Skip to main content

Innovative Mobility

Vanderbilt and its partners are innovating across the spectrum of mobility from roads and vehicles to transportation systems and city services. Learn more about how our collaborative teams are delivering highly advanced, learning-enabled autonomous vehicle control systems that can revolutionize inter-city mobility, artificial intelligence that can efficiently connect communities via multi-modal transit operations, and award-winning, practitioner-centric intelligent solutions for next-level emergency response management.

Real-World Mobility Testbeds: Safer, Smarter, Healthier Roads

Traffic congestion is a major challenge for modern cities, but what if new technologies could help reduce traffic congestion, fuel consumption, and crashes? Vanderbilt’s Associate Professor of Civil and Environmental Engineering Dan Work and the Tennessee Department of Transportation (TDOT) aim to do just that with the I-24 Motion, a world-class mobility testbed that uses hundreds of ultra-high-definition sensors to create digital models of traffic in unparalleled detail (Fig. 1).

Together, Vanderbilt, TDOT, and industry collaborators are well on their way to creating the smartest roadway in the world with advancements like I-24 MOTION, a first-of-its-kind 4-mile testbed for autonomous vehicles and an artificial intelligence (AI)-based decision support system (DSS) that can improve traffic flow and incidence response with coordinated arterial traffic signals and dynamic speed limits that respond to real-world conditions and driver behavior.

I-24 MOTION mobility testbed
Figure 1. High-level overview of I-24 MOTION mobility testbed goals and capabilities currently under development.


The team drove controllers trained by reinforcement learning algorithms on live freeways in 2021. The Congestion Impacts Reduction via CAV-in-the-loop Lagrangian Energy Smoothing (CIRCLES) project is funded by U.S. DOE, project CID DE-EE0008872. Institutions involved are UC Berkeley (Prime), Rutgers University, Temple University, University of Arizona, and Vanderbilt University. Partners include the Tennessee Department of Transportation, Toyota, General Motors, and Nissan. This material is based upon work supported by the U.S. Department of Energy’s Office of Energy Efficiency and Renewable Energy (EERE) award number CID DE-EE0008872. The views expressed herein do not necessarily represent the views of the U.S. Department of Energy or the United States Government.


But our collaborations with TDOT don’t end there. Vanderbilt has already demonstrated how to advance sustainable transportation by using its own campus as a testbed: With MoveVU, we showed how commuters can be incentivized to take public transportation to campus, and expanded their options for safe and healthy travel with better bike lanes throughout campus, and more. Now, Professor of Civil and Environmental Engineering Mark Abkowitz and his team are building on MoveVU with $8.4M in equally shared support from TDOT and Vanderbilt University to scale up sustainability and reduce emissions even further with improved shuttle service transit, bike share services, traffic optimization technologies, and new curbside and micro-mobility best practices.

Figure 2. To better evaluate transportation and traffic congestion impacts the MoveVU research team has integrated a variety of data sources (left panel) for their campus testbed (right panel), which they aim to soon expand to other areas of Nashville. The goal is to evaluate and prioritize the most effective multimodal and micro-transportation solutions to promote public health, social equity, and greater connectivity. The integration of personal trackers allows the team to measure emotional stress responses in commuters for geolocation and time scales as they react to the built environment allowing the team to quantify anecdotal experiences to quantifiable data enabling planners to make data-driven safety decisions.

Our researchers already discovered that even a small percentage of driverless vehicles can increase traffic fuel efficiency up to 40%, which we know is important to maximize efficiency and address electric vehicle range anxiety. Learn more about TDOT’s pioneering I-24 Smart Corridor and Vanderbilt faculty Dan Work, Jonathan Sprinkle, and Mark Abkowitz.

Autonomous Control Systems: Safer, Smarter Land, Air and Water Vehicles

Autonomous vehicles can dramatically improve safety, but they come with their own safety challenges. At Vanderbilt, Professor of Computer Science Gabor Karsai and Vanderbilt collaborators Xenofon Koutsoukos, Taylor Johnson, Abhishek Dubey, and Ted Bapty are tackling these challenges head-on by developing a new autonomous vehicle control system that is capable of continually learning, assessing, and assuring safety in highly uncertain environments with support from DARPA’s Assured Autonomy program. This state-of-the-art architecture implements real-time contingency-management functions and adjustable decision-making logic to ensure safe operations of driverless cars, drones, and even watercraft.

With a team of collaborators at NASA and other partner institutions, Cornelius Vanderbilt Professor of Engineering Gautam Biswas is taking this kind of thinking to the skies with the development of a new electrical, autonomous ‘air taxi’ for advanced inter-city mobility (Fig. 3). To ensure these aircraft operate safely, Biswas and his team at Vanderbilt are building learning algorithms that enable pilotless risk evaluation and real-time decision-making in highly challenging situations. In the future, these services could also enhance connectivity to rural and remote areas.

Figure 3. The NASA-sponsored project for systemwide safety for autonomous air taxis will cover a variety of aircraft classes for operations within cities and along the urban-rural continuum (left panel). The system incorporates in-flight monitoring of aircraft health diagnostics, hazard condition monitoring, and adaptive hazard mitigation (center). The team is developing these capabilities for a variety of aircraft including the Jobe, Tarot-18, and Uber aircraft (right).

Continuing the theme of systems that learn, Vanderbilt Assistant Professor of Computer Science Soheil Kolouri is leading an international DARPA-funded initiative to develop artificial intelligence programs that continually share their lifetime experiences with each other, an approach that enables machines to reuse information and adapt to new conditions much as human beings learn from interacting with each other. Real-world applications include self-driving cars and systems for robotic exploration, emergency monitoring, and cyber security.

Learn more about Vanderbilt faculty Gabor Karsai, Gautam Biswas, and Soheil Kolouri.

Efficient Public Transit, Smart City Services, and Emergency Response Management Solutions

Efficiently operating public transportation is a critical challenge for communities across the country, but what if we can create an integrated multi-modal public transportation service that combines fixed-transit, dynamic microtransit and micro-mobility options such as bikes? Vanderbilt faculty member Abhishek Dubey, the Chattanooga Area Regional Transportation Authority (CARTA), and Nashville WeGo are collaborating to do just that by developing more efficient transit operation algorithms for smart and connected communities (Fig. 4). The team focuses on predicting the expected demand (CARTA, WeGo), operationalizing on-demand microtransit (CARTA), optimizing the schedule and vehicle assignment for energy consumption based on the predicted information (CARTA), and providing support for scheduling additional vehicles to reduce high occupancy levels on fixed line buses (WeGo). The approach includes state-of-the-art machine learning solutions, high-resolution telemetry, and sequential decision support algorithms using reinforcement learning and Monte-Carlo tree search. More information is available at

Smart Transit for All - Abhishek Dubey
Figure 4. The team uses an integrative socio-technical approach (A) to design cyber-physical systems that address modern-day transportation challenges by integrating a range of data. For this project, they developed a decision program enabling real-time vehicle routing and logistics for public transportation in collaboration with Chattanooga’s CARTA (B) and Nashville’s WeGo. The program addresses a variety of challenges such as emission reductions, transit schedule optimization, and demand smoothing.

These days, almost everything from traffic lights to parking meters collects data, but as long as the data from these systems remain disconnected, the most innovative solutions to pressing safety and transportation challenges remain beyond reach. To surmount this challenge, multiple teams at Vanderbilt are developing more intelligent, interoperable cyber-physical systems (CPS), including sensors and data from traffic lights, parking garages, wearable technology, and smartphones. Faculty members Abhishek Dubey and Douglas Schmidt are leading the Siemens-funded development of an application architecture called CHARIOT for a universal cyber-physical component model that allows distributed CPS applications to be constructed using different software and hardware components without being tied down to any specific platform or middleware.

Even more pressing for resilient communities is the need to respond efficiently and effectively to emergency situations and proactively assign first responders where they are most needed. To do that, Abhishek Dubey and his team are developing an open-source, practitioner-centric toolchain to help first responders understand where and when incidents occur, and how to allocate responders in anticipation of incidents (Fig. 5). This innovation could shift the paradigm in Emergency Response Management from reactive to proactive, saving time, equipment, and most importantly, people’s lives. It has already been showcased at multiple global smart city summits, won innovation awards from the Government Technology Magazine and at the International Conference on Learning Representations’ AI for Social Good workshop, and was featured in the Financial Times.

Emergency and Incidence Response Toolkit - Abhishek Dubey
Figure 5. Left Panel: By integrating weather, traffic, accident, and historic emergency data the team developed a self-learning incidence response toolchain able to forecast traffic incidents to allocate response resources based on current conditions, detect incidents via crowdsources and social media data, and create a faster emergency response. Right Panel: Comparison of incidents predicted by the model (A), and real incident distribution (B) over January 2019.

Learn more about Vanderbilt faculty Abhishek Dubey and Douglas Schmidt.

Return Home            Contact Us