They seem to be doing alright for a technology which is supposedly "going nowhere". (Yes I know these cars sometimes stop, but that's better than having them crash).
Cruise did provide video evidence, mentioned in this article, though the contents were not described. A NY Times article on the same subject described the content of the video here (gift link):https://www.nytimes.com/2023/09/02/technology/driverless-car...
> The distance between "sort of works" and "works" for AI is considerable. Not infinite.
> Today you can take a driverless cab in San Francisco [...]
From the outside, it sure does look like driverless is still firmly at "sort of works":
"After California regulators approved the expansion of driverless taxi services in San Francisco earlier this month, it took only a little more than 24 hours for a series of events to begin that seemed to justify the taxis’ detractors.
The day after the vote, 10 autonomous vehicles operated by Cruise, a subsidiary of General Motors, abruptly stopped functioning in the middle of a busy street in the North Beach neighborhood of San Francisco. Posts to social media showed the cars jammed up, their hazard lights flashing, blocking traffic for 15 minutes.
A few days later, another Cruise vehicle drove into a paving project in the Western Addition and got stuck in freshly poured concrete.
And then last week, a Cruise car collided with a fire truck in the city, injuring a passenger in the car.
So it was that last Friday Cruise agreed to a request from the California Department of Motor Vehicles to cut in half the number of vehicles it operated in San Francisco, even though regulatory approval for more remained in place. The company, which has had 400 driverless vehicles operating in the city, will now have no more than 50 cars running during the day and 150 at night."[0]
They’re operating Taxi services in San Francisco. A city that doesn’t experience any real-world weather, with an area of like 50 square miles where speeds generally never exceed 25MPH. They also have humans watching cameras that take over when the self driving breaks down.
It’s a completely different problem space, like claiming someone built a train and therefore they can easily build self driving cars since they are both “driverless”.
Yes. Cruise has been "fake it til you make it" since their original demo video. Now "all vehicles shown on road are renders" in fine print.
Here are the CA DMV's autonomous vehicle accident reports for 2019.[1] Cruise has a lot of them. Mostly being rear-ended by a human-driven car. A surprisingly large number are with the vehicle being driven manually, or right after the human driver took over. True driverless still seems a long way off given those results.
I've seen Cruise cars in SF, most recently making a left turn off Union St in manual mode.
GM's first try at self-driving, Firebird III in 1958, was far cooler than this mini-bus thing.
The mini-bus area seems a good place to start with self-driving. I was once thinking airport parking lot shuttles. Controlled environment, low speed - that could work. But the financial numbers don't work out. You get rid of the driver cost, but you add sizable engineering cost and vehicle cost, and you don't sell many vehicles that way. Local Motors and some other startups have been struggling quietly in that space for years now. Local Motors' Olli has three live installations, with about six vehicles total, all very recent. A college campus, a casino area, and a dedicated track at an industrial park are the public installations. Demos, basically. The thing is so slow that walking might be faster, and any bicycle or electric scooter would be much faster. Not fast enough to get people from Economy Parking to the terminal.
Navya, from France, seems to be further along. They provided the self-driving shuttles for the Las Vegas strip. They have a number of live mini-bus installations in Europe. They've announced an autonomous cab, but it is not deployed yet.
This is still something that can only be done if someone is willing to finance a money-loser. However, this is perhaps the best time in history for that.
> Cruise and Waymo have fully autonomous cars deployed in SF and are scaling up.
Is that it? SF only? After billions invested? There is a laundry list of those that tried and failed especially with burning an insurmountable amount of VC money even with billions of their own money.
Lyft: Scrapped and sold their self-driving project. [0]
Uber: Scrapped their robot-taxi project and sold it off. [1]
Zoox: Once valued at $3BN, acquired by Amazon for $1BN after nearly going bankrupt and is still using specialised cars for self driving only in SF. [2]
Cruise: Acquired by GM and still using specialised cars for self driving in SF [3]
Drive.ai: Ran out of money and almost bankrupt and acquired by Apple. [4] No where to be found on the roads.
Waymo: Same situation as Cruise, but Google keeping them alive.
Comma has lasted longer than these over-valued companies and is already in lots of consumer grade vehicles beyond SF today and not in specialised cars and taxis unlike Cruise and Waymo who are still stuck in SF [5].
Is this what Garry Tan's video (The Truth and Lies About Driverless Cars in SF [1]) was all about? I find him to be pretty credible in general, and he makes a strong case that there are some folks within the city government who are misrepresenting the safety record of self-driving cars. I don't know much about the situation, and I'm glad this testing isn't happening where I live (we do have tons of testing on the peninsula, but it's with humans in the vehicle). But I think it's important for companies and regulators to be honest about what is happening, whichever way the facts come out.
> The California DMV, the agency that regulates autonomous vehicle testing in the state, issued Cruise a permit in October that allows the company to test five autonomous vehicles without a driver behind the wheel on specified streets within San Francisco. Cruise has had a permit to test autonomous vehicles with safety drivers behind the wheel since 2015.
I saw their cars a lot while I lived in San Francisco. But back then they had safety drivers and such.
Cruise operates in a very few markets for which their specialized self-driving cars have been designed for live testing. They have received special attention and care to have extremely up-to-date local information and are closely managed by operators. It is indeed very encouraging that Cruise is operating this local taxi and soon a local shuttle with almost completely driverless capabilities..
But keep in mind, this doesn't mean they work anywhere else in the country or world. What will happen to them in bad weather or some damage to the vehicle or a flat tire or navigating a scene cordoned off by police, or a parade? What happens when they do not have the meticulously groomed information? Likely, Cruise will manually tell the vehicle to avoid the area, shut the vehicle down, or take over control remotely.
Self-Driving cars are designated into technology levels 0 through 5[0], from no capabilities, to "simple" lane-keeping and/or cruise control, to autonomous with caveats
I don't know exactly which technologies Cruise is using to achieve their mostly autonomous taxi, as that information is worth the billions invested in the company, however from what I have been able to gather, they still use a remote safety operator in some instances. The safety operator can take control of the car if a safety problem presents.
The cruise cars in SF are essentially bespoke for their market. If you take it to Wichita or Maryland, these things will likely not be able to manage the environment. The self-driving capabilities are still unable to reach SAE level 5, fully autonomous driving in all conditions.
Its easy to look at Cruise and conclude, "We've arrived!" but I would still caution restraint, as it has been possible to operate autonomous vehicles with extensive constraints for many years, such as autonomous rail.
"It seems odd that self driving cars have been skipping this step." is an opinion referencing well-known current events. You can't expect everyone commenting on this article to re-reference the saga of Cruise in San Francisco; especially when it's frequently covered on the front page of Hacker News.
Furthermore, you should know that a Hacker News discussion is not Wikipedia. These discussions are not the kind of things where every off-the-cuff statement needs to be backed up with references.
I think the success of driverless cars will depend on the experience. If the cars are incredibly slow/cautious and you look like a segway riding dork but just inside a glass cage.. then I don't think it works.
Get the user experience right. The youtube videos of tesla drivers drinking coffee and reading the paper at high speeds looked like great fun to me.
Nah - this happened yesterday (the day before your post). It's unusual for media to be allowed inside autonomous cars, and the conditions are usually controlled. The conditions in this demo were fairly loose (difficult public streets in San Francisco) and also involved relatively little intervention by safety drivers. (Uber did a demo in SF once but it involved very frequent human takeovers. Alternatively, Waymo had a recent demo on a closed course (I think the one at Castle airforce base) and others in Arizona suburbs where conditions are less challenging.)
Fwiw as a human driver I find driving in San Francisco to be pretty challenging/stressful.
They seem to be doing alright for a technology which is supposedly "going nowhere". (Yes I know these cars sometimes stop, but that's better than having them crash).
reply