Here’re the general information for presentations and poster sessions of our group:
Event Type: Workshop
Event Number: 1007
Event Date: Sunday, January 08 09:00 AM – 12:00 PM ET
Prepared for Abrupt Emergence of Vulnerable Road Users at Geofencing Crosswalk
Traffic Control Device Student Challenge 2023: Innovative Traffic Control Devices to Improve Vulnerable Road User Safety Now in its 6th year of competition, the objective of the Traffic Control Device (TCD) Student Challenge is to promote innovation and stimulate ideas in the traffic control devices area with a goal to improve operations and safety. The challenge is sponsored by and conducted cooperatively by the Transportation Research Board (TRB) Standing Committee on Traffic Control Devices (ACP55) and the American Traffic Safety Services Association (ATSSA)
Event Type: Poster Session
Event Number: 2098
Event Date: Monday, January 09 10:15 AM – 12:00 PM ET
Snowflake Schema Based Data Warehouse for Analyzing Crash, Citation, and Warning Traffic Safety Records
Sicheng Fu, Steven Parker, and Andrea Bill
Decision-makers in traffic agencies and police departments require a wide variety of high-quality data to support traffic safety problem identification, program implementation, and result evaluation. But analyzing such a vast amount of data in the traditional transactional database sometimes faces challenges. Because of the complex table structure, a great number of table joins, and constrained to a single application, performing the analytic queries in the database is difficult and the result only contains static and one-time lists. However, the data warehouse technique offers the benefits of large data storage and multidimensional analysis for any number of applications, which would be appropriate to utilize for traffic safety related execution and assessment. This study describes the design and implementation of a Wisconsin traffic safety data warehouse for analyzing traffic safety related crash, citation, and warning records to support the context of the State of Wisconsin’s Traffic Records Coordinating Committee (TRCC) led traffic safety records improvement program. The design of the data warehouse is determined by the selection of data source, description of the data flow architecture, and design of the data warehouse schema. The snowflake schema design has been adopted to integrate multidimensional data sources because it has the ability to address different traffic safety problems with rare changes. The suitability and validity of this data warehouse design would benefit traffic safety analysis research and interest to other state agencies.
Event Type: Poster Session
Event Number: 2228
Event Date: Monday, January 09 06:00 PM – 07:30 PM ET
Distributed Connected Automated Vehicles Control under Real-time Aggregated Macroscopic Car-following Behavior
Estimation based on Deep Reinforcement Learning
Haotian Shi, Danjue Chen, Nan Zheng, Xin Wang, Yang Zhou, and Bin Ran
This paper proposes an innovative distributed longitudinal control strategy for connected automated vehicles (CAVs) in the mixed traffic environment of CAV and human-driven vehicles (HDVs), incorporating high-dimensional platoon information.
For mixed traffic, the traditional CAV control method focuses on microscopic trajectory information, which may not be efficient in handling the HDV stochasticity (e.g., long reaction time; various driving styles) and mixed traffic heterogeneities. Different from traditional methods, our method, for the first time, characterizes consecutive HDVs as a whole (i.e., AHDV) to reduce the HDV stochasticity and utilize its macroscopic features to control the following CAVs. The new control strategy takes advantage of platoon information to anticipate the disturbances and traffic features induced downstream under mixed traffic scenarios and greatly outperforms the traditional methods. In particular, the control algorithm is based on deep reinforcement learning (DRL) to fulfill car-following control efficiency and further address the stochasticity for the aggregated car following behavior by embedding it in the training environment. To better utilize the macroscopic traffic features, a general platoon of mixed traffic are categorized as a CAV-HDVs-CAV pattern and described by corresponding DRL states. The macroscopic traffic flow properties are built upon the Newell car-following model to capture the characteristics of aggregated HDVs’ joint behaviors. Simulated experiments are conducted to validate our proposed strategy. The results demonstrate that the proposed control method has outstanding performances in terms of oscillation dampening, eco-driving, and generalization capability.
Event Type: Lectern Session
Event Number: 3024
Event Date: Tuesday, January 10 08:00 AM – 09:45 AM ET
Conceptual Development for a Generalized Tribal Crash Safety Dashboard
Tianyi Chen, Haotian Shi, Steven Parker, Glenn Vorhes, and David Noyce
Tribal lands have historically experienced higher crash rates and injury severities compared to other areas throughout the United States. Effective data-driven safety analysis methods are critical for tribal lands to allocate resources and develop tribal safety programs. To perform the feasible and reproducible tribal crash analysis, this study identifies the minimum data requirement and creates a generic tribal crash dashboard prototype. The dashboard effectively provides statistical performance measurement approaches, monitors the tribal safety trend regarding different indicators, and gets updated as real-time data flows in, which can be implemented by any state. Specifically, the dashboard capabilities are demonstrated concerning the Wisconsin tribal crash data. It is used to locate the high-risk tribal land in Wisconsin meanwhile analyzing and comparing Wisconsin statewide crashes and tribal crashes based on the crash severity. The findings illustrate that the roads in tribal lands are more dangerous than roads in other areas in Wisconsin, especially the rural areas. The crash types are then analyzed in the dashboard to investigate the reasons and insights behind the Wisconsin tribal crashes. This dashboard concept demonstration shows that it can facilitate a more intuitive understanding of tribal crashes from the statistical perspective, which helps leverage available data sources and better develop effective safety countermeasures.
Event Type: Poster Session
Event Number: 4063
Event Date: Wednesday, January 11 10:15 AM – 12:00 PM ET
Development of the Data Pipeline for a Connected Vehicle Corridor
Keshu Wu, Yang Cheng, Steven Parker, Bin Ran, and David Noyce
The connected and automated vehicle (CAV) technologies have shown great potential in safety, mobility, environmental, and social benefits. Various testbeds have been deployed around the world to investigate and develop CAV-related applications; data sharing between CAVs and infrastructures is one of the focus areas. The Traffic Operations and Safety (TOPS) Laboratory in the UW-Madison College of Engineering has partnered with the City of Madison and the Wisconsin Department of Transportation (WisDOT) to develop a Connected Corridor as a testbed for CAV and cooperative driving automation (CDA) technologies. This paper presents the development of a data pipeline system, which aims to retrieve, process, and archive the real time data from this Connected Corridor. The data pipeline system architecture adopts a virtual edge computing design, which aims to improve system reliability and scalability. The system modules, dataflows, and data visualization are also discussed. The data archived by the data pipeline system will be used for future CAV and CDA applications.