Re/code has a neat article that breaks down autonomous cars by component, and examines which suppliers are building which pieces.
The body, the eyes, and the brains all involve a collection of different, sometimes competing, suppliers. The suppliers range from longtime auto industry giants like Continental and Bosch, to niche firms like Mobileye, to computer hardware firms like Nvidia.
Recommended.
Originally published at www.davidincalifornia.com on October 27, 2015.
As cars go driverless, will auto repair shops go bankrupt?
It’s not an obvious corollary, but the argument is there. Particularly if private vehicle ownership declines and we all begin to rent rides from Uber-like fleet owners, the viability of independent auto shops could be threatened.
But I’m not ready to go there quite yet. My local independent auto shop often has US Postal Service vehicles inside, proving that some fleet owners still outsource their maintenance to small operations.
Of course, USPS is a government operation, and perhaps privately-owned fleets are maintained differently. How do Hertz or FedEx manage maintenance and repair for their fleets?
Originally published at www.davidincalifornia.com on October 26, 2015.
I suppose so, but I doubt it. Progress marches forward.
The Telegraph tries to envision how this nightmare would occur and isn’t very convincing.
I walk everywhere in the city now because traffic is worse than ever. Who needs parking spaces when you can send your driverless car around the block while you do your shopping? And commuting distances have doubled, with vehicle occupants happy to zone out in their pleasure booths, watching movies or working themselves to death, seemingly losing track of time.
The upshot here seems to be that getting more flexibility in parking and commuting is actually a bad thing.
I suppose if the “drive in circles instead of park” problem became bad enough, it could be a real drag. But driving in circles is not free (whether the car is gasoline or electric). And this assumes we all have our own driverless cars that need to idle while we run our errands — a future that seems increasingly unlikely in the age of Uber.
I imagine there are some potential driverless dystopias out there, but let’s get a little more creative. And plausible.
Originally published at www.davidincalifornia.com on October 25, 2015.
Autonomous vehicles could shrink U.S. personal auto insurance sector by 60%:
“Autonomous vehicles are poised to completely transform the auto insurance industry, and underlying market forces, including technology enablement, consumer adoption, and regulatory permission, are already aligning to enable mass change,” said Jerry Albright, principal in KPMG’s Actuarial and Insurance Risk practice. “The risk profile of vehicles is changing daily, and the subsequent drop in industry loss costs would reduce the size of the auto insurance market, trigger consolidation in the personal lines space, attract new competitors, and force dramatic operational changes within carriers.”
Originally published at www.davidincalifornia.com on October 23, 2015.
In place of famous Silicon Valley names, a network of established automotive electronics and software suppliers is developing advanced driver-assist systems, connectivity and autonomous features that are customized to OEM specs instead of being hand-me-downs from the consumer electronics industry.
WardsAuto argues that Silicon Valley companies aren’t going to control the driverless car market, both because car manufacturers are on their toes, and because the supply chains are so important and intricate.
I think kind of probably right, but misses the point.
Looking at the mobile phone industry — which is a somewhat analogous — Apple and Google control the operating systems, and Silicon Valley start-ups come up with the cool applications.
The layer in the middle — actually constructing the phones — is controlled by OEMs and suppliers in Asia.
The middle layer is the most commodified and least profitable layer of the value chain, though.
Originally published at www.davidincalifornia.com on October 22, 2015.
Videos are already showing up on YouTube, demonstrating Tesla’s autopilot feature going awry:
Perhaps this is a #slatepitch, but I think this is a good indication of how much faster Tesla is going to advance beyond other driver-less car manufacturers.
Nobody else is getting their product debugged at this scale.
Originally published at www.davidincalifornia.com on October 21, 2015.
Appropriately enough, yesterday Stanford announced a self-driving Back to the Future super-car that turns donuts.
Pretty awesome.
Originally published at www.davidincalifornia.com on October 21, 2015.
Yesterday I riffed a little bit on what insurance might look like in a driver-less car world.
Today, MarketWatch carries a press release from a report hypothesizing that driver-less cars may kill the auto insurance industry.
Actually, they don’t come out and say that in so many words. Instead, they write:
As the risk of accidents will fall drastically with the advent of autonomous vehicles, the insurance premium to cover that risk too will drop significantly. Nevertheless, original equipment manufacturers (OEMs) and suppliers will increase insurance spend to cover their share of product liability risk, thereby offsetting the shrinkage in consumer-driven insurance revenues.
With OEMs and tier 1 suppliers looking to ensure fool proof product safety, methods to access risk and certify the product will assume greater importance. The traditional method of underwriting that uses historic data will take a back-seat, paving the way for a new breed of underwriters capable of evaluating driving algorithms and assigning a relevant risk priority number.
I think the translation there is that disruption is coming to the auto insurance industry.
“If you could come up with anything involved in driving that cut accidents by 30 percent, 40 percent, 50 percent, that would be wonderful,” he said at a conference in March. “But we would not be holding a party at our insurance company.”
Originally published at www.davidincalifornia.com on October 20, 2015.
A common objection (or at least concern) regarding self-driving cars is the issue of liability. If my self-driving car hits another car, or person, who pays?
This seems to me like any easy problem to solve — it’s just a finance question — as opposed to the actual technology, which is a math and software and robotics question.
Presumably, for enough money, somebody will take on liability.
Nonetheless, it’s an interesting idea through which to work.
An obvious analogy is to commercial airplanes, and in that case the passenger bears no responsibility. Although I’m not knowledgeable of this field, I would imagine liability is shared between the airline, the airplane manufacturer, and the pilots, although all of these groups probably have some form of insurance or re-insurance.
This might even be a near-perfect analogy, if self-driving cars usher in an Uber-centric future in which individuals subscribe to car services, rather than own individual cars.
In a future more like the current world, though, in which individuals own cars, it’s a little less clear who will accept liability.
Volvo and other manufacturers have pre-emptively claimed liability, but since they aren’t actually selling autonomous vehicles yet, it remains to be seen how that will stand up in practice.
My guess is that, at least for a while, drivers will need to maintain personal auto insurance, if only because cars will continue to have manual override features. But I suspect auto manufacturers will be quick to take on as much liability as they can, if only to protect their brands and sell more cars.
This could lead to something of a boom in personal injury law, as lawyers develop more appetite to finance cases against defendants (car companies) with deep pockets, and juries feel less guilty delivering big damages awards against giant corporations. And, that, in turn will be a boon for the insurance industry.
Originally published at www.davidincalifornia.com on October 19, 2015.
Last night, Tesla rolled out a software update to its Model S sedans. The update enables autopilot functionality on highways, as well as auto-parking in some situations.
This is a great step forward for Tesla, and a long-term step forward for consumers, as the new data will give Tesla a huge leg up in race for self-driving cars. Nothing like tens of thousands of users to work out the bugs in your software.
In the short-term, I think it’s probably only a small win for consumers. Tesla has been vocal that any accidents caused by the software will be the responsibility of the driver. This means drivers will have to stay engaged throughout the trip, scaling back a lot of the benefits to a self-driving car.
But, as Tesla gets more and more data and becomes more and more confident in its software, I imagine the day that Tesla takes responsibility for its autopilot is coming fast.
Originally published at www.davidincalifornia.com on October 15, 2015.
David Silver 