Zhi-Li Zhang

What they work on

1. Develop an O-RAN-based framework in which 5G networks use vehicle trajectory and speed to proactively cut delays from poor radio signals and base-station handovers during remote driving.
2. Integrate 5G with cellular vehicle-to-everything (C-V2X) communication so connected vehicles share situation awareness for safer cooperative and remote driving.
3. Build NextMOVE, a prototype autonomous-vehicle teleoperation platform combining fine-grained low-latency networking, a secure edge computing architecture and AI-native designs with safeguards against wrong predictions.
4. Support human-in-the-loop teleoperation by combining faster-than-real-time AI simulation with human operator judgment, including tests in Midwest winter conditions.
5. Re-architect network function virtualization as NFLambda, splitting network functions into small actors in a functional reactive programming style to process traffic at 100 Gbps and beyond.
6. Use smart network cards, programmable switches, hardware accelerators and compiler optimizations to scale NFLambda service function chains on commodity multi-core servers.

More about the research

Why it matters

Reliable low-latency 5G links could let human operators safely guide autonomous vehicles through situations they cannot handle alone, which could speed safe adoption of robotaxis and similar services. The same tools could also help mobile network operators, equipment vendors, smart manufacturing, precision agriculture and telehealth. Faster software packet processing could make 5G and edge networks cheaper and more flexible.

Known grants

From NIH and NSF research grants. Labs often have other funding we can't see.

Closest labs

The school

University of Minnesota: 88 labs on PI Pages, $1.4B a year on research, 722 research doctorates a year.

More about University of Minnesota

Reach out

A start for your email. Fill in the brackets.