When Google is self-driving The pilot car project began in the Bay Area in 2009, with engineers focusing on highways by sending sensor-laden vehicles down Interstate 280, which runs the length of the Silicon Valley peninsula.
More than 15 years later, cars are back on the highway—this time without drivers. On Tuesday, the project, now a subsidiary of the Alphabet we all know as Waymo, announced that its robotaxi service will now ply freeways in the San Francisco Bay Area, Los Angeles and Phoenix.
The new service is another technical leap for Waymo, whose robot taxis currently serve five US metros: Atlanta, Austin, Los Angeles, Phoenix and the San Francisco Bay Area. The company says it will launch in several other US and international cities next year, including Dallas, Miami, Nashville, Las Vegas, Detroit and London.
Waymo also announced Wednesday that it will launch pickup and drop-off services at San Jose Minta International Airport, allowing travelers to theoretically drive autonomously all the way from San Francisco to San Jose — a service area of about 260 square miles. Naomi Guthrie, Waymo’s user experience researcher, says Waymo will be offering its self-driving taxi service on regional service roads starting in the summer of 2023, but the new freeway service could cut the travel time for a robot taxi from San Francisco to Mountain View in half.
“Freeway driving is one of those things that’s very easy to learn but very hard to master,” Waymo CEO Dmitry Dolgov told reporters last week. Highways are predictable, with (mostly) defined signs and lane lines, and a limited set of vehicles and players (trucks, cars, motorcycles, trailers) that a vehicle’s software must recognize and predict. But Waymo executives say that despite a year of testing highways just for employees and guests, highway safety emergencies are relatively rare, so the team wasn’t able to gather enough real-world data to train its vehicles to operate safely there. Complicating the project was the fact that highway accidents, at high speeds, are subject to the laws of physics – and therefore more likely to cause disability or death.
To prepare for highways, Waymo executives say, engineers supplemented real-world driving and training data with data collected in private, closed courses and data generated in simulations. Two on-board computers help make the system “redundant,” meaning the cars have computer backup if something goes wrong. These vehicles are trained to exit highways in an emergency, but they can. Waymo executives also say they will work with law enforcement and emergency responders, including the Highway Patrol, to develop procedures for vehicles and riders stuck on highway shoulders, which kill hundreds of Americans each year.
