Autonomous cars on the roads today are no better than human drivers, according to Professor Gerhard Fettweis, the Vodafone chair at the Technical University of Dresden.
A veteran of several successful wireless startups, Fettweis is currently researching millimeter wave applications that could be part of future connected cars. Fettweis recently joined RCR Wireless News to discuss his research ahead of the Enterprise IoT Summit.
“If you look at autonomous driving like a Tesla or a Google does it, and you go on [Highway] 101 in the Bay Area and see a Tesla or a Google car in autonomous driving mode, it’s horrible,” Fettweis said. “The space it leaves in front of it is so big so that it doesn’t come to this build up … so it is totally ineffective in terms of being able to move a ton of cars over the freeway. So you have to start now going from autonomous vehicles to highly automated driving, meaning that they have to be connected, they have to talk to each other so that they then can actually decrease the distance.”
The answer, said Fettweis, is highly automated driving as opposed to autonomous driving. Highly automated driving means the vehicles are connected to one another. Fettweis said trucks, which are already “platooning” in some tests, need five millisecond latency communication for safety, and the distance between vehicles, or “drag,” is typically five meters at a speed of 55 mph. Cars will need even lower latency, he said.
“Cars are going not at 55, they’re going maybe at 75, and in Germany even faster,” Fettweis said. “Their drag is not as big so you have to get them closer to each other and then you have to get it down to this millisecond.”
Follow me on Twitter.
