Zhi-Li Zhang Leads $7M NSF Breaking Low Project on Improving 5G Networks to Enable AV Teleoperations
Department of Computer Science & Engineering Professor Zhi-Li Zhang is the principal investigator (PI) on a two-year, $7 million project that aims to improve fifth-generation (5G) networks in order to enable autonomous vehicle (AV) teleoperations. This grant is part of the National Science Foundation’s (NSF) Breaking the Low Latency Barrier for Verticals in Next-G Wireless Networks (NSF Breaking Low) initiative, a $17 million investment that aims to identify and solve critical architectural, technical and technological issues that address key barriers in current 5G and next-generation (Next-G) wireless networks.
A joint venture between the University of Minnesota (UMN), University of Utah (Utah), and University of California Riverside (UCR), with industry partners General Motors (GM) and Nokia, the DRIVE-SAFE project is one of three projects funded by the NSF Breaking Low Ideas Lab. The research will focus on enhancing the innovative vertical-aware framework to optimize both 5G networks and the vertical application that will allow for seamless remote operations of self-driving cars.
Despite the progress over the last decade, there are still situations that AVs do not know how to handle, like construction and accident zones. The DRIVE-SAFE project is creating teleoperation systems with a human operating AVs remotely when a complex situation arises. By developing and improving both the vehicle's technology and the network infrastructure, the team aims to create a system that is both reliable and safe.
“In order to successfully build a teleoperations system, you need high-bandwidth connectivity, which is where 5G comes into play,” Zhang said. “It also needs to have low latency, so that the video data being sent to the driver comes in near-real time. Today’s 5G networks cannot support this low-latency, high-bandwidth data transfer. So the project focuses on enhancing today’s 5G network to make this system possible. It is a vertical application of the technology in a specific use-case.”
UMN will lead the project, providing overall management and execution, with a primary focus on developing remote vehicle operation capabilities. Utah and Nokia will support the development of network-aware optimization techniques to achieve low-latency communications. GM will lead the instrumentation of a commercial-grade autonomous vehicle platform to enable and support remote operation. UCR will further enhance the system by integrating collaborative driving capabilities through integrating the V2X communications.
The two-year project will culminate in a live demonstration of the AV teleoperations system. After the final demonstration of their system, the DRIVE-SAFE team will have an opportunity to advance to a second round of funding to further develop their system, so that it can be scaled and ultimately adopted by industry. The NSF initiative aims to establish the U.S. as a leader in 5G next generation technology, and to 6G and beyond.
Zhang has been working on problems related to 5G networks and autonomous vehicles since 2018. As AVs began entering the market, Zhang’s group began researching how computing systems could help support this new emerging technology.
“We have had a number of collaborators at the University of Minnesota looking at how autonomous vehicles could solve commuting problems,” Zhang said. “When we started working on this topic, we quickly realized that fully autonomous vehicles are not quite there yet, which is when we pivoted to the remote driving problem. Since my background is in computing networks and communications, I started thinking about how 5G could support remote driving and we have been working on that since 2020. Now we are working on improving both systems to make teleoperated AVs possible, and have had a number of grants to keep this work moving forward.”
Over the last seven years, Zhang has been putting together the building blocks for this $7M DRIVE-SAFE project through a number of grants from UMN and NSF. He was part of the research group that lobbied for the University to invest in their own autonomous vehicles, MinnCab, which will be used in the demonstrations for the current NSF project. In total, Zhang’s work has received more than $6.5 million in prior research grants and gifts for broad research funding from federal and state agencies, UMN, and industry for related projects that have led to his current grant, including funding from the College of Science and Engineering, Center for Transportation Studies, and various NSF projects.
Learn more about the DRIVE-SAFE Project, one of the three projects awarded by the NSF by reading the award announcement and checking out the sister projects, Agora and XRNet.