Metcalfe’s Law and networked autonomous cars - Embedded.com

Metcalfe’s Law and networked autonomous cars

MADISON, Wis. — When economists discuss the “network effect,” they mean that the value of a product, or service, is dependent on the number of others using it.

Think telephones or fax machines. If nobody else has them, your fax machine or telephone actually has little value, because you have nobody to call, nobody to receive your fax.

Now, the automotive industry is about to step into a brave new world where the network effect rules. The value of a car you just purchased might soon be determined, not by your car alone, but by how many others also have cars like yours.

Specifically, this phenomenon involves cars with V2V (vehicle-to-vehicle) or V2I (vehicle-to-infrastructure) communication features, ultimately including self-driving cars.

As Egil Juliussen, principal analyst, responsible for Infotainment and ADAS at IHS Automotive, put it in a recent interview with EE Times, cars of the future (V2V, V2I, and self-driving) will be “the first automotive product that depends on Metcalfe's law.”

Metcalfe's law, attributed to Robert Metcalfe in reference to the Ethernet, is a theory similar to the network effect. It states that the value of a telecommunications network is proportional to the square of the number of connected users in the system.

Most of the new automotive technologies carmakers and chip suppliers talk about today are those enabling cars of the future to talk to other cars (V2V), while allowing cars to communicate with traffic lights, lamp posts, and any other infrastructure (V2I). The promise of connected cars — and eventually, self-driving cars — is driven by such goals as avoiding accidents (safety) and smoother traffic (efficiency) on the road.

Autopilot in aircraft
As attractive as connected cars sound, one of the biggest hurdles facing the industry is the hard reality that one car with V2V or V2I capability won't do any good.

In a recent phone interview with EE Times, Kaivan Karimi, executive director of global strategy and business development for the Microcontroller group at Freescale Semiconductor, compared self-driving cars with the autopilot function in airplanes.

If a control tower has just issued a warning of a “near miss” between two airplanes in the air, cautioning pilots to watch out, pilots aren't supposed to make independent judgments as to where to turn to avoid collision. That's the autopilot's job. “The whole point of autopilot is to take human judgment out of it,” Karimi told us.

If they're going to work, the same rules have to apply to self-driving cars on highways. Unless every car on the road is equipped with similar self-driving capabilities — just as every aircraft flying in the sky has an autopilot system — one bad human judgment by the driver of a non-self-driving car could cancel a good decision by a self-driving car.

Both Freescale's Karimi and IHS's Juliussen, however, appear fairly confident that various technologies — ranging from camera, radar, and sensors to GPS data, 802.11p, LTE, and more — will become available. This variety of disparate tools and technologies will not only help steer the car down the center of its lane but also help avoid collisions, they say.

But how a potential mandate for V2V and V2I will play into the future scenario of road safety and traffic efficiency remains uncertain. Among the wild cards are politics, government budgets, new legislations, and machismo.

I mean, really, can you picture Chuck Norris in a self-driving Volvo?

This blog was originally published on EETimes News and Commentary.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.