Keping Li

Dynamic Eco-Driving Speed Guidance at Signalized Intersections: Multivehicle Driving Simulator Based Experimental Study

Variations in vehicle fuel consumption and gas emissions are usually associated with changes in cruise speed and the aggressiveness of drivers’ acceleration/deceleration, especially at traffic signals. In an attempt to enhance vehicle fuel efficiency on arterials, this study developed a dynamic eco-driving speed guidance strategy (DESGS) using real-time signal timing and vehicle positioning information in a connected vehicle (CV) environment. DESGS mainly aims to optimize the fuel/emission speed profiles for vehicles approaching signalized intersections. An optimization-based rolling horizon and a dynamic programming approach were proposed to track the optimal guided velocity for individual vehicles along the travel segment. In addition, a vehicle specific power (VSP) based approach was integrated into DESGS to estimate the fuel consumption and CO2 emissions. To evaluate the effectiveness of the overall strategy, 15 experienced drivers were recruited to participate in interactive speed guidance experiments using multivehicle driving simulators. It was found that compared to vehicles without speed guidance, those with DESGS had a significantly reduced number of stops and approximately 25% less fuel consumption and CO2 emissions.

Real-Time Queue Length Estimation for Signalized Intersections Using Vehicle Trajectory Data

Queue length is one of the most important performance measures for signalized intersections. Many methods for queue length estimation based on various data sources have been proposed in the literature. With the latest developments and applications of probe vehicle systems, cycle-by-cycle queue length estimation based only on probe data has become a promising research topic. However, most existing methods assume that information such as signal timing, arrival pattern, and penetration rate is known, an assumption that constrains their applicability in practice. The objective of this study was to propose a cycle-by-cycle queue length estimation method using only probe data without the foregoing assumption. Based on the shock wave theory, the proposed method is capable of reproducing the dynamic queue forming and dissipating process cycles at signalized intersections by using probe vehicle trajectories. To reproduce the queuing processes, the inflection points of probe vehicle trajectories representing the changes of arrival patterns are identified and extracted from the trajectory points of vehicles joining and leaving the queue. A piecewise linear function is then used to fit all the inflection points to estimate the stopping and discharging shock waves. Finally, signal timing data and queue lengths can be calculated on the basis of the estimated shock waves. Under both saturated and oversaturated traffic conditions, the performance of the method is comprehensively evaluated through 60 simulation scenarios, which cover sampling intervals from 5 s to 60 s and penetration rates ranging from 5% to 100%. Results show that compared with the method proposed by Ramezani and Geroliminis in 2015, the proposed method has more robustness for all the sampling intervals and displays more estimation accuracy of queue length and a higher success rate under conditions of low penetration rate.

Modeling Risky Driver Behavior Under the Influence of Flashing Green Signal with Vehicle Trajectory Data

In China, a 3-s flashing green (FG) indication followed by a 3-s yellow (Y) indication is commonly applied to end the green phase at signalized intersections regardless of speed limits. Past research has indicated that the FG may bring more uncertainty and dynamics to drivers’ stop–go decisions at the end of the green phase, particularly at high-speed intersections. Hence, the objective of this study was to propose a probabilistic model for interpreting drivers’ dynamic decision-making processes in response to the combination of FG and Y at rural high-speed intersections, which could help quantitatively analyze the safety effects of FG. The inputs of the proposed model included mainly the acceleration and deceleration rate and the perception-and-reaction time. The outputs were the frequencies of risky behavior, including abrupt stop and aggressive pass on the basis of the cycle. The trajectories of 1,186 vehicles, collected at three rural high-speed intersections in Shanghai, China, with a speed limit of 80 km/h were used for the uncertainty analysis of input variables and model validation. On the basis of the proposed model, three typical scenarios of phase transition periods—3-s FG + 3-s Y, 0-s FG + 5-s Y, and 0-s FG + 3-s Y—were valuated. Results showed that the proposed model was capable of producing an average error of less than 10%. In addition, the scenario of 0-s FG + 5-s Y performed best in reducing risky behavior. It is recommended that a 5-s Y be implemented for high-speed intersections when most of the traffic consists of passenger cars.

A Generalized Assessment Model for the Amount of Information of Variable Message Signs

Variable message signs (VMS) have recently been implemented as a key component of transportation information systems. However, technical guidelines of VMS designing remain absent. Hence, this study proposed a generalized assessment model for the amount of information (AOI) of VMS. In the proposed model, the AOI of VMS is defined as the reduced uncertainty of driver’ selection. The advantages of the assessment model are: 1) to reflect the effect of the VMS on driver behavior, 2) to consider the driving task, and 3) to take drivers as an entity to analyze. Then experiments were performed to validate the proposed model. Learning curve and hierarchical linear models were applied to quantify the relationships between the AOI of VMS, response time, and watching frequencies. The results show that the AOI increase will lead to response time increase. The proposed model can be transferred to properly design VMS and to evaluate existing VMS.

Travel Time Estimation for Recurrent Jammed Flow on Shanghai North-South Expressway

The conventional point-detector based method for travel time estimation does not consider the impact of future downstream traffic conditions and may deviate significantly from real travel times during congestion. Though many newly arising algorithms have been developed, it is difficult for them to be applied in real situation. This paper proposes a novel and easy-to-implement method of travel time estimation using effective capacity at the bottlenecks of expressway based on actual data during recurrent congestion periods. Then the method is applied to Shanghai North-south expressway in the microscopic simulation program VISSIM. The experiment results show the accuracy is satisfying in 5-minute intervals.

A Method of Traffic Flow Forecast and Management

As an important aspect in transportation research, the forecast of traffic state serves as the basis for reasonable traffic control and management. Original prediction models in actual appliance donpsilat consider influencing factors comprehensively when dealing with complex and changeable traffic state. Based on the observation of daily similarity of expressway traffic flow series, the paper recognizes weather and date attribute as classification vectors and designs a new time series clustering algorithm through calculating representatives in all kinds of clusters and building correlative relations between observing time series and representatives. In the end traffic flow under different time scales on both weekday and weekend are forecasted. The experiment result shows that the time series clustering algorithm and forecast model make good use of historical data and the accuracy is quite satisfied under 30-minute time scale with the mean-absolute-relative-error 3.34 percent.