• FuglyDuck@lemmy.world
    link
    fedilink
    English
    arrow-up
    0
    ·
    5 months ago

    Motion sensors don’t provide localization. Gps on cellphones are only really accurate to a few dozen meters.

    You can couple gps and motion (and changes in gps location) to fudge it. Which is why when you diverge off the route navigation provides… it takes it a moment to figure it out. In the display, they “know” you’re on the road so it doesn’t have to be that accurate, they just guess what lane you’re in based on direction and such.

    They’re certainly not going to know what seat you’re in.

      • FuglyDuck@lemmy.world
        link
        fedilink
        English
        arrow-up
        0
        ·
        5 months ago

        And how does the phone know if you just take left turns faster? How does it know if you’re in the left or right lane? It has no way of knowing what the forces are, or if that variation is caused by something else.

        Your phone has no idea which side of the car it’s on, and insurance companies and their apps really don’t care.

      • ShepherdPie@midwest.social
        link
        fedilink
        English
        arrow-up
        0
        ·
        5 months ago

        How do centrifugal forces determine which seat you’re sitting in inside of a car? Everyone in the car is going to be experiencing the same forces.

        • Wrench@lemmy.world
          link
          fedilink
          English
          arrow-up
          0
          ·
          5 months ago

          He is correct that the forces are different. The equation for centripetal force is Fc = Mv2/R.

          Radius is the distance from the focal point, and each seat will be different distances.

          So he is technically correct that seat position could be calculated in perfect conditions with accurate measurements.

          But none of the data that reaches this service will be remotely accurate or complete enough to make that determination. It will only have one passengers phone data, and even if it collected everyones phone data, phone sensors have a margin of error well above what the difference would be. GPS data is only even marginally accurate up to something like 6ft, and really not even then. Then cars have a lot of other factors like suspension and compression in seats, etc, that would absorb enough of the forces to muddy the data even if accurate sensors were everywhere.

          Tl;dr; another cocky person that took a few physics courses but walked away with a poor understanding of real world applications talking out their ass.