Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×
Transportation

Self-Driving Cars Aren't Going To Be So Great Until We Make Our Maps Better (theverge.com) 146

Uber is debuting its self-driving cars in Pittsburgh this month, a move that has many taxi drivers upset. The Verge's Nilay Patel argues that this move should change the way we think about maps and addresses. He adds that Uber is currently unable to pinpoint his home, and often ends up at the door of a "widely different address." Citing the CEO of a "large ridesharing company", Patel writes that this issue is known as the "egress problem" -- "the way we locate buildings on a map doesn't really describe how people move in and out of those buildings." Though there are workarounds and inventive ways to pinpoint your exact address, Patel argues that when we grow reliant on self-driving cars, things will get far more complicated and futile if we don't make our maps and navigation services better. He writes: Driverless cars are one of the ultimate signifiers of the future -- the real Jetsons stuff. And we're so close to making them happen: tons of cars have sophisticated adaptive cruise control that can basically keep you going on the highway, prototypes of true self-driving cars from Google and others are already on the road, and the momentum is only increasing. But maybe we shouldn't hand control of how we get somewhere to the machines until we're entirely sure the robots know where we're going.
This discussion has been archived. No new comments can be posted.

Self-Driving Cars Aren't Going To Be So Great Until We Make Our Maps Better

Comments Filter:
  • by Anonymous Coward

    This should be something that is only an issue once, or if it is an issue simply have you go to the pick up spot and press a button so they can find it

    • we have a food delivery service in my city that when you go to order shows your location on the map, if its wrong you correct it. done and done

      • by Mr D from 63 ( 3395377 ) on Wednesday August 24, 2016 @05:43PM (#52765267)
        Its not that hard to give out your coordinates if needed. That's a small issue for self driving cars, they have much harder challenges.
        • Its not that hard to give out your coordinates if needed.

          NAD27, NAD83, WGS-84, NAVD88, UTM, or state plane? Or any of a thousand other datums in use all over the world?

          That's a small issue for self driving cars, they have much harder challenges.

          It is a small part of the very large problem of knowing where it needs to be and where it cannot go.

          • by Octorian ( 14086 )

            Don't forget that most software engineers seem to assume that coordinates only exist in the most naive latitude/longitude implementation, while remaining completely oblivious to everything you just mentioned above.

            • by sl149q ( 1537343 )

              Please forward some suggestions to Google Maps engineers and Google Auto guys. I'm sure that they have not considered this.

              • Ahh, this same stupid, arrogant, insulting response when problems are mentioned.

                So, pray tell, what exactly is the solution to the problem of someone giving a Google AV a coordinate in NAD27 when Google expects WGS-84? How does Google differentiate? Do these over-smart all-knowing Google engineers think they'll teach everyone about datums and to always always use WGS-84?

                • by rhyous ( 1727822 )

                  Ahh, this same stupid arrogant, insulting response when problems are blown out of proportion by an engineer who can't see the forest through the trees or think simple.

                  So, pray tell, what exactly is the solution to the problem of someone giving a Google AV a coordinate in NAD27 when Google expects WGS-84? How does Google differentiate? Do these over-smart all-knowing Google engineers think they'll teach everyone about datums and to always always use WGS-84?

                  No. They will probably instead teach everyone to push a button on their phone. Or teach everyone to drag an icon on an online map. The map may even be 3d someday.

                  I agree that sl149q's response could come across as sarcastic, but take the sarcasm out and he is right. Let the engineers know your use cases. Let them figure out how to deliver them

          • by PybusJ ( 30549 )

            NAD27, NAD83, WGS-84, NAVD88, UTM, or state plane? Or any of a thousand other datums in use all over the world?

            There are attempts to make easier, and less ambiguous, ways to quote locations such as:

            https://map.what3words.com/ [what3words.com]

            My browser's location thinks I'm at "simply.pitch.punchy" (entirely wrongly as it happens, but that's just the location services on my laptop). Each 3m grid square gets a unique three-word pronounceable name.

            It is quite a neat system, though it does suffer from using a proprietary algorithm and word list to create/lookup the names. It is aiming at exactly this market of being able to identify

            • I've seen this system mentioned before. It is yet another example of "we'll teach everyone in the world a new system so that we can make programming AVs easier." A side-effect is that you make use of a computer mandatory to encode and decode locations into something usable, like lat/lon.

              Your computer doesn't tell you the right answer, but this will solve the location identification issue for everyone else?

              Your "location services" almost certainly works in lat/lon from GPS (when it actually works), or some

        • Maybe the problem is that we've designed cities that have a rate of errors which is fine for humans but doesn't work very well for machines. Maybe the solution is to just fix the addresses. If the address of your door is a "wildly different address", then why isn't that just your actual address?

          • There is no problem so there is no fix. Machines don't work well in a human world with lots of variables. Heck, the fastest computer can barely beat people at chess - a game with pieces that only move here and there, not 456543233435! different ways.

            • Heck, the fastest computer can barely beat people at chess

              Not anymore, there are programs written for smart phones that compete at grandmaster level. Algorithms that can search 20,000 moves per second can win tournaments, let alone the supercomputers searching 200 million moves per second. And that was 7 years ago.

            • A computer recently beat the top champion at go, which is a great deal more computationally deep than chess; but I agree with your point. All of this is really just a simple calculation done over and over on a very grand scale. Experiencing the real world is opposite, there is an almost infinite number of rules that need to be understood and utilized in order to understand it even on a fundamental level.
              • Experiencing the real world is opposite, there is an almost infinite number of rules that need to be understood and utilized in order to understand it even on a fundamental level.

                And this most of this "understanding" is irrelevant to self-driving cars.

                Humans and animals aren't doing anything particularly special when they navigate terrain.

                Classic games like chess and go were once considered "special" tasks that computers could not perform. And that remained true only until we developed the processing power and the algorithms necessary to perform the tasks well.

                Today, a human cannot beat a standard desktop computer at chess. All of the games like Chessmaster have to deliberately redu

                • That's actually incorrect, you just don't realize the things that humans are doing because you take it for granted. The example I always use is that I as a human know to be extra careful pulling out of the driveway if there are garage doors open down the street, because I know the neighbors and their habits and if garage doors are open there are probably kids running around the neighborhood playing. An automated car would not detect things in this way, by comparison any sensor that it has is a relatively
                  • by rhyous ( 1727822 )

                    Great example of something that is a valid use case.

                    Issue:
                    Don't hit kids

                    You are assuming a car has to solve the problem the way you are solving it. You don't see the kids, but you guess they might be there by the open garage. So you drive slower. Why would a computer do that?

                    A self-driving care could be equipped with an infrared sensor and doesn't have to wonder if kids are nearby. It doesn't have to guess that their might be kids by open garages. The 360 degree sensor detects all warm life in the vicinity,

                    • Well I just searched on 'vehicles with infrared sensors' and came up with nothing.
                    • To be fair, self driving cars like the Google ones use Radar and Lidar to do navigation. They sometimes however consider a plastic bag blowing in the wind as a person, so it isn't perfect, but it has been able to stop just fine for people stepping out into traffic.

                      Check out answer 2 here:

                      http://ai.stackexchange.com/qu... [stackexchange.com]

                      It has a video of what the car sees as a combination of all the sensors. You can clearly see the people in the video. The second video is even more impressive, the car is responding to a

                    • But the bag in the wind example is exactly one of the wide breadth of reasons why automated driving is a completely different problem. In a game of chess, the system has no need to consider the environment of the room the chess board is in. When you are diving, you need to consider everything in the environment and determine whether it is important or not. Google needs to figure out how to determine that bag in the wind in the air isn't important while not driving into something else heavy that that may
            • Heck, the fastest computer can barely beat people at chess

              No. Even a relatively slow computer, such as a typical laptop, has more power than Deep Blue had in 1996, and can easily beat a grandmaster.

            • A slow off-the-shelf chess computer from the '80s can beat well over 90% of the population at chess. In the chess club when I was at school, I think that there was only one person who could beat it on its hardest difficulty setting, and he was the under-13s UK chess champion. The fact that it took Deep Blue to beat the best human player in the world is irrelevant: self-driving cars don't have to be better than the best possible human driver, they just have to be better than most human drivers to be a big

            • Heck, the fastest computer can barely beat people at chess

              (1) That was 20 years ago. Computers and algorithms have improved tremendously since then.

              (2) It wasn't playing random people; it beat the world champion.

              (3) In 2006---ten years ago---a desktop computer beat the world champion. It was a 2P Core 2 Duo system, so it would be difficult to buy something slower off the shelf today.

              a game with pieces that only move here and there, not 456543233435! different ways

              The technical term you're looking for is "degrees of freedom", and while cars have more than chess pieces, the problem is not as intractable as you seem to believe. The number is not i

          • If the address of your door is a "wildly different address", then why isn't that just your actual address?

            Wildly different than what?

            I live on a corner. My "address" is on one street, but if I walk out the side door I'm on the wrong street from what my address says. And I've seen buildings that are ells, having faces on two streets with addresses that wouldn't logically be contiguous.

            How do you fix that? Isn't this a situation where "address" is NOT the same as "location", and AV need to know "location" instead of "address"? Coordinates, right?

            How do you fix the "coordinate" problem of having ten different

            • > I've seen the result of telling someone a coordinate for something and they wind up in the wrong place.

              That is not really the most difficult problem, if Uber supplies the app to communicate to them. Because as your likely aware that issue does not apply to the GPS/phone firmware level, only at the application level. For example, The road I live on was incorrect in the county survey. I have used waze to add the road to my house, now my road is in their system. Google had a survey vehicle drive my roa

            • I live on a corner. My "address" is on one street, but if I walk out the side door I'm on the wrong street from what my address says.

              This is resolved with standards. A building can only have one address, so in the case of a building on a corner, you have to pick one. This probably varies by country or state, but I think in many places in the US, residential houses' addresses are determined by which road the driveway enters from. I lived in a house like that years ago: the front door faced street A, but t

              • This probably varies by country or state, but I think in many places in the US, residential houses' addresses are determined by which road the driveway enters from. I lived in a house like that

                I think you are creating generalities from your specific situation. There are many places where the "driveway" enters from an alleyway in the back.

                Um, the default? Almost everything is WGS-84 AFAIK.

                AFAYK. But it's not that way in real life. There is no "default". You have to know. For example, I'm looking at the USGS topo map for a nearby area and it is NAD27. The map lists an almost 100m difference between it and NAD83 for east/west measurements.

                According to the Wikipedia article for WGS-84, it is the datum used by the GPS system itself,

                That's funny, because I can get my location in any number of datums using GPS. Wikipedia isn't always right.

                • I think you are creating generalities from your specific situation.

                  No, I specifically said it varies; did you miss that? You even quoted it. I also said that addressing is controlled by local governments, so places with alleyways are obviously going to be handled differently.

                  AFAYK. But it's not that way in real life. There is no "default". You have to know.

                  No, you'd don't "have to know". Enter some lat/lon coordinates into Google Maps, and it'll show you a location on the map. It doesn't ask you for yo

            • Wildly different than what?

              I'm quoting the article. He said that the address of his building and the location where the actual door are are "wildly different" addresses. I assume he means a different number and different street.

              if I walk out the side door I'm on the wrong street from what my address says.

              That's what TFA describes. I'm suggesting that he should be able to give an address for the particular door. And yes, I'm using address and location more or less interchangeably.

              How do you fix the "coordinate" problem of having ten different coordinate systems in use just in one place?

              You use the correct one. If you take a sphere and stick a pin in it, there's only one correct way to refer to where that pin is

          • If the address of your door is a "wildly different address", then why isn't that just your actual address?

            To add to the other reply:

            I used to live in a house in a row of terraced houses. My address was a the number of my house along that street and the street name. There's only one problem: there were two ways to get to my house and neither of them was from that street. The houses were all a bit above the street, with their front gardens raised above the street and the only way to the front door of the first 9 was to go around the corner at the end of the street then walk along the footpath that ran along t

          • There are two conflicting use cases for addresses given a physically large urban location: consumer and delivery. I've found several cases where I've been dumped at an inaccessible delivery address by GPS while looking for the public entrance of something.
  • I live on a fairly busy highway and for some reason Google Maps has the street numbers swapped. I should be odd numbers on the west side and even on east, but they have it swapped. I sent them a message a couple years ago and they replied "You're right, it's incorrect. We'll send an email when it's corrected". It's never been corrected. I've sent messages since, but no reply and no action.

    • This is not the worst problem. I can cite two examples within 10 miles of my residence where electronic maps show roads that aren't really there. One is a grade level crossing of the local railroad yards. It was abandoned in the 1930's. The other is a cow path (really) that is mapped as a road running North to South between two major roads running East to West. Both of these cause problems with the Turn-by-Turn navigation program (Co-pilot) that I use. I've sent several notices to the software authors but
      • I've come across a "road" shown as a back route up to the top of a local mountain. The main road was closed one time so I tried it. It took me through many one lane roads and ended at a locked gate onto forest company land.

        At the top level, most people have good addresses. The problem is that people confuse "address" with "location". And changing how we do things to make life easier for Uber is just nonsense. Also for autonomous vehicles in general. These uber-smart cars need to understand how humans do th

        • These uber-smart cars need to understand how humans do things

          There's a problem with that. Often the way that humans do things is completely arbitrary and prone to errors. That doesn't translate well to a machine. The more logical choice is in fact to reduce errors and make the things we do less arbitrary. It will make sense for more than just the machines that we build to help us.

          Let's face it, a major reason why people want autonomous cars is because the way that humans do things doesn't always work that well. It would be kind of pointless to try to program the

          • Often the way that humans do things is completely arbitrary and prone to errors. That doesn't translate well to a machine.

            Yep. I understand that.

            The more logical choice is in fact to reduce errors and make the things we do less arbitrary.

            Nope. The logical choice is to remember that humans do things the human way and will continue to do so even after a perfect engineering-based solution is created. Building a system that depends on humans doing things the machine way is building a system designed to fail.

            Let's face it, a major reason why people want autonomous cars is because the way that humans do things doesn't always work that well.

            There are two major reasons. The biggest, as far as I can determine, is that "I hate to drive". Period. The other one is an unfounded and as-yet unsupported belief that autonomous vehicles will eliminate traffic death

            • "The other one is an unfounded and as-yet unsupported belief that autonomous vehicles will eliminate traffic deaths and accidents."

              Not eliminate. Reduce.

              It's a reasonable expectation I think. Long term, anyway. Autonomous cars probably aren't going to speed, run red lights, try to beat trains to level crossings, etc, etc, etc. Yes, there will be a large number of accidents -- some serious, some fatal -- while the cars learn to recognize open manhole covers, moose, hand lettered signs that say "BRIDGE OU

            • Building a system that depends on humans doing things the machine way is building a system designed to fail.

              I'm not suggesting that we do things "the machine way", I'm suggesting that we do things "the logical way". What does it say about us that "the human way" and "the logical way" are 2 different things? Why can't they be the same thing? That's not something worth trying to correct? In 20,000 years from now are we still going to be converting between pounds and kilograms, and miles and kilometers, when we're calculating how much thrust we need to escape gravity? Are we still going to have to give turn-by-

      • My best GPS has no clue was in rural Ohio. Headed to a semi-major town, the maps guided me to smaller and smaller roads until I ended up in a driveway to a barn of an Amish family. I guess they never called to get that changed.
      • A few months ago, on a whim, I followed my GPS's directions to the next town instead of just driving down the road I knew I should take. I wound up on a dirt road a couple of km back in the hills looking at the (closed) doors of someones' barn.

    • The difference is that it costs them nothing to leave it unfixed. If however you couldn't give Google $80 a month to come pick you up they would fix it really quick. Amazon and FedEx and others keep a database attached to addresses which allow for manual corrections and they implement them instantly because it costs them money to go to the wrong place or fail to find your address.

      Uber's mapping is horrifically aweful, but they rely on a GPS waypoint to set your pickup location so they don't care either.

  • There are two ways to solve this problem.
    A) Map every single building including ingress and egress.
    B) Rely on the user being able to pick a route from the road to the building entrance.

    B) doesn't even require the user to actually drive from the road to the entrance. They can simply pick an entrance on a map, or high resolution ariel imagery.

    Plus, entrances remembered from other users may be usable.

    • B) Rely on the user being able to pick a route from the road to the building entrance.

      This is the easiest solution for now.

      Bus and trains already have that problem. And that hasn't stopped those services from being useful to a segment of the population.

    • C) Let Patel walk half a damn block.
  • by Darinbob ( 1142669 ) on Wednesday August 24, 2016 @05:44PM (#52765277)

    There will always be something not on the map. The AI on the cars will need to be good enough to figure out what to do in many cases or else allow the passenger to manually maneuver the car. That is why many of the plans from google are about driving from a known location to another known location, as it may be decades or more before they'll be able to figure out how to get into and out of a condo garage, negotiate mountain roads, deal with temporary obstacles (the dog is in the middle of the driveway so there's no way to recalculate a route), etc. But pickup from a street corner and get dropped off at a street corner, that's much more doable.

    • by Anonymous Coward

      Exactly. Nilay Patel is letting his Google shill out to play too much here. (The Verge has an incredibly valuable quid pro quo with Google, their exclusive access to Google is second to none.) Google's cars simply aren't the only way to do self-driving cars. In fact, they're not even a GOOD way to do self-driving cars. They're just a really expensive marketing scam that keeps people believing Google is an innovative company.

      Google's self-driving cars are basically Google Maps clients with motors on them. Th

    • The AI on the cars will need to be good enough to figure out what to do in many cases or else allow the passenger to manually maneuver the car.

      Why not? I agree AI is not close enough now, but there's not reason why all your decision making processes can't be mimicked given enough technology. All that is required is more detailed data and greater processing power. Humans do not posses some magic ability that is unable to be copied.

      • "Not now" may mean hundreds of years in the future too. The current state of AI is very far off to do what an average drive has to do. It's more likely that we change roads than the AI catching up (ie, sends out electronic signals that the auto can read).

        • "Not now" may mean hundreds of years in the future too. The current state of AI is very far off to do what an average drive has to do. It's more likely that we change roads than the AI catching up (ie, sends out electronic signals that the auto can read).

          Certainly - and we can even include a mechanical interlock device so that the car can cannot physically deviate from a lane even if it wanted to. Intersections could be implemented via signalling so that cars will not t-bone each other.

          In fact, to make thing efficient, the cars can even narrow the following distance to something lower than the reaction time of humans - say... a metre or so? Actually, just do away with the following distance and have the cars attached to each other. If the linked-cars leave

          • Yes, but that's the easy stuff, being on major roads. Now move the self driving car to dirt roads, driveways, condo complexes, parking lots, grocery stores, etc.

            • A solution doesn't have to cover all use cases to become mainstream. I think robot cars are mostly hype, and 'mainstream' is decades away, not months or years. However once you reach a tipping point the change will be rapid.
              Buses and trains or broadband don't go everywhere either, but all that happened with their invention was that people moved to the places that were covered by their services. The same will happen here. Some streets/locations will be considered robot car friendly, the convenience will dra
  • Comment removed based on user account deletion
    • Oh good. I thought this was an impossible problem of accessing a bunch of existing, disparate data and entering it into a system. Nothing as simple as a team of minimum wage drones could accomplish with a massive budget.

      Instead, I find self-driving cars are limited only by natural language processing, and the ability to cross-reference natural language with what's happening in the physical world.

  • I think one of the frustrations people will have with autonomous cars is the lack of mind-reading for their specific preferences. Do you prefer to park in a particular section of the parking lot? Do you want to park in the shade today? Which entrance to the store/mall/school/etc do you want to go into today? Even with perfect maps it is not possible to fully know, or easily get at the subtle desires of the occupant.

    The autonomous future might be rather frustrating as HAL drives past the parking spot you

    • How about the joy of finding a way to tell a fully autonomous car to dart into a gap at the airport arrivals/departure scrum?

      There is no scrum, the traffic computer scheduled everybody's access already and instead there is a timer on the dash telling you how long the car will be stationary. The scrum is caused by two things, ignorance (of where everybody is and where they are going, etc) and indecision. Traffic computers will solve both of these.

      Even in the shorter term, where there won't be an airport traffic computer, and there will be a mix of human and computer drivers, there won't be much problem because generally you will b

      • Maybe in 50 years, when all manual cars are banned. Until then the meatbags will be in the mix.

  • by Okian Warrior ( 537106 ) on Wednesday August 24, 2016 @06:11PM (#52765443) Homepage Journal

    I was helping a startup in the next town over, who had a package that FedEx couldn't deliver, so I agreed to drive over to the local FedEx office to get the package.

    While I was there, the person at the counter pointed out that the delivery person couldn't find the address, and I explained how to go into the parking lot, down and around the building, to the front door of the startup.

    It was indeed a weird situation where you can't see the front door from the road, and you had to know beforehand where to go to make deliveries.

    The point here is, the FedEx person at the counter typed in my instructions in the "notes" section of their database and then assured me that further deliveries should go through OK.

    Will it *really* be that hard to do something similar with self-driving cars? By which I mean, report an error to the company along with the correct data, or manually direct the car to the correct destination and note the error, and similar work around.

    I'm not sure this is a terribly important issue. I mean, it sure *seems* like there's a simple solution and the problem will quickly be self-correcting.

  • Let's get real.

    Driverless cars make sense if you live in a neighborhood full of drunk adults, or a retirement community, or if you're severely handicapped.

    And that's it.

    Now stop playing Pokemon GO on your smartphone and running over neighborhood kids.

    • by Tablizer ( 95088 )

      I'd have to disagree.

      I'd LOVE to be trollin' around slashdot etc. instead of actually driving during my commute.

      And my lazy son won't get a driver's license, making me drive. I'm thinking of telling him to Uber, but the idea of riding with a creepy stranger kind of bothers me. I'd rather it be creepy robot.

      There are people too young, too old, or too ill to drive, and many that just don't want to. I'd say that's at least 1/4 the population (excluding younger than say 12). Big potential market.

  • The system doesn't have to be perfect, only be equal to or better than typical human taxi drivers. Human drivers make mistakes and/or have bad maps also.

  • Sure, self-driving cars will frustrate us in new ways and it's always great to address potential issues prospectively.

    However, just because you are used to all the frustrations and inconveniences of the current system and you're just thinking about the annoyances of self-driving cars for the first time doesn't mean the new annoyances are worse than the old.

    Those of us who live on obscure streets or on divided streets have had to talk taxi drivers, friends and delivery people to our houses for years. GPS h

  • Reliance on maps is never going to cut it. All maps have errors and the cost of updating to maintain all current details is high and never done. What we need are many advances in AI algorithms like the ability to ask for directions from the occupants and translate that contextually into where the vehicle needs to go, perhaps using street view information and vision algorithms.
    Thousands of little problems like this will plague autonomous vehicles and it will likely be two at least decades before they func
    • How about just reading the house number written on letter boxes? Relatively trivial compared to everything else.

      Most people assume that the machines are as completely unintelligent as the programs that they are working on. But being able to see and work in the real world is the whole point of this exercise.

      • by Imrik ( 148191 )

        Around here this would lead to a dozen different houses being considered directly in front of my building, mine would not be one of them.

  • There is already a solution to this problem. Whether it's the best solution remains to be seen, but it seems like a great step in the right direction to me.

    This was posted on slashdot many months (over a year?) ago.

    http://what3words.com/

    Every 3x3m square in the world has its own unique 3 word code. So you can give people the 'address' of your garage, or your front door, or the lampost outside the park down the road, without having to provide the exact GPS co-ordinates.

  • So that the vision systems of the cars can help them stay in their lane.

  • by Anonymous Coward

    Why not the Russian ones?

    http://www.popularmechanics.com/military/news/a16530/these-soviet-maps-were-the-top-secret-google-maps-of-their-day/

    • by k6mfw ( 1182893 )
      That is a really interesting article, though here is link to entire article at http://www.wired.com/2015/07/s... [wired.com] and I've squandered much of my work time reading this. It seems USSR allocated a lot of people with lots of time to do this detailed field work. Google can do the same as they have tens of $billions$ of extra cash.
  • by Anonymous Coward

    If a driver brings you from point A to B in exchange for cash, it is a taxi service.

  • How hard can this be... when you request a pickup, the self-driving car begins driving to you, and a pin is placed on the map where it *thinks* it should pick you up. You review the pin, and move it to whatever door / street corner you want. The self-driving car then re-routes to pick you up where you placed the pin.

    On future visits through the same provider, it will remember where you dropped the map pin and default to picking you up there. If you choose to move the pin again (say, you're coming out of
    • That was my thought as well. Hell, you could even allow the user to input some pathing for the last little bit, in the case of the map being inaccurate or to access a slightly different area than the exact address.

      One place where I lived the house was on a steep hill, with no real access between the front door and the road. Yet the street address was directly in front of the house. To get picked up at my doorstep, I'd need to set the path so the car would drive past the house, around the building ne

  • "And we're so close to making them happen". No, we're not. Stop believing the hype, we're literally still in the infancy of self driving, the problems that still need to be addressed are numerous and not easy.

    "What NASA could teach Tesla about the limits of autopilot" is a great read. http://www.pbs.org/newshour/rundown/time-seem-fly-faster-age/

  • I work in a building where the entrance faces a pedestrian walkway. It is a half-block walk to either of two streets. The problem is generic to the whole multi-block walkway.

    With Uber I can position the cursor on the map to identify the pickup location. This is not that hard, and there is no reason for the ride-sharing company to not learn from it, and sell information to or trade information with a mapping company.

  • Patel argues that when we grow reliant on self-driving cars, things will get far more complicated and futile if we don't make our maps and navigation services better.

    No, Mr Patel, a significant number of your potential market/ audience will not become reliant on your product until after map and navigation services are better.

    And incidentally, some of us are used to spending time where you don't have electrical power or any mobile phone/ data signal (Iridium excepted, all 9600 bps of it) , and the magnetic

Keep up the good work! But please don't ask me to help.

Working...