Categorizing Driving Style Using Connected Vehicle Data: Application of Unsupervised Learning
Amin Mohammadnazar, University of Tennessee, KnoxvilleShow Abstract
Ramin Arvin, University of Tennessee, Knoxville
Asad J. Khattak, University of Tennessee
Driving styles can be broadly classified as ranging from aggressive to calm. Such classification can be used to design customized driver assistance systems, assess mobility, crash risk, fuel consumption, and emissions. While a range of methods has been used in the past for driving style classification, the emergence of connected vehicles equipped with communication devices, provide a new opportunity to classify driving style using microscopic real-world data. This can enable providing drivers with better assistance as they drive. The main objective of this study is to propose a framework to harness Basic Safety Messages (BSMs) to classify driving styles in different spatial and temporal contexts using unsupervised learning methods. To this end, a subset of the BSM data from the Safety Pilot Model Deployment (SPMD) is used to analyze the driving styles of 1300 individuals making trips on diverse roadways and through several neighborhoods in Ann Arbor, Michigan. Driving volatility is used as a key measure for clustering the data. Then analysis was performed on the data, which showed that the k-means method provides better classification results compared to the k-medoids method. Additionally, the thresholds for aggressive, and calm driving vary across different neighborhoods. Subsequently, a Driving Score was created to measure driving performance consistently across drivers. The results show that the proportion of aggressive driving style on commercial streets seems higher than on highways and in residential neighborhoods.
Performance Evaluation of Field-Test Vehicles Platooning: Application of Adaptive Cruise Control and Cooperative Adaptive Cruise Control
Iman Mahdinia, University of Tennessee, KnoxvilleShow Abstract
Ramin Arvin, University of Tennessee, Knoxville
Asad J. Khattak, University of Tennessee
Amir Ghiasi, Leidos, Inc.
Connected and automated vehicle technologies have the potential to significantly improve transportation system performance. In particular, advanced driver-assistance systems, such as Adaptive Cruise Control (ACC) and Cooperative Adaptive Cruise Control (CACC), may lead to substantial improvements in sustainable mobility by decreasing driver inputs and taking over control of the vehicle. However, the impacts of these technologies on the vehicle- and system-level energy consumption, emissions, and safety have not been quantified in field tests. The main goal of this paper is to study the impacts of automated and cooperative systems in mixed traffic of conventional, ACC, and CACC vehicles. To reach this goal, experimental data based on real-world conditions are collected (in tests conducted by FHWA) with adaptions of ACC, CACC, and conventional vehicles in a vehicle platoon scenario and a merging scenario. Specifically, the platoon of five-vehicles with different vehicle type combinations is considered and analyzed to generate new knowledge about potential safety, energy efficiency, and emission improvement from vehicle automation and cooperation. Results show that adopting the CACC system in a five-vehicle platoon substantially reduces the driving volatility and reduces the risk of rear-end collision and consequently improves safety. Furthermore, it decreases fuel consumption and emissions compared with the CACC system and manual-driven vehicles. Regarding the merging scenario, the results show that while the cooperative merging system reduces the driving volatility slightly, the fuel consumption and emissions can increase due to sharper accelerations of CACC vehicles compared with manually-driven vehicles.
Characterization of Drivers’ Engagement in Secondary Tasks: Application of Deep Learning for Data-Driven In-Vehicle Systems
Osama Osman, University of Tennessee, ChattanoogaShow Abstract
Hesham Rakha, Virginia Polytechnic Institute and State University (Virginia Tech)
Distracted driving is an epidemic that threatens the lives of thousands every year. Data collected from vehicular sensor technologies and through connectivity provide comprehensive information that, if used to detect driver engagement in secondary tasks, could save thousands of lives and millions of dollars. This study investigates the possibility of achieving this goal using promising deep learning tools. Specifically, two deep neural network models (a multilayer perceptron neural network model and a long short term memory networks model) were developed to identify three secondary tasks: cellphone calling, cellphone texting, and conversation with adjacent passengers. The Second Strategic Highway Research Program Naturalistic Driving Study (SHRP 2 NDS) time series data, collected using vehicle sensor technology, was used to train and test the model. The results show excellent performance for the developed models, with a slight improvement for the LSTMN model, with overall classification accuracies ranging between 95 and 96%. Specifically, the models are able to identify the different types of secondary tasks with high accuracies of 100% for calling, 96 to 97% for texting, 90 to 91% for conversation, and 95 to 96% for the normal driving. Based on this performance, the developed models improve on the results of a previous model developed by Osman et al. ( 1 ) to classify the same three secondary tasks, which had an accuracy of 82%. The model is promising for use in in-vehicle driving assistance technology to report engagement in unlawful tasks and/or take over control in level 1 and 2 automated vehicles.
Different Level Automation Technology Acceptance: Older Adult Drivers
Sanaz Motamedi, University of FloridaShow Abstract
Alaa Masrahi, University of Rhode Island
Tobias Bopp, University of Rhode Island
Jyh-Hone Wang, University of Rhode Island
The increase in the number of older adult drivers in developed countries has raised safety concerns due to the decline in their sensory, motor, perceptual, and cognitive abilities which can limit their driving capabilities. Their driving safety could be enhanced by the use of modern Automated Driver Assistance Systems (ADASs) and might totally resolved by full driving automation. However, the acceptance of these technologies by older adult drivers is not yet well understood. Thus, this study investigated older adult drivers’ intention to use six ADASs and full driving automation through two questionnaires. A four-dimensional model referred to as the USEA model was used for exploring older adult drivers’ technology acceptance. The USEA model included perceived usefulness, perceived safety, perceived ease of use, and perceived anxiety. Path Analysis was applied to evaluate the proposed model. The results of this study identified the important factors in older adult drivers’ intention to use ADASs and full driving automation, which could assist stakeholders in improving technologies for use by older drivers.
Connected Vehicle Real-Time Traveler Information Messages for Freeway Speed Harmonization Under Adverse Weather Condition: Trajectory-Level Analysis Using Driving Simulator
Guangchuan Yang, North Carolina State UniversityShow Abstract
Mohamed Ahmed, University of Wyoming
Sherif Gaweesh, University of Wyoming
Eric Adomah, University of Wyoming
This paper employed a high-fidelity driving simulator to investigate the impacts of the Wyoming Department of Transportation (WYDOT) Connected Vehicle (CV) Pilot’s Traveler Information Messages (TIMs) on drivers’ speed management behavior, and the safety benefits of their speed harmonization. Three driving simulator experiment scenarios were developed to simulate the typical traffic and weather conditions on rural Interstate 80 (I-80) in Wyoming. A total of 25 professional truck drivers from the WYDOT and trucking industry were recruited to participate in the driving simulator experiments. Participants’ instantaneous speeds at various locations were collected to reveal the effects of CV TIMs on their speed management behavior. Simulation results showed that average speed profiles under CV scenarios were generally lower than under baseline scenarios, particularly under winter snowy or severe weather conditions. In addition, the variance of speed under CV scenarios were found to be significantly lower than the baseline scenarios, indicating that CV TIMs have the potential to harmonize the variations in speed. These findings suggest that CV TIMs can bring potential safety benefits, since reductions in average speeds and speed variances are considered as effective surrogate measures of safety, which are associated with lower risk of crashes. Research findings would also provide the WYDOT with early insights into the effectiveness of CV TIMs, which could assist the WYDOT with developing more efficient transportation management strategies under adverse weather conditions.
DISCLAIMER: All information shared in the TRB Annual Meeting Online Program is subject to change without notice. Changes, if necessary, will be updated in the Online Program and this page is the final authority on schedule information.