Abstract
Traffic oscillations, known more commonly as ”stop-and-go” driving conditions in heavily congested traffic, result in numerous negative impacts. Not only do they reduce effective capacity, but they also increase safety risks, travel delay, and fuel consumption and emissions. Researchers have hypothesized that strategies intended to harmonize the speed of vehicles will reduce traffic oscillations and reduce the adverse impacts. A speed harmonization system that utilizes the capabilities of connected and automated vehicles (CAV) has been developed that dynamically and automatically adjusts the speeds of vehicles based on current traffic conditions. While researchers have evaluated speed harmonization in simulation studies, there have been no large-scale field evaluations of speed harmonization systems. This thesis presents the results of the continuation of the first large-scale field test of a speed harmonization system using CAVs. The objectives of this research effort are threefold: to demonstrate the feasibility of the implementation of speed harmonization in a real-world environment, to determine the effectiveness of speed harmonization in reducing traffic oscillations, and to investigate the potential of speed harmonization in improving traffic performance.
In a field test of a speed harmonization system prototype, a fleet consisting of three CAVs and five probe vehicles was deployed on a roadway segment that experiences daily recurring traffic congestion. The fleet was led by one probe vehicle, followed by the three CAVs driving nearly parallel to one another, followed by the four remaining probe vehicles. The objective of the CAVs was to regulate traffic upstream of the bottleneck so that vehicles move with uniform speed, thereby creating a steadier flow of traffic.
To demonstrate the feasibility of the implementation of speed harmonization in a real-world environment, recommended and actual speeds of the CAVs were compared. Results of these comparisons showed that the speeds of the CAVs compared favorably to recommended speeds, thereby implying that the implementation of speed harmonization using CAVs is feasible. To evaluate the effectiveness of speed harmonization on smoothing the flow of traffic by reducing traffic oscillations, the oscillatory behavior of the leading probe vehicle, which represents the “general” flow of traffic unaffected by the CAVs, and the following probe vehicles that followed the CAVs were analyzed and compared. Power spectral density (PSD) analysis was used to analyze traffic oscillations. Results showed that there was a statistically significant difference in each PSD comparison between the lead probe vehicle (P0) and the following probe vehicles. The PSDs of P1, P2, P3, and P4 were reduced by 29%, 36%, 90%, and 84%, respectively, when compared to the PSD of P0. Therefore, it can be concluded that speed harmonization can reduce speed oscillations.
Finally, to determine whether speed harmonization had additional benefits, specifically with regards to mobility and the environment, travel time and fuel consumption were measured. Results showed that the average travel time of each probe vehicle was very similar, thus implying there is no statistically significant difference in travel time. On the other hand, it was found that there were statistically significant differences in the fuel consumption of the probe vehicles. However, these results showed increases in the fuel consumption of the following probe vehicles in comparison to that of the leading probe vehicle. This was not unexpected, as the speed harmonization algorithm primarily focused on smoothing the flow of traffic by reducing traffic oscillations, not explicitly considering other performance measures. The results presented in this thesis will provide more knowledge and a better understanding of the potential benefits of speed harmonization using CAVs.
