Driverless Cars Need Smarter Roads: A Tale From San FranciscoAs Jonathan Bartlett notes, the recent Frisco foul-up shows the need for roads adapted to include self-driving cars
The future was here, briefly at least. The driverless cars of GM’s autonomous driving unit, Cruise, started charging fares early last month in a limited area in San Francisco. Google’s Waymo also operates driverless cars in Frisco but hasn’t yet started charging fares. With the regulators and the tech media, it certainly seemed like all systems were go:
The era of commercial autonomous robotaxi service is here — Cruise officially became the first company to offer fared rides to the general public in a major city as of late Wednesday. The milestone comes after Cruise received official approval from the California Public Utilities Commission in early June to operate driverless in a commercial capacity.
Initially, Cruise’s driverless autonomous offering will operate only between 10 p.m. and 6 a.m., and only on designated streets in the city.Darrell Etherington, “Cruise’s driverless autonomous cars start giving rides to paying passengers” at TechCrunch (June 24, 2022)
UPDATE: As of last night, fared rides are now rolling out to our customers in SF.
If you’re waiting to take your first driverless ride, we’re inviting more people into our AVs each week, so sit tight— it’ll be worth it! https://t.co/UpjuQ9K81W pic.twitter.com/CwkD1LftnV
— cruise (@Cruise) June 23, 2022
But then, less than a week later,
More than a half dozen Cruise robotaxis stopped operating and sat in a street in San Francisco late Tuesday night, blocking traffic for a couple of hours until employees arrived and manually moved the autonomous vehicles…
“We had an issue earlier this week that caused some of our vehicles to cluster together,” a Cruise spokesperson said. “While it was resolved and no passengers were impacted, we apologize to anyone who was inconvenienced.” The vehicles were recovered through a combination of remote assistance and manual retrieval.Rebecca Bellan, “Cruise robotaxis blocked traffic for hours on this San Francisco street” at TechCrunch (June 30, 2022)
Some @Cruise robotaxis appeared to be stuck in SF last night at the corner of Gough St. and Fulton St.
Human ops apparently had to rescue them. Still some kinks to iron out. pic.twitter.com/eXDocjVfHU
— Taylor Ogan (@TaylorOgan) June 30, 2022
Upvoted comments on the mess at Reddit San Francisco ranged from “The first thing I say to my coworker is that they’re getting together to murder us. It was a pretty surreal event. Humans had to come and manually take the cars away. Cruise should get fined to sh*t for blocking the street off for so long. They even made it so the street sweeper couldn’t hit an entire block.” through “They’re unionizing.”
Electrek, a news site that tracks the development of electrical vehicles, provides a bit of perspective: There are only 30 Cruise vehicles and they are only allowed to operate in a 20 square-mile area in the easiest part of the city to navigate (“the Northwest corner of the city is the least complex portion, with mostly residential areas arranged into a grid of low-speed streets ”). So a convoy blockade at the intersection of Gough and Fulton Streets within a week means failing a fairly simple test.
Programmer Jonathan Bartlett, who writes frequently here at Mind Matters News about driverless cars, maintains that, for successful integration into the traffic pattern, they need roadways to be specially adapted (“virtual rails”):
● Signage to make clear that the roadway is a virtual rail, thus the cars may not have drivers
● Beacons to mark the roadways clearly at night
● A publicly available map of the roadways that is updated when the roads are updated
● Traffic laws specific to virtual rail
● Test criteria with which vehicle manufacturers must comply if their vehicles are to be approved for virtual rail
● Beacons embedded in the roads that tell the cars things like things like (a) where the road is, (b) mile markers, (c) other relevant details, possibly including road conditions, using a defined, accepted communication and behavior protocol
● Appropriate (and well-defined) exception policies for exceptional events
● Entrance/exit ramps to the self-driving roadway where the car/driver can transition safely from ordinary driving to self-driving
● An autonomous vehicle “rest stop” at the end of a virtual rail roadway where the car can be parked if the driver cannot immediately reassume control. – March 21, 2019
Bartlett adds, “These policy recommendations would transform autonomous vehicles from a “celebrity inventor” cultural statement into a practical engineering project. Self-driving roads would require quite a bit of investment capital but not significantly more than other kinds of roads.”
The current focus appears to be more short term, however. Cruise CEO Kyle Vogt says that his firm’s driverless taxis will be cheaper than regular taxis. Electrek is a bit more cautious:
So far, Cruise’s paid driverless taxi is less expensive than competing ride-hailing apps, but not by a lot. Cruise said that a trip would cost 90 cents per mile and 40 cents per minute, plus a $5 base fee. For a sample 1.3 mile trip, this would cost a total of $8.72 including taxes, whereas Uber would cost $10.41 for the same trip. So a cost reduction, sure, but not a tremendous one. Vogt claims that the cost will eventually drop to “far below” the cost of ride-hailing apps today as technology develops, but given that ride-hailing apps currently run at a loss anyway, we don’t really know how the economics of any of this will shake out in the long run.Jameson Dow, “GM Cruise takes first fares for paid driverless taxi in San Francisco” at Electrek (June 23, 2022)
Automation done by halves with short-sighted goals risks avoidable foul-ups and even tragedies.
You may also wish to read: Self-driving cars need virtual rails; the alternative is more needless fatalities. A virtual rail is essentially a road that is built expressly for driverless cars.