The Next Step in Driverless Cars

Elon Musk took a drive in one of his cars using the latest version of its full driverless software. The current software uses 300,000 lines of code to tell the cars how to turn left, how to respond to pedestrians and other vehicles, and so forth. Instead of code, the latest beta software is based on neural networks, meaning the computer has learned and will learn how to deal with various situations based on past experiences.

“It’s all nets, baby,” he said during the drive, “nothing but nets.” As I understand it, this is the approach George Hotz was taking when he was working on an his autonomous car software. Hotz tried to train his software without going through all of California’s state licensing requirements, and when the state and the National Highway Traffic Safety Administration tried to force him to comply with their regulations, he left the business. Continue reading

Getting the Most from Driverless Cars

Back in 2009, when I first suggested that urban transportation would soon be taken over by driverless cars, a friend of mine told me that he would never use a driverless car because “I get a testosterone thrill from having my hands on the wheel of a car.” I point out to him that, if cars drove themselves, we wouldn’t need open container laws and he and his girlfriend could make out in the back seat while the vehicle was in motion. His eyes bugged out and I knew I had won a convert.

What’s happening in the back of this van? Only the occupants and Waymo know for sure. Photo by Rob Pegaroro.

So I wasn’t surprised to read a report that “San Franciscans are having sex in robotaxis,” with some people doing it so often it is almost routine. Many of the vehicles have darkened windows that make it difficult to see inside, especially at night, creating a comfortable feeling of privacy. The travelers may not realize that Waymo and Cruise have cameras recording both outside and inside the vehicles, but the companies say no one looks at the videos unless there is some kind of an incident such as an accident. Continue reading

The State of Driverless Cars

Cruise founder Kyle Vogt takes OpenAI CEO Sam Altman on a virtual 75-minute drive in a driverless car through the streets of San Francisco in the video below. Cruise vehicles use LIDAR, radar, and optical sensors and are connected to to Cruise offices through the 4G cell network. The video demonstrates that autonomous vehicles are ready to enter ride hailing service, but not quite ready to go on sale to consumers.

To save time, the video has been speeded up five times so that 75 minutes are compressed into 15. During this time, the car handily deals with pedestrians, cyclists riding the wrong way, unprotected left turns, double-parked cars, and other road hazards. Vogt notes that the car’s computer not only tracks every other vehicle and pedestrian in its view, it simulates that vehicle or person’s movement and tries to predict where they are going so as to avoid any collisions. Since the accompanying map shows that the vehicle is tracking dozens of other moving objects at any given moment, this is pretty impressive. Continue reading

New Flyer Announces Level 4 Bus

Bus manufacturer New Flyer has announced that it has level 4 autonomous buses ready for sale. Level 4 means no human driver will be necessary as long as the bus says on its designated route.

New Flyer’s autonomous bus will have at least a dozen LIDAR, radar, and optical sensors. Click image to download a brochure about the bus.

The intelligence for the buses will be provided by Robotic Research, which until recently focused mainly on developing small delivery vehicles that can operate on sidewalks. Moving from tiny delivery carts to 40-foot buses is quite a leap. Continue reading

Will COVID Kill Robotaxis?

One of the victims of COVID-19 may be robotaxis and with them one path towards a future of autonomous vehicles. Before the pandemic, there were two views of how driverless cars would take over the road.

One model, which I’ll call the Waymo model but it was also endorsed by Uber, General Motors (through its Cruise subsidiary), and Ford, was that robotaxis would replace privately owned automobiles, especially in the urban areas that house 80 percent of the nation’s population. These robotaxis would rely heavily on maps, and would only work in areas that had been mapped. Since many people would be unwilling to buy a car that could only go on some roads, Waymo and other software companies planned to put them in robotaxi or ride-hailing services, at least until the entire country was mapped.

The other model, which I’ll call the Tesla model but it was also endorsed by Volvo and perhaps Volkswagen, continued to rely on private ownership of automobiles. Instead of depending on precise maps, the autonomous vehicles would rely mainly on their own sensors, which would enable them to go anywhere, even potentially off-road. To get to that point, Tesla and other companies planned to incrementally improve the on-board electronics until the computers could completely take over driving. Continue reading

One Leaves, Another Enters

Early this month, Uber announced that it has given up on its efforts to make a driverless car to replace its human-driven mobility services. A few weeks later, Reuters reported that Apple plans to have its driverless car ready by 2024(though some say it likely won’t be ready for one to four years after that).

Too much can be made of either of these reports. Uber didn’t halt its driverless car program, it merely transferred it, along with $400 million, to another company, Aurora. If and when Aurora brings Uber’s technology to fruition, Uber will apply it to its various services. The New York Times called this transaction a “fire sale” since Uber actually paid Aurora to take the program off of the ride-hailing company’s hands, but another view is that Uber is investing in the work Aurora has already done.

Apple, meanwhile, is as tight-lipped as ever about the products it may or may not bring to the market. The Reuters report is apparently based on a few statements by current or (more likely) former Apple employees. The most important statement they made is that Apple plans to have an electric car using a new battery technology, but batteries are far from the most important part of an autonomous vehicle. The real earth-shaking news would have been any indication that Apple’s autonomous system has a chance of competing with Waymo, not to mention GM and Ford, in the marketplace.

super cialis You need to take this medicine under guidance of the doctor. Norepinrephine is this hormone that works on the same. respitecaresa.org order uk viagra They also make higher quality decisions and stick to those decisions by eliminating politics and confusion among themselves and the people they lead. canadian viagra online The psychological impotence, which is simply not getting excited for some reason. viagra cost in canada All we really know is that Apple is one of 58 companies that have a California permit to test autonomous vehicles with a back-up driver at the wheel. Only five, including Waymo, Cruise (GM), and Zoox (Amazon), but not Apple, have permits to test cars without a back-up driver.

In the software business, the first is not always the winner. Google wasn’t the first internet search engine, Excel wasn’t the first spreadsheet, and Word wasn’t the first word processor. With nearly $200 billion in cash, Apple has both the opportunity and the means to become a player.

On the other hand, in the hardware business, sometimes first is best: look at iPod, iPhone, and iPad. If Apple really wants to sell cars, and not just software to carmakers, it will have to have something more than a better battery.

Tesla’s Self-Driving Beta Test

Tesla released what it calls “full self-drive beta” software to selected Tesla owners last week, and while it does not really make a Tesla into a true driverless car, it works pretty well under most conditions and provides a glimpse of what driverless cars will be like in the near future.

Tesla has taken a different approach to autonomy from other manufacturers. While Waymo, Ford, and GM driverless cars rely heavily on extremely precise maps, which means they can only be used within “geofenced” (i.e., mapped) areas, Elon Musk has criticized this approach. In technical terms, an autonomous car that relies on maps is called a level 4 vehicle while Tesla wants to go straight to level 5, meaning a vehicle that can go anywhere based on the geography that it detects with on-board sensors. Continue reading

Waymo Is Back in Business

Last March, Waymo suspended its driverless taxi operations in Arizona due to the pandemic. Now, it has not only started up again, it has opened the service to all members of the public. Previously, it had been available only to about a thousand people who had signed up for and been accepted as “early riders.”

Waymo’s taxis are geofenced, meaning they can only operate in an area that has been carefully and precisely mapped. At the moment, that includes about 50 square miles of Chandler, Mesa, and Tempe, Arizona. Technically, autonomous cars that need maps to work are classed as level 4 vehicles. Level 5 vehicles should be able to drive themselves anywhere without maps.

In response to a tweet about Waymo’s progress, Elon Musk tweeted that level 4 “gives a false sense of victory being close” and that Tesla is aiming for level 5; he claims its “new system is capable of driving in locations we never seen even once.” Continue reading

Waymo Begins Driverless Ride-Hailing–Sort Of

Waymo has been promising to start a commercial driverless ride-hailing service in the Phoenix area by the end of this year. With only a few weeks left in the year, the company announced that, starting yesterday, it will provide rides for hire over a 100-square-mile area that includes Chandler, Gilbert, Mesa, and Tempe, Arizona. The service will work like Uber or Lyft: potential users will launch an app on their smart phones, enquire about the cost of going to a particular destination, and then choose whether to hire the car.

Yet there are a couple of important caveats. First, Waymo will have a test driver behind the wheel ready to take over in case of an emergency. Second, only people who are pre-screened by Waymo will be eligible to hire a car for the service. Continue reading

Can Transit Survive Driverless Ride Hailing?

Should cities be building new transit infrastructure when driverless cars may be just around the corner? That’s the question asked by a New York Times article last Friday. The answers provided by a range of experts were far more balanced than a previous Times article about transit, which took for granted that transit was good and anyone skeptical of it was bad.

Last Friday’s report was far more broad minded. “Don’t build a light rail system now,” it quoted a venture capitalist as saying. “Please, please, please, please don’t” until we see how driverless cars “plays out.”

The article also quoted driverless car supporter Brad Templeton, who has promoted “robocars” for years, arguing that driverless cars can move people far more efficiently than forms of transit. Templeton claims that transit supporters are people who “just believe there is something pure and good about riding together, that it must be the right answer.” Continue reading