Waymo just doubled the number of cities they operate with nobody in the vehicle. How do the various teams decide where to expand next, and will it be your town?
Waymo ’s recent expansion to freeways and most of the San Francisco Peninsula has prompted many to ask when they will get robotaxi service? Waymo has announced intentions for a number of towns. They just started unsupervised operation inWashington, San Diego, Detroit, Nashville and Las Vegas Tesla has no working robotaxi yet, but keeps saying it will offer several soon, and Elon Musk imagines they will grow even faster than Waymo once they get it working.
In China, Baidu is doing almost as many rides as Waymo and the 3 core companies there support a number of cities, plus there are plans for the UAE and Europe. Even in the Bay Area, there are major areas not served, including most of San Jose, the entire East and North bay regions, and places like Santa Cruz, Pacifica and the wealthy hill towns. Still lots of L.A. to be served as well. So how do Waymo and other robotaxi companies decide where to deploy, and what limits them and gets in the way of you getting service at your location?Uber was found in 2009. 16 years later, while Uber is in most major towns in the USA, there are still many smaller towns without it, and it took some time to get there. For Uber to spread to a new town only requires sending a team of 2 to do localization and local relations and recruit drivers Minimal capital must be spent. Expanding a robotaxi service is capital and labour intensive. Though companies like Alphabet have some of the world’s greatest capital reserves, even they can’t spend money that fast. There’s just a ton of stuff to do and build, as well as making and paying for the vehicles and depots. I outlined that in my article on infrastructure. One thing needed for every area is to both handle all the local situations, but certify that you have done so. Even if you think you have a very general car that can handle just about every type of road, you’re not going to put customers in a car going down a street you’ve never put a car down before. Usually there will be no surprises, but sometimes there will be and that could burn you. All this takes time, money and management bandwidth. That management bandwidth, just the work to organize all these tasks for a new city and fund them, is a big bottleneck. Since there’s a limit as to how fast new territory can be conquered, companies will pick the territory that makes the most sense. Either because it’s lucrative, or it’s an important competitive foothold, or it’s easy. Right now though, none of the companies make money or expect to make money for some time. Their R&D and other non-operating costs are so high that cost and revenue won’t be nearly as important as they will be in the future. Indeed, no company has really settled on what sort of pricing they will have; for now they just duplicate Uber style taxi pricing, even though none will imagine doing that long term. They want pricing models that change the game, and create whole new markets and uses. The global Uber market is nice but not large enough to justify this effort.One key market is car replacement: Convincing customers to give up owning one or more of their cars, and using robotaxi service to replace most of what that car did. That’s easiest in dense cities and with people who don’t go too far afield. You need to supplement this with human driven rides, car sharing and car rental, and other modes like public transit. Uber style pricing of $2-$3 per mile won’t work. It’s much more than the 50-70 cents/mile cost of car ownership for those who drive near or above the 10,000 mile/year average. Even if you can price your robotaxi much less riders will resist it. We don’t quite think rationally about this. We tend to think of the cost of a ride in our own car as only costing the price of energy with perhaps a few other things, we don’t factor in depreciation, insurance and some maintenance. For the 100 mile round trip from Silicon Valley to San Francisco, drivers may think the primary cost is just the $15 of gasoline or $4 of electricity, not the $70 “all-in” cost including depreciation. Asked to pay $70 for such a trip just for a party or a dinner, they would rarely take it. So to get products that people use like private cars, and thus are satisfied by in the same way, you need to price so they will enjoy it--while getting revenue.While revenue is the long term goal, all companies are still very much in the learning phase. Initially, companies tested where it was easy but they know the big money requires driving in snow, in complex crowded cities, and where governments are unfriendly. Teams will deliberately tackle places that represent new classes of territory that will be large, vital classes in the future. However, they will try to do one problem at a time, and focus on it. We’ve already seen Waymoin Detroit and deep complexity in New York and London. In London and Tokyo they will learn left-side driving and different international regulatory and pubic opinion issues. Since nobody can deploy everywhere, there will be some desire to just be first over in virgin territory. Why compete when you can be the only player? Some cities, though, such as the big prizes of New York, London, San Francisco and Tokyo can’t be ceded, so you will see competition. Some places just won’t have high demand. Suburbs with high car ownership and free parking don’t get a lot of taxi/Uber business and they won’t get much robotaxi business either, until they can compete with car ownership on price. To do car replacement, however, you need to get people everywhere, which means that the suburbs around a prime urban territory will get covered. It’s no accident the first countries for deployment with the USA and China, and in the USA while states with friendly regulatory environments like Texas, Arizona, Florida were first. California has been fairly friendly but it’s the home of Waymo, Cruise, Zoox and formerly of Tesla, as well as the richest place in the world that’s not a nation. Europe, on the other hand is a regulatory quagmire, though some European teams are pushing there because it is home. I’m often asked by cities and other regions how they can get a robotaxi company to come there. The problem is, the leading teams have many places asking them this. It will take a lot to get them to change their priorities, though one way is to clear regulatory hurdles and let them test and learn something important to their R&D plan. One thing that won’t happen is “everywhere at once.” Even if you can build a robotaxi that can drive on almost any street, that’s not the same as certifying it safe enough to bet customer lives on streets you’ve never tested on, or dealing with local regulators you haven’t done a lot of talking to. It’s a lot more work than just bringing Uber to a new city, and that was harder than they thought. It takes a lot of management effort, hiring and capital to bring a robotaxi fleet. Tesla’s plan that ordinary private car customers will hire out their own cars in an Uber-like network faces a lot of challenges. Tesla has gained a lot of generality because owners of private Teslas drive on all roads and upload some data, but this is not enough. Tesla recently announced they would deploy 1,000 non-autonomous robotaxis with safety drivers in California. That’s not to make money, it’s to learn, and learn what you only can with a professional safety driver, while riders subsidize the testing with their fare payments. Robotaxis and employee driven cars will do more than private owners will, particularly the complex situations of Pick-up and Drop-off and better documentation of interventions. Even so, take heart. Waymo’s accelerating expansion indicates they are ready to move from pure R&D to scaling. Tesla hopes to get a working robotaxi and use their position as an auto OEM to also scale quickly. Outside the USA, the Chinese players will be eager to go places they can win, since they are locked out of the USA. They all want to come to your city as soon as they can. More rural places will have much longer to wait. It’s hard to do a taxi service or car replacement for rural locations. But watch the streets.What about just getting your own personal robocar? That’s harder than making a robotaxi, but some companies like Tensor are working at it on the luxury end, and Tesla has always wanted to do it. Waymo and Toyota have said they are partnering to eventually do it. In a town that has robotaxis from a company, it’s not too hard to make a car that can serve that town. Serving a town without robotaxis requires a lot of work, unfortunately for limited return, because you only meet the needs of a few people in that town who buy your particular car. Private robocars will handle the freeway system though, and major arterials in most towns, but you will probably need to drive the small streets in towns without a robotaxi service. That will still be pretty valuable, and can come sooner. Mercedes sells a car now that will drive the freeways in California and Nevada, but at first it only did traffic jams, and now it’s moving to doing only the right lane at 50mph when following a truck. But it’s a start.
United States Latest News, United States Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
Dallas getting autonomous Waymo robotaxis in ‘coming weeks’Soon, according to a release, Waymo will begin fully autonomous operations in several new cities, including Dallas. Autonomous operations will start in Miami...
Read more »
Zoox will let public riders use its robotaxis in San FranciscoZoox will offer robotaxi trips to people on its public waitlist in San Francisco, the Amazon-owned company announced.
Read more »
Amazon's Zoox robotaxis service to give free rides in San Francisco as its expansion acceleratesAmazon’s Zoox will start giving free robotaxi rides through parts of San Francisco as it accelerates its attempt to challenge Waymo’s early lead in the race to transport passengers cities in self-driving vehicles.
Read more »
Amazon's Zoox robotaxis service to give free rides in San Francisco as its expansion acceleratesAmazon’s Zoox will start giving free robotaxi rides through parts of San Francisco as it accelerates its attempt to challenge Waymo’s early lead in the race to transport passengers cities in self-driving vehicles.
Read more »
Amazon's Zoox robotaxis service to give free rides in San Francisco as its expansion acceleratesAmazon’s Zoox will start giving free robotaxi rides through parts of San Francisco as it accelerates its attempt to challenge Waymo’s early lead in the race to transport passengers cities in self-driving vehicles.
Read more »
Amazon’s Zoox robotaxis service to give free rides in San Francisco as its expansion acceleratesAmazon’s Zoox will start giving free robotaxi rides through parts of San Francisco as it accelerates its attempt to challenge Waymo’s early lead in the race to transport passengers in s…
Read more »