Tesla’s robotaxi service, touted by Elon Musk as the future of his flagging electric-car maker, launched in the company’s home city of Austin, Texas, on Sunday with about 10 vehicles and a human safety driver on board amid regulatory scrutiny of its self-driving technology.
Shares in Tesla have risen about 50 percent from this year’s low in early April, with investors hopeful the autonomous ride-hailing service will help revive a company that has suffered declining sales and a consumer backlash against Musk’s political activism.
Despite the hype surrounding Tesla’s robotaxi, the launch—with a company employee seated in the passenger side for safety while leaving the driver’s seat empty—was low-key, and the initial service was open only to a select group of social media influencers.
Shortly before the launch, Musk said on social media that the robotaxi service would begin “with customers paying a $4.20 flat fee.”
According to Musk, who has stepped back from his US government role to focus on the electric-car maker and the robotaxi, the self-driving Tesla Model Y vehicles will only operate in limited areas, avoid challenging intersections, and have teleoperators who can intervene if problems arise.
The limited launch comes as the National Highway Traffic Safety Administration continues to carry out multiple investigations into Musk’s claims about the capabilities of Tesla’s autopilot and “full self-driving” systems. Despite its name, the full self-driving system still requires humans to sit in the driver’s seat and pay full attention—unlike Google’s Waymo taxis.
The NHTSA wrote a letter in early May seeking additional information about technologies that would be used in Tesla’s robotaxi service. The regulator said it had received Tesla’s response and was reviewing its content.
Musk said in a social media post this month that the company was being “super paranoid” about safety. But he has also claimed there would be 1,000 robotaxis “in a few months,” and that the service would expand to cities such as San Francisco and Los Angeles.
I let FSD drive ~most~ of the time around town on the routes I use daily as I love testing new tech and I find it interesting. It works well enough now for what I want it for, but it definitely needs to be monitored. I probably intervene once a week or I just choose to not use it on very specific sections of road during rush hour. Which is a vast improvement over last year, but it still feels half baked.
Oddly I find all the complicated traffic moves / intersections etc. it has no issues with. I've forced it into hard situations just to see what it would do, obviously with hands on the wheel ready to take over. The problems I see with it regularly are really small stuff or just lack of forethought / knowledge of the area that a human would have. Things like the left lane is moving slower let me get into the right lane, well the right lane is about to be a standstill in 1000 yards, but it doesn't know this, it sees that traffic "currently" is moving faster in the other lane so it chooses to try and use that lane, which actually ends up slowing the trip down, along with making it near impossible to merge back into the left lane. The only way to teach this is to have the car self-learn from its own routes just like a human driver does, because you can't teach every hour of every day for every road in the USA to every car.
In heavy traffic situations it LOVES to try and use the zipper merge lanes, which in my area of the south are considered rude to use even though it would help traffic flow. There is a road on my commute where 3 lanes reduce down into 2. About a mile from that merge people stop using the lane that disappears. If you were to ride the lane that merges in all the way to the end people will not let you merge in. Guess what the Tesla wants to do? It sees an empty lane and tons of traffic in the other two so let's take the fast route and save 10 mins! Effectively boxing itself out.
The other issues I feel are partially camera related along with no radar and improper characterization of things its sees. Football sized objects in the road it will not avoid, so anytime I see a dead animal or tree or whatever in the road I have to manually avoid it. If a mail truck is stopped on a 2-lane road half off the road, it wants to go around it, which is what everyone does, but it can't see or predict the oncoming traffic well enough to make that move, so you have to be ready to either let it or cancel the attempt based of what you see.
The camera's get blinded all the time if the car is dirty at all. My right side B pillar constantly gets blinded. It flood rains hard enough here that I've seen it get blinded in the rain, but it does see better than I expected considering what it has to work with. Not good enough for Unsupervised.
I think this year I've probably logged 5000+ miles on FSD alone, including a few 3 hour+ road trips. 90% of the time it works great, but the other 10% is the problem and the hardest to solve. Until its 100% the taxis won't work.