Research

Piloting Augmented Reality Cues for Enhancing Driving Safety in Glaucoma

Time: 2022 – 2023

Sponsor: NIH (CTSA)

Description:

This project aims to improve driving safety and performance in older drivers with peripheral vision loss due to glaucoma. Leveraging the ACT Lab’s high-fidelity driving simulator, the potential benefits of incorporating visual/auditory Augmented Reality (AR) cues within the in-vehicle environment will be tested. This investigation will specifically focus on driving performance and mental workload to avoid introducing distraction to the driving task. The experiments will assist in optimizing the modality and quantity of the cues provided to attain a balance between improving driving safety and increasing distraction in an already cluttered driving environment.

Piloting Augmented Reality Cues for Enhancing Driving Safety in Glaucoma

Time: 2022 – 2023

Sponsor: NIH (CTSA)

Description:

This project aims to improve driving safety and performance in older drivers with peripheral vision loss due to glaucoma. Leveraging the ACT Lab’s high-fidelity driving simulator, the potential benefits of incorporating visual/auditory Augmented Reality (AR) cues within the in-vehicle environment will be tested. This investigation will specifically focus on driving performance and mental workload to avoid introducing distraction to the driving task. The experiments will assist in optimizing the modality and quantity of the cues provided to attain a balance between improving driving safety and increasing distraction in an already cluttered driving environment.

SCC-IRG Track 1: Fostering Smart and Sustainable Travel through Engaged Communities using Integrated Multidimensional Information-Based Solutions

Time: 2021 – 2025

Sponsor: NSF

Description:

This project aims to develop systematic deployment tools that smart and connected communities can use to achieve their sustainable travel goals in a quantifiable manner by leveraging advances in information, communication, and sensor technologies. While the deployment of advanced technological solutions offers great promise for communities to improve residents’ quality of life and prosperity, they are faced with significant challenges in realizing these aspirations due to the diversity in technological and travel needs and barriers faced. Solutions to enhance travel mobility, safety, equity, and access will be developed using the City of Peachtree Corners (GA) as a living lab. Solutions will include building partnerships of emerging micromobility services, behavioral interventions, and public policy interventions. These solutions will be developed using data collected from community residents and other sources, and deployed using an information design system that provides targeted information delivery to various stakeholders. This project draws on methods from multi-objective and multi-agent optimization, machine learning, behavioral economics, and data and policy analytics, to generate multidimensional solutions for the community.

Incorporating Situational Awareness Cues in Virtual Reality (VR) Environments for Users of Autonomous Vehicles

Collaborators: Bruce Walker and Nadia Fereydooni (School of Psychology; School of Interactive Computing)

Time: 2021

Description:

Technological advances are attracting an increasing number of individuals to use VR in vehicles. However, users report concerns of losing track of real-world occurrences. Given that VR devices completely block the road environment from the user, this study aims to develop a compensating mechanism that will allow passengers to stay aware of their real-world surroundings while engaging in the VR environment. Situational awareness cues are embedded within the VR environment of one group of participants, but not the other. Using validated questionnaires such as SART (Situational Awareness Rating Technique) and NASA TLX (Task Load Index), participants’ informational needs are examined, along with their perceived risk, perceived comfort (physical and psychological), task workload and trust in technology.

Driving Simulator-Based Study of the Impacts of Various Roadway Design Modifications on the Curiosity Lab Test Track

Time: 2020 – 2021

Sponsor: City of Peachtree Corners, GA

Description:

This study aims to understand the role of urban infrastructure modifications on the performance of Autonomous Vehicles (AVs) in mixed traffic environments, and what infrastructure designs can enhance transportation safety in smart-city environments by developing a digital twin of Curiosity Lab in a high-fidelity driving simulator. Data from simulator-based experiments on traffic conditions and driver behavior/response under different infrastructure modifications (e.g., dedicated AV lane, lane width, barriers, and markings) will be collected and analyzed to develop implementation guidance to the Curiosity Lab for potential future roadway modifications to enhance consumer acceptance of AVs and their adoption in typical urban environments. The study aims to develop a virtual AV test track of Curiosity Lab with varied roadway characteristics, investigate the impacts of roadway modifications on human drivers using driving simulator-based experiments and develop design guidelines for AV testing and deployment. 

Simulation of Competitive Ride-Hail Services

Time: 2020 – 2022

Sponsor: USDOT (T-SCORE)

Description:

This study will develop a simulation model capable of representing tradeoffs between independent ride-hailing operators and coordinated demand-responsive transit. The proposed simulation model aims to capture the intricate relationship between supply for and demand of on-demand transit and independent ride-hail services. Based on the inputs of ride-hail requests, the model can not only predict the rides and routes for ride-hailing operators and demand-responsive transit, but also estimate wait time and cost for travelers. The results will be implemented in the open-source MATSim simulator to help researchers and transit operators evaluate transit strategies to achieve maximum complementarity across modes.

Using Driving Simulator Environment to Determine Interactions between User Behavior and Infrastructure Design Under Autonomous Vehicles

Time: 2019 – 2021

Sponsor: USDOT (CCAT)

Description:

This study will use immersive driving simulator environments in conjunction with stated preference surveys to analyze the interactions between user behavior and infrastructure design changes under AVs while also factoring emerging trends of share-mobility services, electrification, and the promotion of sustainable transportation modes (such as mass transit, biking, and walking). A road corridor in Atlanta, GA with current road infrastructure design and human-driven vehicle environment will provide the base case for the driving simulator environment. Further modifications are made to the base case driving simulator environment to include AVs infrastructure design, shared-mobility services, electrification, and sustainable transportation modes. The study insights will be used to develop guidelines that can aid state and local transportation agencies to develop near- and medium-term infrastructure design modifications to enable efficient, smooth transition to an AV future that additional factors other emerging transportation-related trends.

Smartphone-Based Incentive Framework for Dynamic Network-Level Traffic Congestion Management

Time: 2019 – 2021

Sponsor: USDOT (STRIDE)

Description:

By leveraging advances in smartphone-based personalization, big data availability for traffic, and network-level integration through information-based connectivity, this study will develop a smartphone-based framework to develop real-time incentives (monetary, value-based, travel-related credits, etc.) to influence drivers’ en route routing decisions to manage network-level system performance in congested dynamic traffic networks. Analytical models and algorithms will be developed to identify and implement the specific incentives to be deployed in real-time, using techniques from game theory, optimization, and machine learning. This study will ensure that the proposed incentives are behavior-consistent and tailored based on travelers’ smartphone response. The study will also explore how public and private sector transportation entities can collaborate through the use of new forms of incentives that leverage the emerging transportation future, in addition to the currently-used ones.

CPS: Medium: Emulating Emerging Autonomous Vehicle Technologies

Time: 2019 – 2022

Sponsor: NSF

Description:

The study aims to develop an analytical and numerical framework to emulate the impacts of current automated vehicle (AV) technologies on transportation networks in the near future. Empirical data from Level 2/3 AVs on the market will be collected to train the machine learning models that the industry is implementing. The stability of the machine learning models will be tested using ACT full cab driving simulator in different situations. Based on the results of machine learning models, the corresponding car-following models will be formulated to establish the macroscopic dynamic at the network level. The impact of this study is expected to be significant as it will establish the connection between machine learning models and car-following models, and will steer research and development of future AV technologies towards artificial intelligence models that are guaranteed to be stable.

Statistical Learning, Driving Simulator-Based Modeling and Computational Tractable Dynamic Traffic Assignment

Time: 2017 – 2021

Sponsor: NSF

Description:

This study is to develop a new dynamic traffic assignment modeling methodology by leveraging statistical learning and driving simulator-based experiments. The proposed methodology will enhance not only modeling realism of agent behavior through observations from the driving simulator-based experiments, but also computational tractability by replacing the computationally expensive dynamic network loading process with response surfaces constructed based on field data and agent behavioral response in the driving simulator experiments. In addition, this study will integrate learning process models to enable modeling flexibility to capture disequilibrating traffic dynamics that can be impacted by innovations in information technologies and/or based on observations from the driving simulator experiments. This will lead to new theories of driver behavior and adaptation to emergent travel environments under advanced and emergent technologies.

SCC-IRG Track 1: Fostering Smart and Sustainable Travel through Engaged Communities using Integrated Multidimensional Information-Based Solutions

Time: 2021 – 2025

Sponsor: NSF

Description:

This project aims to develop systematic deployment tools that smart and connected communities can use to achieve their sustainable travel goals in a quantifiable manner by leveraging advances in information, communication, and sensor technologies. While the deployment of advanced technological solutions offers great promise for communities to improve residents’ quality of life and prosperity, they are faced with significant challenges in realizing these aspirations due to the diversity in technological and travel needs and barriers faced. Solutions to enhance travel mobility, safety, equity, and access will be developed using the City of Peachtree Corners (GA) as a living lab. Solutions will include building partnerships of emerging micromobility services, behavioral interventions, and public policy interventions. These solutions will be developed using data collected from community residents and other sources, and deployed using an information design system that provides targeted information delivery to various stakeholders. This project draws on methods from multi-objective and multi-agent optimization, machine learning, behavioral economics, and data and policy analytics, to generate multidimensional solutions for the community.

Incorporating Situational Awareness Cues in Virtual Reality (VR) Environments for Users of Autonomous Vehicles

Collaborators: Bruce Walker and Nadia Fereydooni (School of Psychology; School of Interactive Computing)

Time: 2021

Description:

Technological advances are attracting an increasing number of individuals to use VR in vehicles. However, users report concerns of losing track of real-world occurrences. Given that VR devices completely block the road environment from the user, this study aims to develop a compensating mechanism that will allow passengers to stay aware of their real-world surroundings while engaging in the VR environment. Situational awareness cues are embedded within the VR environment of one group of participants, but not the other. Using validated questionnaires such as SART (Situational Awareness Rating Technique) and NASA TLX (Task Load Index), participants’ informational needs are examined, along with their perceived risk, perceived comfort (physical and psychological), task workload and trust in technology.

Driving Simulator-Based Study of the Impacts of Various Roadway Design Modifications on the Curiosity Lab Test Track

Time: 2020 – 2021

Sponsor: City of Peachtree Corners, GA

Description:

This study aims to understand the role of urban infrastructure modifications on the performance of Autonomous Vehicles (AVs) in mixed traffic environments, and what infrastructure designs can enhance transportation safety in smart-city environments by developing a digital twin of Curiosity Lab in a high-fidelity driving simulator. Data from simulator-based experiments on traffic conditions and driver behavior/response under different infrastructure modifications (e.g., dedicated AV lane, lane width, barriers, and markings) will be collected and analyzed to develop implementation guidance to the Curiosity Lab for potential future roadway modifications to enhance consumer acceptance of AVs and their adoption in typical urban environments. The study aims to develop a virtual AV test track of Curiosity Lab with varied roadway characteristics, investigate the impacts of roadway modifications on human drivers using driving simulator-based experiments and develop design guidelines for AV testing and deployment. 

Simulation of Competitive Ride-Hail Services

Time: 2020 – 2022

Sponsor: USDOT (T-SCORE)

Description:

This study will develop a simulation model capable of representing tradeoffs between independent ride-hailing operators and coordinated demand-responsive transit. The proposed simulation model aims to capture the intricate relationship between supply for and demand of on-demand transit and independent ride-hail services. Based on the inputs of ride-hail requests, the model can not only predict the rides and routes for ride-hailing operators and demand-responsive transit, but also estimate wait time and cost for travelers. The results will be implemented in the open-source MATSim simulator to help researchers and transit operators evaluate transit strategies to achieve maximum complementarity across modes.

Using Driving Simulator Environment to Determine Interactions between User Behavior and Infrastructure Design Under Autonomous Vehicles

Time: 2019 – 2021

Sponsor: USDOT (CCAT)

Description:

This study will use immersive driving simulator environments in conjunction with stated preference surveys to analyze the interactions between user behavior and infrastructure design changes under AVs while also factoring emerging trends of share-mobility services, electrification, and the promotion of sustainable transportation modes (such as mass transit, biking, and walking). A road corridor in Atlanta, GA with current road infrastructure design and human-driven vehicle environment will provide the base case for the driving simulator environment. Further modifications are made to the base case driving simulator environment to include AVs infrastructure design, shared-mobility services, electrification, and sustainable transportation modes. The study insights will be used to develop guidelines that can aid state and local transportation agencies to develop near- and medium-term infrastructure design modifications to enable efficient, smooth transition to an AV future that additional factors other emerging transportation-related trends.

Smartphone-Based Incentive Framework for Dynamic Network-Level Traffic Congestion Management

Time: 2019 – 2021

Sponsor: USDOT (STRIDE)

Description:

By leveraging advances in smartphone-based personalization, big data availability for traffic, and network-level integration through information-based connectivity, this study will develop a smartphone-based framework to develop real-time incentives (monetary, value-based, travel-related credits, etc.) to influence drivers’ en route routing decisions to manage network-level system performance in congested dynamic traffic networks. Analytical models and algorithms will be developed to identify and implement the specific incentives to be deployed in real-time, using techniques from game theory, optimization, and machine learning. This study will ensure that the proposed incentives are behavior-consistent and tailored based on travelers’ smartphone response. The study will also explore how public and private sector transportation entities can collaborate through the use of new forms of incentives that leverage the emerging transportation future, in addition to the currently-used ones.

CPS: Medium: Emulating Emerging Autonomous Vehicle Technologies

Time: 2019 – 2022

Sponsor: NSF

Description:

The study aims to develop an analytical and numerical framework to emulate the impacts of current automated vehicle (AV) technologies on transportation networks in the near future. Empirical data from Level 2/3 AVs on the market will be collected to train the machine learning models that the industry is implementing. The stability of the machine learning models will be tested using ACT full cab driving simulator in different situations. Based on the results of machine learning models, the corresponding car-following models will be formulated to establish the macroscopic dynamic at the network level. The impact of this study is expected to be significant as it will establish the connection between machine learning models and car-following models, and will steer research and development of future AV technologies towards artificial intelligence models that are guaranteed to be stable.

Statistical Learning, Driving Simulator-Based Modeling and Computational Tractable Dynamic Traffic Assignment

Time: 2017 – 2021

Sponsor: NSF

Description:

This study is to develop a new dynamic traffic assignment modeling methodology by leveraging statistical learning and driving simulator-based experiments. The proposed methodology will enhance not only modeling realism of agent behavior through observations from the driving simulator-based experiments, but also computational tractability by replacing the computationally expensive dynamic network loading process with response surfaces constructed based on field data and agent behavioral response in the driving simulator experiments. In addition, this study will integrate learning process models to enable modeling flexibility to capture disequilibrating traffic dynamics that can be impacted by innovations in information technologies and/or based on observations from the driving simulator experiments. This will lead to new theories of driver behavior and adaptation to emergent travel environments under advanced and emergent technologies.