C-V2X would let vehicles communicate with, well, everything
Companies investing in autonomous driving have logged thousands of hours and millions of miles on the road to test out sensor arrays, Lidar, radar, cameras, etc…But these elaborate systems still face limitations, particularly when it comes to tracking and predicting the movement of pedestrians, bicyclists and other dynamic objects. With C-V2X seeing global traction among operators, auto makers and other interests, Li Erra Li, chief scientist of Pony.ai, called out the role of wireless networking in advancing autonomous driving.
Li last week presented at the Texas Wireless Summit, hosted annually by the University of Texas–Austin’s Wireless Networking and Communications Group. In addition to his role at Pony.ai, Li serves as an IEEE and ACM Fellow.
He said that, despite the benefits Lidar realizes, there are still problems around sensitivity to fog, rain, car exhaust, vegetation and the like. Coupled with radar and cameras, “each of these sensors give us some piece of the puzzle, some piece of the information. How can we use wireless communication networks to increase the reliability and the accuracy of our perception system?”
Li described the ability to detect and track moving objects wirelessly through the ability for a vehicle to communicate with other vehicles, pedestrians and roadside infrastructure. Through in a connection to the cloud and that’s C-V2X in a nutshell. “We could have communication devices on these dynamic objects,” he said. “Essentially we solve the problem very easily. You also have the case where objects may not all have the communication devices. There what we can do is leverage the detection of nearby vehicles or actually infrastructure. The good thing about outfitting the road infrastructure is you’re looking down.”