CAREER Award Fuels Work on Control Systems for Autonomous Vehicles

Autonomous vehicles need to process enormous volumes of constantly changing and unpredictable data from sensors and detectors in order to make important decisions, such as whether to change lanes.

Developing control systems and networks that can manage these massive data inputs is a major challenge for getting vehicles to their destinations safely and efficiently, said Dr. Tyler Summers, associate professor of mechanical engineering at The University of Texas at Dallas. Summers has received a five-year $500,000 Faculty Early Career Development Program (CAREER) Award from the National Science Foundation to support his and his students’ research aimed at developing such control systems and networks.

“Our challenge is determining how to architect control systems and networks to produce desired behaviors in the face of uncertainty and adversarial influences,” said Summers, who concentrates on the abstract mathematical aspects of this research area in the Erik Jonsson School of Engineering and Computer Science.

As an example of an application of his and his team’s work, Summers said robust control systems and networks are needed to reduce the risk of cyberattacks that have caused malfunctions in some autonomous vehicles. They also are working on solutions that address inaccurate measurements or misclassifications that could compromise a vehicle’s performance. His research applies to the operation of a single autonomous vehicle or the coordination of fleets of vehicles.

Summers’ research also has a broader range of commercial and military applications.

“Control networks need to allow the grid to operate with large amounts of wind and solar energy. We’ve had days when wind provided half of the state’s power, but there’s a huge challenge of how to reliably operate the grid as intermittent and unpredictable energy sources become more prevalent. We need to rethink the control architecture of the grid.”

Dr. Tyler Summers

Associate Professor of Mechanical Engineering in 
the Erik Jonsson School of Engineering and Computer Science

Networks have become increasingly complex with the integration of machine learning, he said. For example, electricity grids must interface with a growing number and type of data and algorithms to be capable of integrating renewable, fluctuating power sources.

Summers leads the Control, Optimization, and Networks Lab, where he and students conduct research on autonomous systems, including Super COMO, an autonomous artificial intelligence vehicle. The project, funded in part by a $350,000 grant he received in 2017 from the U.S. Army Research Office‘s Young Investigator Program, is part of his research on connecting sensors and actuators into networks.

After earning a PhD in aerospace engineering at The University of Texas at Austin in 2010, Summers served as a postdoctoral fellow at the ETH Zürich, a research university in Switzerland, before joining UT Dallas in 2015.