> But nonetheless, Tesla has been rumored to be working on an augmented reality system in the form of a heads-up display for the Model 3. Tesla CEO Elon Musk has been referring to the interior controls of the vehicle as “spaceship-like” and the prototypes unveiled earlier this year didn’t feature any instrument cluster.
That's reassuring. One of the things about the model 3 that bothered me was that there was only one display between the driver and passenger. It seemed very weird to have to look way off to the side to read the instruments. If they can add a heads-up display, that's a great improvement. If they can make the focal distance appear to be past the windshield, that's even better. (I've heard that it takes our eyes longer to refocus from far to near and back than it takes to physically move our eyes down and up when we glance at the dashboard.)
I expect another eventual step in the evolution of the automobile to come with the introduction of head-mounted displays so that the driver can "see through" the car without obstruction. Windows could be made smaller or go away entirely to make the car stronger and more survivable in an impact, and perhaps cheaper as well.
> so that the driver can "see through" the car without obstruction.
I'd rather have the driver "see through" the environment than the car, resembling something I long wished cars would have something to the effect of this HUD[0] from GR:AW, where the road boundaries would be delimited and highlighted, the car projected trajectory and braking/safety distance markers, as well as other vehicles and pedestrians being pinpointed, or even outlined through fog/night/objects. Obviously, nav waypoints and path-to-take trail too, and why not an indicator not too dissimilar to Forza's braking lines helper[0] that goes red when you're going too fast.
> One of the things about the model 3 that bothered me was that there was only one display between the driver and passenger.
That's been a problem with low-end Toyotas for years, starting with Scion back in 2004. Apparently Toyota wanted to save costs by having the same dash configuration for both left-hand drive and right-hand drive models, so they put the display in the middle.
> so that the driver can "see through" the car without obstruction
It would certainly mitigate the effects of very stringent rollover and SUV-centric standards, such as high belt lines and massive pillars.
I seem to recall reading about one of the German marques prototyping a system that used AR to project camera imagery onto the A-pillars to make them "see-through". It was a pretty cool effect.
I still prefer the windows, they are much less likely to malfunction than a dispaly. Can you imagine going down the highway and having your display suddenly stop working.
I hope this means that we will have new posts on Andrew's blog (Minimally Minimal). Microsoft asked him to remove nearly all of his new posts related to design thus some of his last articles (e.g. about the Xbox One S) have been online only for a week or so.
It's been wild following Andrew for so long. What's really interesting about him is that his blog gained recognition quite early and he was already featured on HN[0] before graduating college. Watching his growth has been fascinating as readers of his blog "grew up" with him.
I need to play devil's advocate here -- and hoping everyone will stop and think about this in a more practical way for a minute. This has less to do with VR/AR, and more to do with pre-existing overlap:
o What if Tesla is interested in improving sensory accuracy for its self-driving vehicle program?
o What if Tesla is interested in Hololens' pixel greying on transparent surfaces?
This seems like the most viable use of this kind of tech in a car. In an extreme case I could see something like head/eye tracking on the windshield, though that isn't in any way guaranteed or even necessarily practically useful.
Improving external sensor accuracy, or even pixel shading on transparent surfaces, however, is probably a FAR bigger overlap between Hololens and Telsa, than anything for augmented reality. Hololens has immense requirements for sensory precision and pixel blacking, and I could see why Tesla would look to an industry expert to carry over any useful architectural guidance for that.
1. Factory. Highly customised cars will require a very flexible workforce. Much easier to do by equipping each worker with a hololens
2. Inside the car - turn all the flat surfaces into displays. No headmounting glasses needed. In a driverless car those big windows are ideal for projecting things onto.
Certainly interesting, especially as ODG partnered with BMW to bring AR to the Mini about a year ago [1] which appears to have not really gotten much commercial success. I could go into why but that's a broader conversation about AR interfaces today.
We also demoed a "see in your driveway" low poly Tesla long ago [2] which is better used for consumers to evaluate before purchase.
Musk's businesses may be spread too thin. The Solar City merger was iffy; that's a bet Solar City can beat China on solar panel price.
Tesla needs to get their production line for the Model 3 going. They need to develop self-driving that works in the real world. The Gigafactory battery factory, which right now just has the battery assembly line moved from Fremont, needs to be finished.
Space-X needs to get their launch rate up so they can stop losing business to other launch providers. They need to finish the man-rated Falcon 9 and Dragon capsule.
They also need to get the Falcon Heavy, which has slipped from 2014 to 2017, to first launch. There's also the Texas launch facility, which has slipped.
Musk has a lot of expansion underway, and most of it is behind schedule. Not a good time to go off in search of the next cool thing.
You forgot to mention the 4400 satellite internet constellation he is planning.
Now i do get the sense that his businesses are spread a little too thin, but most of his businesses are pack leaders in what they are doing. He definitely sets overly optimistic goals but that happens when you are developing new technologies that no-one has done before.
He's definitely pushing the technology on a lot of fronts. It would help if at least one of the businesses was a cash cow. But none of them are, yet. Space-X ought to be - they're the lowest cost launch provider. But they're having trouble building and launching enough Falcon 9 rockets to bring in the money.
As someone who doesn't own a car anymore I was fairly shocked recently at some of the stuff that you find in current BMWs. They do have a HUD-like navigation that's a windshield layover for example. I know they do quite a bit of HCI research. I'll be curious to see how many Tesla employees will pop up at CHI next time around. I think there's some interesting research to be done automotive AR...especially if there's an automated driving component.
Everyone here is focused on what this means for Tesla, but this may be intelligence on what is going on with the Microsoft HoloLens program if senior staff is beginning to jump ship.
What's going to happen with HoloLens? The team doesn't appear to be very large, and it's been stuck in research-land for a while now.
>this may be intelligence on what is going on with the Microsoft HoloLens program if senior staff is beginning to jump ship.
Kim seems by all evidence to be a good designer, but I'm not sure I'd call him "senior staff". He's been at Microsoft for 3ish years, which is also how long he's been not-a-student. "Senior staff" implies an importance that a "senior" title does not. I am a senior engineer at Microsoft and I assure you if I left it would mean nothing about the future of the company or the product I work on.
Kim also doesn't seem to have been working on HoloLens very long.
I'd actually see it as validation of the technology they are working on. Hurts to lose people but Microsoft can refill research positions fairly well (anecdotally).
Furthermore, several researchers from MSR working on HoloLens have left Microsoft and founded a startup around HoloLens this year: http://perceptiveio.com/team/
I'm not sure whether this is a sign of Microsoft abandoning HoloLens, or whether they are building an ecosystem of developers and integrators around HoloLens.
It could just mean Tesla is good at making irresistable offers? MS may be sufficiently confident in the current team to not worry about making counter-offers to keep hold of some members.
With full self driving, telepresence-from-the-back-seat will be an interesting application. Especially with SpaceX low latency high speed internet, whose phased array antennae should work well on the roof of a car.
I think they are very good at taking things others more or less meddled in with little efford/interest in improvements and perfecting it (by applying a more software oriented approach). It's a doubly nice strategy as there's some talent basically waiting to be hired away to work on stuff they have championed (often falling on deaf ears). AR is a bit different since that's interesting to work on and other companies are more invested.
E-vehicles were a thing but mostly a bothersome side project for most car makers. Space was cost+ and noone really cared to upset that cozy environment etc.
What's challenging about it from a design perspective? You just need a transparent display, which is a hard engineering problem (though being worked on as far as I know).
That's reassuring. One of the things about the model 3 that bothered me was that there was only one display between the driver and passenger. It seemed very weird to have to look way off to the side to read the instruments. If they can add a heads-up display, that's a great improvement. If they can make the focal distance appear to be past the windshield, that's even better. (I've heard that it takes our eyes longer to refocus from far to near and back than it takes to physically move our eyes down and up when we glance at the dashboard.)
I expect another eventual step in the evolution of the automobile to come with the introduction of head-mounted displays so that the driver can "see through" the car without obstruction. Windows could be made smaller or go away entirely to make the car stronger and more survivable in an impact, and perhaps cheaper as well.