“Collision Avoidance Of Arbitrary Polygonal Obstacles.”
As usual... Apple tries to patent something with boatloads of prior-art...
Google's lawyers are going to have a field-day with this...
Apple is the latest player to enter the race to test out self-driving cars, alongside other big names including Tesla, Uber and Google’s Waymo. It received a permit on Friday from the California Department of Motor Vehicles to toy with autonomous vehicles on the US state's roads. The company is notoriously secretive about …
As the saying goes...
There are many ways to skin a cat
If Company B does the same thing as Company A but using a method that is NOT described in the patent then Company A's patent attourneys will tell them that they don't have a leg to stand on.
The devil is in the detail.
The article did give a link to the actual patent so you can see which means of skinning a cat they claim is new:
As you can see it is indeed as @malle-herbert said.
Vectors of where the car is.
Vectors and polyhedrons to describe the obstacle.
Bounding circles to describe what must be avoided.
Vectors to describe where the obstacle is in relation to the car.
Computer systems with memory, display units, memory and one or more processors.
"As usual... Apple tries to patent something with boatloads of prior-art..."
It seems you don't understand the patent system at all.
The reason for the existence of patents is that the patent owner gets a time limited monopoly in exchange for publication, so that others can improve on the published patent. Yes, boatloads of prior art _is supposed to be there_. And then you improve the prior art and get a patent on the improvement.
And not only that, you are allowed to come up with other ways to solve the same problem!
People who assume a patent on "Collision Avoidance Of Arbitrary Polygonal Obstacles" must have prior art, or be obvious, don't understand how patents work. You don't patent ideas, you patent a particular method of a general idea like "Collision Avoidance Of Arbitrary Polygonal Obstacles".
Of course, the lawyers always try to write patents in as general a manner as possible to cover multiple methods, so maybe a prior patent already covers the method(s) Apple's patent listed. If so, that's a question for a judge if one of those earlier patent holders sues Apple (if/when Apple comes out with a product that uses this patent, since you can only sue someone for infringement if they use one of your patents, not if they try to patent what you think is the same thing)
Frank, same is true no matter who comes up with the car -- or any product or service, any industry, cars, computers, TVs, music.
Company comes up with a good one, people want it, and they make a boatload of cash and it puts them ahead of their competitors.
But a question: What does it mean for a car to 'handle well' when nobody is handling it, when it is being handled by a computer? Do you mean a car that 'rides well'?
Or maybe it is a difference between UK and North American English?
I still don't think Apple intends to build cars as the margin isn't really there. They're more likely looking to expand their patent portfolio so they can partner with or license it to companies already in the industry. Of course I could be wrong since they traditionally like to have full control over the hardware. Perhaps they see a less competitive future for the auto market where the automobile market is split like the mobile phone market. Sounds a bit too dystopian for me.
Alright Siri, no damn it! Turn left!
From what I have read many companies are fiddling with autonomous vehicles including automakers. Given the potential rewards for solving the computing issues are great it is not surprising that Fruit is interested. For Fruit to make a pile of cash it is not necessary for them to have a car but to have the technology for sale/license to the automakers. A variation of how Slurp really makes money from Android.
"For Fruit to make a pile of cash it is not necessary for them to have a car but to have the technology for sale/license to the automakers."
"Nice self driving car you've got there. Now if you license our tech, you you won't have to spend years in court with us before you can sell it."
I'm not sure licensing automotive software will be particularly lucrative for very long, since there is likely to be a lot of competition and not much to distinguish them. Why should you prefer Apple's self driving software over Google's or Bosch's or Ford's? It would take a while for safety differences to shake out, and that would be evolving over time as software gets updated. Assuming they all take you from point A to point B in much the same way, what difference does it make?
Now maybe the Apple's UI is nicer, but they wouldn't need a permit to test self driving cars if they only intended to put a pretty UI over someone else's self driving software.
Think about it in terms of phones. If Apple had built iOS and licensed it to others, how much could Apple charge to license a copy of iOS when Google gives Android away for free? They wouldn't be worth 3/4 of a trillion dollars today, that's for sure!
If they want to make real money, they have to "sell the hardware", which means selling cars that like the phones are intended to derive benefit from having the hardware and software designed to work well together, rather than being glued together from different sources as is in the case in the Android and PC markets.
However, Apple can develop the car behind closed doors and never needs to let it see the light of day (at least not in the US where the press might see it, they could open a subsidiary in Europe somewhere with some weird name and make people think it was going to be just another boutique automaker) They can use existing cars - I think their permit application says they'll be testing using Volvos - to develop the software, and only move to testing with their own cars when they have all the bugs worked out of it using the testbed cars. That will be several years away, minimum.
Or just hold sufficiently vague and general patents to convince the lawyers of those actually making the cars that it's cheaper to pay a licence fee than to go to court and prove that the patents don't apply or are invalid.
Lost of companies make plenty of money that way, without any of the risks involved in having real product in the field.
Yes, various trolls will file the patents of the form "X, but in an autonomous car" variety where X is something that already existed in manually driven cars or in other forms of computer. They'll never spend a dollar on R&D for actual products, let alone try to sell any. That's just built in friction you have to assume, unless/until patent laws are reformed. It won't change the profitability equation for those who do field products in the market, whether self-driving software only or cars themselves.
> Why should you prefer Apple's self driving software over Google's or Bosch's or Ford's? ... Now maybe the Apple's UI is nicer, but they wouldn't need a permit to test self driving cars if they only intended to put a pretty UI over someone else's self driving software.
The UI won't stop with the car. Indeed, an iPhone (with access to messages and calender appointments) would have a fair idea of where the user wants to go before they even get in the car. The UI would have provision for organising travel for different family members (if we're on the assumption that the private car ownership model will still hold). The UI would be linked to related functions, such as the iPhone waking the user with an alarm.
Pretty much if you want to put your guidance software in someone else's car and test it and tweak it, you're going to need a permit to build and test self-driving cars.
The fact is that the software being tested is going to have huge safety implications. It is going to change the safety aspects of the car.
Consider aircraft. You install an untested unapproved piece of electronics that does pretty much anything, let alone autonomous guidance, and you've got to treat and label your aircraft as "Experimental", get approval, and not permit passengers.
Aircraft or car, it is a public safety issue when you're testing safety features in public. Testing guidance control systems is going to require a permit even if the car used is a standard production model from Ford or GM.
I was about to cast a vote in favour of self-driving vehicles - as long as EVERY vehicle was self-driving. But then remembered (as others here will) "driving" beaten up dodgems* at [your choice of rundown English coastal resort's even more rundown funfair] when the brakes didn't work, the steering wheel barely made a difference and the accelerator was worse than useless. I changed my mind.
On reflection I prefer a human brain - with all its wonderful defects - in charge.
*Hi, kids. Imagine Grand Theft Auto played out for real** on a basketball court.
** No, not virtually real. Really real.
Biting the hand that feeds IT © 1998–2019