My general area of research lies at the intersection of control, optimization, and machine learning. My current applications include mobile robotics, transportation, and engineering. My past applications include connected health and sensing of water systems. The problems I am generally interested in focus on the integration of novel data sources into mathematical learning models. They also involve the application of machine learning algorithms to large scale mobility problems. The techniques I use include game theory, convex optimization, network optimization, deep reinforcement learning, partial differential equations, and numerical analysis.

The research lab currently has two major focuses:


FLOW is a deep reinforcement learning framework implemented on AWS EC2 and used for learning and optimization over microsimulation tools for traffic flow. Its main application includes mixed autonomy traffic, in which we are studying the impact of a small proportion of self-driving vehicles on the rest of traffic flow. FLOW is a traffic control benchmarking framework. It provides a suite of traffic control scenarios (benchmarks), tools for designing custom traffic scenarios, and integration with deep reinforcement learning and traffic microsimulation libraries. To this day, it already includes two of them: SUMO and AIMSUN. Recently, we deployed benchmarks that are now available to the research community to benchmark their different algorithms.

Network Optimization and Analysis of the Impact of Information on Traffic Flow

This project focuses on analysis of the impact of routing apps, such as Google, Waze, Apple traffic, INRIX, etc. Our approach develops new network traffic flow models that incorporate the impact of routing information on traffic flow and routing. We provide theoretical analysis of the resulting mathematical framework, as well as numerical simulations for practical cases of the impact of such apps on congestion.

We have two testbed applications we are currently working on:

Trucking automation

This project focuses on the design, prototyping, deployment, and testing of novel algorithms for truck platooning. We are also simultaneously using classical control techniques for the coordination of multiple trucks, and novel applications of deep learning and deep reinforcement learning for the same coordination problems. We are currently working in collaboration with Volvo and KACST, the King Abdulaziz Center for Science and Technology, on these problems both in the United States and Saudi Arabia.

Connected Corridors

The Connected Corridors is a collaborative program to research, develop, and test an Integrative Corridor Management (ICM) approach to managing transportation corridors in California. ICM views the corridor as a total system to be managed as an integrative and cohesive whole. It seeks to address the corridor’s overall transportation needs rather than the needs of particular elements of agencies alone. This project relies on large scale microsimulation and optimization of the I-210 freeway in Los Angeles.

Past projects include the following:

The Floating Sensor Network

Over the course of five years, the lab developed an entire fleet of a hundred aquatic motorized robots capable of navigating currents and transmitting water quality and water river flow measurements in real time to a backend system. This system performed large scale data assimilation to integrate mobile measurements into hydrodynamic models. The development of this system led to several field operational tests involving up to a hundred floating sensors at the same time in the Sacramento-San Joaquin River. The project was concluded in 2014.

Mobile Millennium

Mobile Millennium was the first traffic app deployed in North America to run on RIM and Symbian operating systems (Blackberry and Nokia phones) in November 2008. It ran in Northern California to enroll up to 10,000 users to participate in traffic data collection using GPS-enabled cellular phones. The project ultimately led to the creation of a traffic information system used jointly by the California Department of Transportation, UC Berkeley, and Nokia. The project was concluded in 2010.

Mobile Century

On February 8th, 2008, UC Berkeley ran a test involving 100 cars to measure traffic on the I-880 freeway in California to demonstrate the possibility of traffic reconstruction using cellular phones. This project contributed to the launch of the Mobile Millennium project, a longer partnership with Nokia.


The videos below show some research activities of the group, more videos and photos in the gallery section

Fung Institute, Master of Engineering Capstone Project Video

FSN Centerline and Lane Splitting


FSN El Cerrito Pool test