dedicated CACC lane is 30% MPR, below which the priority usage of the current HOV lane for CACC traffic is found to be more appropriate. It is also observed that DLA strategy is able to consistently increase the percentage of platooned CACC vehicles as MPR grows. The percentage of CACC vehicles within a platoon reaches 52% and 46% for DL and DLA, respectively. When it comes to the impact of vehicle-to-vehicle (V2V), it is found that DLA strategy provides more consistent transmission density in terms of median and variance when MPR reaches 20% or above. Moreover, the performance of the MOOP-based cooperative driving is examined. With average 75% likelihood of obtaining a feasible solution, the MOOP outperforms its counterpart which aims to minimize the headway objective solely. In UML, MML, and DL strategy, the proposed control algorithm achieves a balance spread among four objectives for each CACC vehicle. In the DLA strategy, however, the probability of obtaining feasible solution falls to 60% due to increasing size of platoon owing to DLA that constraints the feasible region by introduction more dimensions in the search space. In summary, UML or MML is the preferred managed lane strategy for improving traffic performance when MPR is less than 30%. When MRP reaches 30% or above, DL and DLA could improve the CACC performance by facilitating platoon formation. If available, priority access to an existing HOV lane can be adopted to encourage adaptation of CACC when CACC technology becomes publically available.