Editor’s note: Our story was based on the premise that the advanced radars discussed below were intended for autonmous cars. GM contacted IEEE Spectrum after publication to say that the radars are not intended for autonomous vehicles. According to GM:
The FCC filings referenced in the IEEE Spectrum story are not part of our autonomous vehicle development program. They are related to further advancement of technologies featured on our vehicles today.
The original story continues below:
General Motors (GM) is likely building a fleet of 725 self-driving taxis for its partnership with Lyft, with an intended launch date of January 2019, according to documents filed with the U.S. Federal Communications Commission (FCC).
This is the first concrete indication of the scale and intended deployment date of GM and Lyft’s autonomous on-demand network since GM announced a US $500 million investment in Lyft in January.
Any organization that wants to use a radio frequency device that has not been certified by the FCC needs to apply for an experimental license. This can include the radar systems that many autonomous and semi-autonomous vehicles rely on to detect other road users.
Most experimental licenses involve only a handful of units. Earlier this year, for instance, Google applied for permission to test its 76-GHz long-range radars on its couple of dozen self-driving cars.
In late July, however, engineers at General Motors filed applications for thousands of millimeter range radar systems. The first involved 725 units of a device made by Conti Temic, a subsidiary of Continental, whose part number indicates that it is a long range radar. The second application was for 2175 units of a short range radar.
The 76-GHz radars represent “a new hardware generation”, according to a written request to the FCC for confidentiality by Jeffrey Clark, a GM engineer in charge of long range radars.
Self-driving cars typically have one long range radar facing forward, to detect cars up to several hundred meters away, and three short-range radars on the sides and back of the vehicle, with a range in the tens of meters. (2175 divided by three is 725.)
Such a high number of vehicles suggests a significant pilot program, probably in a single American city. (The licenses granted by FCC cover the entire continental United States.) This aligns with reports in May that Lyft would start testing a fleet of electric Chevrolet Bolt taxis in an undisclosed U.S. city within a year. GM is already testing a couple of autonomous Bolts in California, using technology from Silicon Valley self-driving start-up Cruise Automation which it recently acquired for just under $600 million.
In a call with investors in July, GM CEO Mary Teresa Barra said:
“We’re working aggressively. The fact that we have the vehicles and we’re able to so quickly integrate the Cruise Automation software speaks to the speed at which we’re moving.”
In his letter to the FCC, Jeffrey Clark requested that “the duration of the experimental license shall cover the time until the market introduction, which is January 2019.” In the end, however, the FCC granted GM permission to use the new radars only until 1 August 2018.
Ride-share rival Uber, which is currently testing an autonomous Ford Fusion in Pittsburgh, says that the company is “still in the early days of our self-driving efforts.”
Neither GM, Lyft, nor Continental responded to requests for a comment.
This story was corrected on 10 August to give the right end date for the FCC permission.
Mark Harris is an investigative science and technology reporter based in Seattle, with a particular interest in robotics, transportation, green technologies, and medical devices. He’s on Twitter at @meharris and email at mark(at)meharris(dot)com. Email or DM for Signal number for sensitive/encrypted messaging.