Dr. Cong Liu, assistant professor of computer science in the Erik Jonsson School of Engineering and Computer Science, is producing algorithmic and system solutions that process massive amounts of data in real time. This allows designers across several industries to develop GPU-accelerated embedded cyber-physical systems — intelligent engineering systems that interact with the physical environment and make autonomous control decisions such as autonomous vehicles and robotics.
Liu’s project, D3: Addressing Emerging Data-Induced Challenges in Embedded and Real-Time Systems, earned him a Faculty Early Career Development (CAREER) Award from the National Science Foundation, providing him more than $500,000 in funding over the next five years to further his research in autonomous driving.
“If successful, my research will provide a set of hardware/software-combined solutions that take the first step for making the execution of autonomous driving workloads controllable and predictable in a resource-constrained embedded automotive platform,” he said.
In other words: Your car will make driving decisions for you.
Embedded systems must monitor humans and the environment to make intelligent decisions, Liu said.
“The embedded system receives tens or maybe hundreds of data streams, which are recorded by a series of sensors and cameras,” said Liu, who has been at UT Dallas since 2013. “Each data stream needs to be processed in real time so that a decision can be made on whether the vehicle should accelerate, brake, turn left or turn right. If the data is not processed in real time, then the system fails.”
Liu said the camera sensors have to process and analyze huge amounts of data using a limited amount of storage space. The data is not in the cloud, or on a desktop or server.
“It needs to be in a small space in a car, using small wattage,” he said. “This becomes a significant challenge because it’s difficult to have enough computing capability to process all these complicated data streams in real time, and intelligently figure out what to do for the next 15 milliseconds. My proposal puts forth a set of techniques ranging from the application level to run-time resource scheduling and location, all the way down to system-level implementation and design.”
Liu is confident that driverless vehicles will be in full swing in 10 years, with limited routes within the next five.
“I think vehicles containing a good set of driver-assisted or even driverless tools will be the norm in five to 10 years as researchers overcome major obstacles in different domains, including computer vision, operating systems and algorithms. Some of the tools such as lane following and auto parking are already available in many of the cars we can buy today,” he said.
Source | Jonsson School Research eNews
ABOUT THE UT DALLAS COMPUTER SCIENCE DEPARTMENT
The UT Dallas Computer Science program is one of the largest Computer Science departments in the United States with over 2,800 bachelors-degree students, more than 1,000 master’s students, 190 Ph.D. students, 52 tenure-track faculty members, and 41 full-time senior lecturers, as of Fall 2018. With The University of Texas at Dallas’ unique history of starting as a graduate institution first, the CS Department is built on a legacy of valuing innovative research and providing advanced training for software engineers and computer scientists.