At the Tesla Cyber Rodeo, Elon Musk said, “There’s going to be a dedicated robotaxi that is going to look quite futuristic-looking.” Then, in the recent earnings call, he briefly discussed the “futuristic” robotaxi. He declined to give details, but told us that they plan to do a robotaxi event of some kind next year. He also added, “We are aiming for volume production in 2024.”
That’s all that we really know for sure right now, unfortunately, but there’s speculation that it could look like this “People Mover” sketch that is associated with Boring tunnels.
I saw a lot of Tesla owners, investors, and fans questioning the necessity of a specialized robotaxi model. After all, Teslas going back several years are supposed to have the hardware necessary to act as autonomous vehicles. With all of those vehicles, especially the leased Model 3 vehicles that have no purchase option at end of lease (and the same with every other Tesla model going forward), Tesla could easily come up with a bunch of taxis without needing to build vehicles just for that purpose, they’re saying.
But, in my experience doing rideshare driving in 2018–2019, I can tell readers a few things that make a special robotaxi model a necessity.
Carrying Strangers Around Is Hard On A Car
After doing about a year of full-time rideshare work, my Nissan LEAF had seen better days. During that time, I put on about 40,000 miles and the car wasn’t left in as good a condition as other 2018 LEAFs that weren’t used for this.
Vehicle wear and tear is a big one. Sure, an EV needs less maintenance, but that doesn’t mean it’s never going to wear out. I put the battery pack through over 300 DC fast charging sessions, and because Nissan didn’t put a cooling system on the LEAF, and because this year included a summer in Phoenix, Arizona, the battery lost over 10% of its original range. Tesla vehicles aren’t going to see this kind of degradation this quickly, but we also have to keep in mind that a robotaxi doesn’t have to stop to eat, sleep, or have a family life.
It’s reasonable to assume that a robotaxi would probably drive 80,000 or more miles per year, and possibly well over 100,000.
There’s more mechanical wear and tear outside the batteries and drive units. CV axles, suspensions, tires and wheels, bearings, and many other things every car (electric or gas) have will go through accelerated wear as a robotaxi. Service that a normal vehicle only needs every few years might happen every few months.
Then, there’s the passenger-facing systems. My LEAF has a worn out power door lock in the door that passengers got into the most. The seats? They don’t look too nice, and even if they were leather, they’d probably be more worn than average. I have a USB charging port that keeps trying to come out from people plugging and unplugging. I have two door handles that are trying to come off. I had to replace the 12-volt battery after only two years. The cargo area? It has a lot of scratches from people loading and unloading suitcases at Sky Harbor.
So, in reality, the car is still fairly new, but it’s unusually rough. And, I took breaks because I’m a human and need sleep. And, I was there to kick people out who weren’t behaving (one violent passenger only left because he was afraid of getting shot), hand drunk people a bucket to puke in, and do other things to monitor and take care of the car.
Let’s talk about puke for a minute. Even in a car with leather seats (I’ve done rideshare in a Chevy Volt with leather seats), it’s a huge pain to clean up and get rid of the smell. Carpets, seat belts, seat belt clickers, the spaces between cushions, door handles, and sometimes even the ceiling needed thorough cleaning.
Building A Car That Can Handle All This
Taking stock Model 3s and putting them through all of this abuse could end up being expensive. There’d be an unusual amount of repair work just to keep a car on the road and acceptable to passengers for 2–3 years. But, Tesla could take care of all this with either a specialized model or by upfitting/retrofitting used Teslas that come off lease.
The battery systems will probably need some minor upgrading for retrofitting taxis, perhaps by upgrading the cooling system a bit. For a specialized models, it would be a good idea to go with a battery chemistry like LiFePO4 or LFP (the Stans often call these “iron” batteries, but you’re going nowhere without the lithium and phosphate), because they’d degrade a lot less under this kind of abuse. They couldn’t go as fast or as far on a charge, but they’d work fine for a taxi and last a lot longer. They may even just swap packs to LiFePo if they wanted to upgrade an existing off-lease Model 3.
Upgrading other things, like the suspension, would be a great idea. For example, upgrading the shocks to something that will last longer would be a good idea. I’d leave this up to the professional engineers, but it would be a good idea to work on upgrading everything to last longer and stand up to more abuse.
The interior could also be simplified and built more for “hosing out,” like a police car. This would probably look a little less luxurious, but it could be done tastefully and in a way that’s aesthetically pleasing with some effort. When the inevitable puke, pee, and other bodily fluid mishaps occur, whoever cleans these things out would have a lot less trouble if they could spray the whole thing out and then dry it. Upgraded door handles, plugs, and anything else passengers use would also go a long way.
One other thing? There had better be some sort of in-car monitoring. People aren’t going to be happy at all if they summon a robotaxi and it shows up with a mess inside of it or some freshly-broken thing. I don’t know if this could be automated, but it might be a good idea to have the cars stop in a few times a day for an inspection and cleanup.