User:Tuttlec/sandbox
The Center for Connected and Automated Transportation (also known as CCAT) is the United States Department of Transportation Region 5 University Transportation Center. Founded under the Fixing America's Surface Transportation Act (or FAST Act), CCAT is the UTC for focused research on comprehensive transportation safety, congestion, connected vehicles, connected infrastructure, and autonomous vehicles. The Center is led by Dr. Henry Liu with consortium members including the University of Michigan Transportation Research Institute, the University of Akron, Central State University, the University of Illinois at Urbana Champaign, Purdue University, and Washtenaw Community College.
Contents
History[edit]
The Center for Connected and Automated Transportation received a $2,470,600 grant through the United States Department of Transportation's University Transportation Center program on December 1, 2016. The grant was one of 35, five-year grants awarded with partners including the University of Akron, Central State University, the University of Illinois at Urbana-Champaign, Purdue University, and Washtenaw Community College. In addition to conducting research, CCAT is focused on promoting education, workforce development, and technology transfer activities. In February of 2019, Robert Hampshire was interviewed by MLive on a CCAT-funded research project focused on increasing the speed of autonomous vehicle deployment utilizing artificial intelligence. In July of 2019, Henry Liu testified before Haley Stevens, Chairwoman of the House Committee on Science, Space, and Technology, Subcommittee on Research and Technology, to highlight the importance of the FAST Act in reducing traffic fatalities. In January of 2022, Henry Liu was appointed as Director of Mcity.
Areas of Research[edit]
CCAT research is focused in six main thrusts: Control & Operations, Enabling Technology, Human Factors, Infrastructure Design & Management, Modeling & Implementation, and Policy & Planning.
Control & Operations
Traffic control systems are a critical component of the transportation infrastructure. With the development of CAV technologies, traffic control in a CAV environment will be significantly different from the conventional framework. Despite substantial efforts in investing and developing CAV technologies in the past decade, large-scale CAV systems, have not yet been deployed. Over the next ten years, the CAV penetration rate is expected to remain at a low level. Therefore, optimizing traffic control systems for a low penetration rate is necessary. CCAT will design new traffic control models and algorithms that can be applied in a low CAV penetration rate environment for both off-line and real-time applications. If successful, this research would be available for implementation at once in the Ann Arbor Connected Vehicle Test Environment (AACVTE) and/or Mcity.
Enabling Technology
In early 2016, two separate events sparked excitement in the fields of autonomous vehicles and artificial intelligence. NVidia announced that their camera-based vehicle, driven by their Drive-PX computer, was trained to drive, using no prior knowledge and with just 72 hours of training video data; and AlphaGo, a computer program designed to play the game Go, defeated Lee Sedol, one of the world’s top-ranked players. These two events highlight the potential of artificial intelligence, as an alternative to the current focus on expensive sensors like Lidars. In both cases, the key enablers included rich training data sets, a learning algorithm, and a test platform.
Human Factors
Automated driving allows for at least some degree of vehicle control to be shifted from the driver to the vehicle. It changes not only the driving behavior but also the travel decision making process. The change of driving behavior will impact the design of the automated vehicle system that further impacts travelers’ adaption of CAV. In addition, the change of travel behavior may affect the system performance. The objectives of this research topic are to (i) leverage the capability of driving simulators at both UM and Purdue to investigate the factors that impact drivers’ decisions to take over driving tasks, (ii) optimize in-vehicle alarm systems to inform drivers to take over, (iii) investigate the psychological effect on users’ acceptance of CAVs using driving simulator and biosensors, and (iv) understand route choice behavior of CAV users. The research will be conducted by applying analytical models (psychological models, driver behavior models) to driving simulator experiments integrated with biosensors. The research results are expected to foster safe driving of CAV and increases users’ acceptance of CAV technology.
Infrastructure Design & Management
The advent of CAVs promises enhanced safety and mobility of highway operations. However, the impacts of CAVs on highway infrastructure have not been fully studied. As such, highway asset managers are generally not adequately prepared to make the needed investments in the physical infrastructure to accommodate these new technologies. Therefore, CCAT aims to examine the impact of driverless vehicles on the design and management of highway infrastructure. The research approach will be to identify the various dimensions of the impact of CAVs on highway infrastructure with reference to the pre-CAV and post-CAV eras. The first dimension is the physical dimensions of highway in the infrastructure (the dimensions of certain assets will need to be increased, others decreased). The second dimension is the obsolescence of certain existing infrastructure (and their subsequent removal from the asset inventory). The third dimension is the introduction of new types of assets. The fourth dimension is the effect of increased or decreased amounts of highway travel on (a) physical deterioration of the infrastructure (b) user costs associated with highway investments and their effect on economic analysis and (c) highway revenues whether through indirect charging (the fuel tax) or direct charging (VMT fees). In addressing these issues, the proposed project is expected to yield a wide range of benefits to the departments of transportation at every level of government. Agencies will be better informed to make or prioritize investments to prepare for CAVs and will be able to make more reliable assessments of the impacts of CAVs operations on highway infrastructure expenditures and user revenues.
Modeling & Implementation
As CAV technologies provide new paradigms for the future transportation system, there is an urgent need for models that can predict traffic flow under different levels of vehicle connectivity and automation. Hence, CCAT’s modeling and implementation cross-cutting thrust area is motivated by two key aspects. First, consistent with a future transportation system characterized by CAV technologies, there is the need for models that can evaluate the performance and characteristics of CAV-based transportation systems. Second, CCAT’s unparalleled resources such as Mcity and Ann Arbor Connected Vehicle Test Environment (AACVTE) provide the team immediate opportunities to test models in a real-world environment. Hence, the cross-cutting activities related to model development and field implementation, provide us unique opportunities to implement and test models in the field, and use the field data to refine, calibrate and validate models. This enables the development of models that can test “what-if” scenarios of capabilities before they are deployed in the real-world by transportation agencies and the auto industry. The modeling research will focus on bringing the automation and connectivity capabilities for the evaluation of transportation system safety and performance in connected environments, and with different levels of automation.
Policy & Planning
In the era of post-autonomy when automated vehicles are prevalent, there will be revolutionary changes to the transportation system. From auto ownership to auto usage, from trip planning to road traffic management, even from the perspective of law enforcement, almost every aspect of an autonomous transportation system will differ from the existing one. To prepare for these changes, it is desirable to study a variety of essential policy and technical issues an autonomous vehicle system may face, including automated vehicle adoption pathway, logistic operations of fleets, social acceptance, consumer choice and business operation, as well as sustainability assessment. Although an automated future is defined as desirable, simply saying so does not map a route from here to there. The lack of viable deployment pathways was a huge impediment to past automation efforts and many obstacles remain despite improvements in technology. Shared use of automated vehicles is one promising form of automated vehicle fleet adoption. However, even within the narrower field of shared automated vehicles (SAVs), all academic work to date assumes relatively mature systems. CCAT researchers will seek to fill this gap by identifying a pathway for sharing to emerge in an automated vehicle context, providing practical and actionable knowledge to planners, governments, and private companies on what types of SAV systems can effectively be deployed and where.
Principal Investigators[edit]
Principal Investigators under the CCAT program are as follows:
- University of Michigan
- Shorya Awtar
- Shan Bao
- Debby Bezzina
- Tierra Bills
- Robin Brewer
- David W. Eby
- Nicole Ellison
- Fred Feng
- Shuo Feng
- Carol Flannagan
- Vineet Kamat
- Sridhar Lakshmanan
- David LeBlanc
- Brian Lin
- Henry Liu
- Z. Morley Mao
- Bernard Martin
- Neda Masoud
- Carol Menassa
- Lisa Molnar
- Gabor Orosz
- Huei Peng
- Anuj K. Pradhan
- Matt Reed
- Paul Richardson
- James Sayer
- Siqian Shen
- Yafeng Yin
- The University of Akron
- Ethan Shajie
- Yilmaz Sozer
- Ping Yi
- Central State University
- Ramanitharan Kandiah
- Krishnakumar Nedunuri
- University of Illinois at Urbana-Champaign
- Imad Al-Qadi
- Hadi Meidani
- Yanfeng Ouyang
- Hasan Ozer
- Jeffery Roesler
- Purdue University
- Darcy Bullock
- Sikai Chen
- Daniel DeLaurentis
- James Eric Dietz
- Yiheng Feng
- Jon D. Fricker
- Konstantina "Nadia" Gkritza
- James Krogmeier
- Samuel Labi
- Yongfu Li
- Mohammad Miralinaghi
- Shaoshuai Mou
- Christos Mousas
- Srinivas Peeta
- Brandon Pitts
- Kumares Sinha
- Dengfeng Sun
- Shreyas Sundaram
Funded Research[edit]
The following are research projects funded by CCAT since its inception in 2017.
University of Michigan
- 2021 International Symposium on Transportation Data and Modelling (ISTDM)
- A Data-Driven Autonomous Driving System for Overtaking Bicyclists
- A Naturalistic Bicycling Study in the Ann Arbor Area
- Accelerated Training for Connected and Automated Vehicles Based on Adaptive Evaluation Method
- Adapting Land Use and Infrastructure for Automated Driving
- AI-enabled Transportation Network Analysis, Planning, and Operations
- Anomaly/Intrusion Detection: Design, Development and Testing
- Augmented Reality Testing Environment Development – Phase II
- Automotive Cybersecurity Industry Consortium CAN Bus Scanning Tool Project
- Autonomy in Transportation Education
- AV Occupant ID Optical Based Occupant Identification and Classification for Autonomous Vehicles
- Big Data Drive Deployment in a Connected Vehicle Environment
- CAV-Based Intersection Maneuver Assist Systems (CAVIMAS) and Their Impact on Driver Behavior, Acceptance, and Safety
- Conformance to Clarifications for Consistent Implementations (CCIs) to Ensure Interoperability of Connected Signalized Intersections in the Ann Arbor Connected Environment with a National Deployment
- Connected and Automated Vehicle (CAV) Data Infrastructure and Access
- Connected and Automated Vehicle (CAV) Testing Scenario Design and Implementation using Naturalistic Driving Data and Augmented Reality
- Cybersecurity of Transportation Infrastructure in a Connected-Vehicle Environment
- DeepScenario: City Scale Scenario Generation for Automated Driving System Testing & Evaluation
- Deployment of Preemption based Motion Sickness Prevention Technology on a Testbed Vehicle in Mcity
- Design and Operation of Efficient and Budget-Balanced Shared-Use Mobility Systems
- Developing Decision-Making Models for AV Movements at the Unsignalized Intersections
- Development of An Augmented Reality Environment for Connected and Automated Vehicle Testing
- Development of an Integrated Augmented Reality Testing Environment and Implementation at the American Center for Mobility (ACM)
- Development of Machine-Learning Models for Autonomous Vehicle Decisions on Weaving Sections of Freeway Ramps
- Driving Etiquette
- Enhancing Network Equilibrium Models for Capturing Emerging Shared-Use Mobility Services
- Expand Deployment of Applications for Vulnerable Road Users in Pillar 1 (Pedestrian in Crosswalk)
- Guidelines for Development of Evidence-Based Countermeasures for Risky Driving
- How Vehicle Connectivity Based Eco-Routing Choices Will Impact Driver Decision Making
- Improving the Efficiency of Trucks via CV2X Connectivity on Highways
- Investigation Into U.S. Real World Lane Change Behavior for Automated Freeway Driving
- Leveraging Connected and Automated Vehicles for Participatory Traffic Control
- Machine Learning, Human Factors and Security Analysis for the Remote Command of Driving: An Mcity Pilot
- Mcity Ann Arbor Connected Environment Operations and Maintenance
- Mcity Infrastructure Data-Collection and Management System Development
- Modeling Naturalistic Driving Environment with High-Resolution Trajectory Data
- Motion Sickness Alleviation via Anticipatory Control of Active Seats in Autonomous Vehicles
- On Transportation Equity Implications of Connected and Autonomous Vehicles (CAV): A Review of Methodologies
- Predicting Driver Takeover Performance in Conditional Automation (Level 3) through Physiological Sensing
- Promoting Inclusive Design and Deployment of Connected and Automated Vehicles for Older Adults Through Education and Training of Engineering Students and Older Drivers
- Real‐Time Distributed Optimization of Traffic Signal Timing
- Reliable V2V Communication Networks: Applications in Fuel-Efficient Platooning
- Road-side Based Cybersecurity in Connected and Automated Vehicle Systems
- Supporting People with Vision Impairments in Automated Vehicles
- Targeted Real-World Heat Map
- Trajectory Based Traffic Control with Low Penetration of Connected and Automated Vehicles
University of Akron
- Access Control at Major-Minor Intersection through CAV in Mixed Traffic
- Development of a Prototype Safety Advisory System to Aid Older Drivers in Gap Selection
- Impact Analysis of Roadway Surface and Vehicle Conditions on Fleet Formation for Connected and Automated Vehicles
- Promoting Inclusive Design and Deployment of Connected and Automated Vehicles for Older Adults Through Education and Training of Engineering Students and Older Drivers
Central State University
- CAV Developed Vehicles as Real-Time Sensors for Assessing Greenhouse Gases
- CAV Systems Incorporating Air Pollution Information from Traffic Congestion
University of Illinois at Urbana-Champaign
- Impact of Autonomous Freight Delivery on Trucking Operations
- Multifront Approach for Improving Navigation of Autonomous and Connected Trucks
- Operations of Connected and Autonomous Freight Trucks under Congestion and Infrastructure Cost Considerations
Purdue University
- A Virtual Reality Framework to Measure Psychological and Physiological Responses of the Self-Driving Car Passengers
- Adapting Land Use and Infrastructure for Automated Driving
- Behavioral Intention to Ride AVs and Impacts on Mode Choice Decisions, Energy Use, and Emissions
- Changes in Highway Agency Expenditures and Revenue in an Era of CAVs
- Cooperative Control Mechanism for Platoon Formation of Connected and Autonomous Vehicles
- Design and Management of Highway Infrastructure to Accommodate CAVs
- Design of Urban Landscape and Road Networks to Accommodate CAVs
- Develop In-Vehicle Information Dissemination Mechanisms to Reduce Cognitive Burden in the Information-Rich Driving Environment
- Development of A Cooperative Perception System
- Development of AI-Based and Control-Based Systems for Safe and Efficient Operations of Connected and Autonomous Vehicles
- Development of Dynamic Network Traffic Simulator for Mixed Traffic Flow Under Connected and Autonomous Vehicles
- Development of Situational Awareness Enhancing Systems for AV-Manual Handover and Other Tasks
- Economical Acquisition of Intersection Data to Facilitate CAV Operations
- Enhanced Methodology for Exploring Autonomy-Enabled Multi-Mode Regional Transportation
- Exploring the Prospective Role of Connected Vehicles in Monitoring and Response to Pandemics and Disasters
- Facilitating Electric Propulsion of Autonomous Vehicles Through Efficient Design of a Charging-Station Network
- Impacts of In-Vehicle Alert Systems on Situational Awareness and Driving Performance in SAE Level 3 Vehicle Automation
- Intelligent Sidewalk De-Icing and Pre-Treatment with Connected Campus Maintenance Vehicles
- Investigation of AV Operational Issues Using Driving Simulator Equipment
- Lane Management in the Era of CAV Deployment
- Large Network Multi-Level Control for CAV and Smart Infrastructure: AI-based Fog-Cloud Collaboration
- Leveraging Control Theory to Facilitate UAV Application for CAV Deployment
- Non-Connected Vehicle Detection Using Connected Vehicles
- Pedestrian-Vehicle Interaction in a CAV Environment – Explanatory Metrics
- Promoting CAV Deployment by Enhancing the Perception Phase of the Autonomous Driving Using Explainable AI
- Public Acceptance and Socio-Economic Analysis of Shared Autonomous Vehicles: Implications for Policy and Planning
- Ridesharing, Active Travel Behavior, and Personal Health: Implications for Shared Autonomous Vehicles
- Ride-sharing with Advanced Air Mobility
- Smart Interaction – Pedestrians and Vehicles in a CAV Environment
- The Impact of COVID-19 on User Perceptions of Public Transit, Shared Mobility/Micro-Mobility Services, and Emerging Vehicle Types
- Translation of Driver-Pedestrian Behavioral Models at Semi-Controlled Crosswalks into a Quantitative Framework for Practical Self-Driving Vehicle Applications
- Using Driving Simulator Environment to Determine Interactions Between User Behavior and Infrastructure Design Under Autonomous Vehicles
- Using Virtual Reality Techniques to Investigating Interactions Between Fully Autonomous Vehicles and Vulnerable Road Users
Notable Projects[edit]
Development of an Integrated Augmented Reality Testing Environment and Implementation at the American Center for Mobility (ACM)[edit]
In this project, Henry Liu developed an integrated solution for autonomous vehicle testing, in which a naturalistic adversarial driving environment (NADE) was integrated with an augmented reality (AR) testing system. The integrated solution was implemented at the American Center for Mobility (ACM). With the AR techniques, the real-world testing AVs can be tested at a test track and interact with virtual background vehicles. With the NADE, the maneuvers of the virtual background vehicles will be generated purposely, in that most of the maneuvers are generated from naturalistic driving data, and only at selected moments, selected vehicles execute adversarial moves to challenge the AVs. The project aims to drastically reduce the number of vehicle miles driven needed to prove the worthiness of a system for real roadways. In February of 2021, this research was published in Nature Communications with authors including Shuo Feng, Xintao Yan, Haowei Sun, Yiheng Feng & Henry Liu.
Behavioral Intention to Ride in an Autonomous Vehicle and Implications on Mode Choice Decisions, Energy Use, and Emissions[edit]
The objectives of this project are threefold using the results of a stated preference survey: (i) assess the behavioral intention to ride in AVs, (ii) investigate the effect of the emergence of shared AVs on mode choice decisions in the short and long run and the corresponding effect on value of travel time savings (VTTS), and (iii) assess the energy and environmental implications due to the emergence of AVs offering single-passenger rides. In November of 2021, this research project was published in Transportation.
