Principal Investigator

Sean Qian

Status

Completed

Start Date

Jan. 1, 2015

End Date

Dec. 31, 2015

Project Type

Research Advanced

Grant Program

MAP-21 TSET - Tier 1 (2012 - 2016)

Grant Cycle

2015 TSET UTC

Visibility

Public

Abstract

Transportation infrastructure is quickly moving towards revolutionary changes to accommodate the deployment of AV. On the other hand, the transition to new vehicle technologies will be shaped in large part by changes in performance of roadway infrastructure. This research aims at understanding the relationship between AV technology and infrastructure performance, which leads to fundamentals of future infrastructure design.

We attempt to tackle two fundamental questions, 
1)	How would vehicle automation/communication, with different sensing and control specifications, change the vehicle speed and headway under various traffic conditions, and therefore change traffic congestion and crash patterns in the network? 
2)	How would the vehicular technology change the flow capacity of the roadway infrastructure networks, under different crash rates that are expected to be achieved by different vehicular control strategies?  How does the change vary at different levels of AV penetration rates? 

             As a seed project, this research primarily addresses the mobility concerns of AVs. We propose to study potential car-following behavior of AV, which results in a new fundamental diagram of vehicle density, speed and volume. We develop a novel two-class traffic flow model where both AVs and conventional vehicles are mixed. We examine four types of network topology, lane drop, merge junctions, diverge junctions and non-stop intersections. The expected outcome of this research is to lay out a framework of traffic flow modeling in the presence of AVs, and ultimately to allow flexible extensions of various vehicular control specifications for systematic assessment of mobility and safety. Furthermore, the new flow models will be incorporated into an open-source network simulation toolkit, TNM (Transportation Network Modeling) co-developed by PI Qian, to simulate traffic evolution in large-scale roadway networks.  Insights on AV impact to the network flow capacity and congestion patterns will be provided for policy indications and infrastructure design.    

Description

Transportation infrastructure is quickly moving towards revolutionary changes to accommodate the deployment of AVs. On the other hand, the transition to new vehicle technologies will be shaped in large part by changes in performance of roadway infrastructure. This research aims at understanding the relationship between AV technology and infrastructure performance, which leads to revolutionary change in transportation infrastructure design in the both short and long term. 

For nearly a century, traffic flow in the roadway networks is operated purely by human beings. Human’s reactions to preceding vehicles and side vehicles almost dominate the driving behavior, which is to be replaced by vehicle automation/communication in the future. The technology for autonomous and connected vehicles is rapidly approaching the point of commercial implementation. AVs are cars that can be fully controlled by computers, instead of people, relying upon on-board advanced sensors and computers to observe and interpret road conditions and determine a safe course of action. Connected vehicles, on the other hand, receive data from other vehicles, or a central system, that then instructs them how to operate safely. Generally there is a significant developmental overlap between the two, with future autonomous connected vehicles able to receive data from itself, other cars and systems, and capable of driving themselves or accepting control from external systems. With an assigned time and path, these lightweight, self-guided cars would proceed steadily through crowded infrastructure without all the stop-and-go that chokes roadways and saps fuel efficiency. Many of the enabling technologies, such as adaptive cruise control and lane departure warning systems, already exist. We envision that the pathways of AV and connected vehicle development are likely to converge in the long run. This research uses ‘autonomous vehicle (AV)’ to represent ‘autonomous connected vehicle’.

To assess the vehicular technology impact to the traffic flow, two of the most important questions we attempt to tackle in this research are, 
1)	How would vehicle automation/communication, with different sensing and control specifications, change the vehicle speed and headway under various traffic conditions, and therefore change traffic congestion and crash patterns in the network? 
2)	How would the vehicular technology change the flow capacity of the roadway infrastructure network, under different crash rates that are expected to be achieved by different vehicular control strategies?  How does the change vary at different levels of AV penetration rates? 

This project primarily addresses the mobility concerns of AVs, while establishing a modeling framework that allows future extensions to assess both mobility and safety. In particular, this research helps determine the impact of vehicle automation on the effective road capacity and operating efficiency of transportation networks. It also provides insights for design of the vehicle control strategies targeting mobility and safety. Research will be devoted to address knowledge gaps related to the operations of automated vehicles and the existing road infrastructure, and the policy implications for transportation planning, system design, and the economy.

While there is already a great deal of study on improving existing roads’ capacities and speeds using autonomous cars, two of the ideas that promise the greatest improvements are autonomous platoons and managed non-stop intersections. In autonomous platoons, or Cooperative Adaptive Cruise Control groups, vehicles communicate with each other to enable them to travel closer together and, possibly, at higher speeds than would otherwise be safe. In managed intersections the vehicles communicate their locations and planned paths with each other or a central system to allow themselves to pass through an intersection without the traditional stop-go light cycle. However, both of the ideas lack modeling support on how they would influence roadway flow capacity and thus time-of-day traffic flow evolutions. Therefore, in order to utilize such concepts to improve the performance of existing road networks, explicit modeling of traffic flow mixed with both conventional vehicles and AVs is required. 

The research approach is as follows. First we will review state-of-the-art AV control specifications, such as expected crash rate, reaction time, headway and lane-changing rules. The new driving behavior will be used to develop new traffic flow models, i.e. the fundamental relationships between flow rate, speed and density, through microscopic car-following models. It will then be compared to the classical traffic flow theory based upon conventional vehicles. 

Second, we consider the heterogeneity of traffic flow where AVs and conventional vehicles are mixed, and develop a two-class traffic flow model. One key idea of modeling heterogeneous flow is that, rather than an aggregated flow-density relation of mixed flow as treated in the well-known LWR model, we approximate the mixed flow by explicitly modeling the interactions of two classes of traffic streams. Each class possesses identical vehicle attributes and car-following rules, which are encapsulated by a unique well-defined (least requirements usually include continuity and concavity) fundamental diagram. Within this general framework, we develop a generic, yet simple, class-specific capacity allocation and flux scheme to capture inter-class flow interactions. Some preliminary work has been done, and a relevant paper has been accepted for presentation in the TRB annual meeting 2015. 

Third, we will develop a dynamic network loading model incorporating the two-class traffic flow. Four types of network topology will be examined, lane drop, merge junctions, diverge junctions and a general urban intersection. Mixed traffic flow in the first three types will follow merging and diverging rules of AVs, while the general intersection requires intensive modeling and simulation work to enable non-stop settings.  The flow capacity changes in all those topology types will be reported with respect to the penetration rate of AVs. 

Fourth, we examine our dynamic network loading model in three test networks, a grid network, the Sioux Falls network, and a large-scale Philadelphia Metropolitan Network. The new traffic flow models serve as the fundamental theory to derive system performance of the existing roadway infrastructure under AVs. Furthermore, the new flow models will be incorporated into an open-source network simulation toolkit, TNM (Transportation Network Modeling) co-developed by PI Qian, to simulate traffic evolution in the roadway networks.  The output will provide the spatio-temporal distributions of both automatous vehicles and conventional vehicles, depending on the penetration rates. Insights on AV impact to the network flow capacity and congestion patterns will be provided for policy indications and infrastructure design.

The expected outcome of this research is to lay out a framework of traffic flow modeling in the presence of AVs, and ultimately to allow flexible extensions of various vehicular control specifications for systematic assessment of mobility and safety. The optimal vehicle control strategies and the resultant system performance are oftentimes a compromise of mobility and safety. The modeling framework can lead to systematic assessment of AV technologies.

The funding from this seed grant will be used to support a PhD student for initial establishment of research. This will allow the student to create a preliminary framework to develop traffic flow models that considers AVs. The student will also study and develop computationally efficient network loading model applicable to large-scale networks. The funding will also support the PIs and the student to attend a conference, which will enable them to disseminate the results and interact with experts and potential collaborators/funding sources in this research field.   

We plan to actively seek both industrial and federal funding based on the initial development. While we focus on particular car-following strategies for AVs to demonstrate the generality of the modeling framework and network simulation tools, other vehicular control strategies are applicable in this framework, which can be scalable to real-world networks of cities or regions. This generality will attract the attentions from various agencies and private firms. Potential funding agencies/collaborators include the Department of Transportation, Federal Highway Administration, National Science Foundation, National Institute of Standards and Technology, and vehicle manufacturers. The proposed research is closely related to on-going AV research at Carnegie Mellon. Interactions with those groups will have synergistic effects. 

Timeline

1. Review of AV control specifications: Jan-Feb
2. Two-class traffic flow model: Feb-May
3. Dynamic network loading model: May-Sept
4. Test on three networks: Sept-Dec

Strategic Description / RD&T

Deployment Plan

We will run the new network loading models on the large-scale Philadelphia Metropolitan Network to gain insights on AV impact to the network flow capacity and congestion patters with respect to the AV penetration rates. Certain assumptions on travel demand and travelers’ choices will be made to simplify the computation. We will present the findings and results to DVRPC and PennDOT to gain their comments and help their decision making. 

We plan to actively seek both industrial and federal funding based on the initial development. While we focus on particular car-following strategies for AVs to demonstrate the generality of the modeling framework and network simulation tools, other vehicular control strategies are applicable in this framework, which can be scalable to real-world networks of cities or regions. This generality will attract the attentions from various agencies and private firms. Potential funding agencies/collaborators include the Department of Transportation, Federal Highway Administration, National Science Foundation, National Institute of Standards and Technology, and vehicle manufacturers. The proposed research is closely related to on-going AV research at Carnegie Mellon. Interactions with those groups will have synergistic effects.

Expected Outcomes/Impacts

The expected outcome is a modeling framework to model AV traffic flow that can be potentially extended in the future with general vehicular control strategies. The traffic flow model developed in this research will be implemented into an open-source network simulation toolkit, TNM (Transportation Network Modeling) co-developed by PI Qian, to simulate mixed AV and conventional vehicles’ evolution in the roadway networks.  The output will provide the spatio-temporal distributions of both AVs and conventional vehicles, depending on the penetration rates. Insights on AV impact to the network flow capacity and congestion patterns will be provided for policy 

Expected Outputs

TRID

Individuals Involved

Email	Name	Affiliation	Role	Position
cth@cmu.edu	Hendrickson, Chris	CIT/Heinz	Co-PI	Faculty - Tenured
weima@cmu.edu	Ma, Wei	CEE	Other	Student - PhD
seanqian@cmu.edu	Qian, Sean	Carnegie Mellon University	PI	Faculty - Research/Systems

Budget

Amount of UTC Funds Awarded

$46299.00

Total Project Budget (from all funding sources)

$46299.00

Documents

Type	Name	Uploaded
Data Management Plan	Qian_Autonomous_Vehicle_Proposal_sGkkZLD.doc	Oct. 13, 2017, 7:57 a.m.
Final Report	What_do_autonomous_vehicles_mean_to_traffic_congestion_and_crash.pdf	March 21, 2018, 8:08 a.m.
Publication	Traffic simulation and transportation engineering.	Nov. 27, 2020, 6:47 p.m.
Publication	Some Thoughts on the Future of Transportation Engineering.	Nov. 27, 2020, 6:48 p.m.
Publication	Exploring the economic, environmental, and travel implications of changes in parking choices due to driverless vehicles: An agent-based simulation approach.	Nov. 27, 2020, 6:49 p.m.
Publication	Examining driver injury severity in intersection-related crashes using cluster analysis and hierarchical Bayesian models.	Dec. 2, 2020, 9:07 a.m.
Publication	Lane Management with Variable Lane Width and Model Calibration for Connected Automated Vehicles.	Dec. 2, 2020, 9:34 a.m.

Match Sources

No match sources!

Partners

No partners!