• maniacal_gaff@lemmy.world
    link
    fedilink
    English
    arrow-up
    30
    ·
    8 days ago

    Even a perfect sensor will accumulate errors in the nav solution over time because there’s no such thing as a perfect gravity model. No free-running INS will ever replace GPS long term. This shit is so frustrating to see in the press.

    • Darkassassin07@lemmy.ca
      link
      fedilink
      English
      arrow-up
      27
      ·
      8 days ago

      It will definitely require corrections from GPS or other systems, but if made sufficiently accurate; it could be months or even years before the accumulated errors necessitate a correction.

      What seems more concerning to me is a system like this would require 100% up time between outside corrections.

      A gps receiver can acquire its position from a completely powered off state. Inertial guidence though, needs to be told its current position; then it can keep track of where it goes from there. If there’s any hiccup with power, you’ve completely lost your location fix and can’t reacquire it alone.

      Put the two together though, and the inertial guidence can accurately fill in the gaps between gps service while also getting regular updates/corrections when you do have that signal available.

      • maniacal_gaff@lemmy.world
        link
        fedilink
        English
        arrow-up
        7
        ·
        8 days ago

        I don’t think you’ll ever see an INS going months without needing a correction. Imperfect gravitational compensation applies directly to the specific force measurements and those errors are then accumulated twice.

    • fartsparkles@sh.itjust.works
      link
      fedilink
      English
      arrow-up
      9
      arrow-down
      1
      ·
      8 days ago

      Especially since, to calculate current location, it needs an input of initial location (i.e. it needs GPS coordinates to begin with so it can track direction and velocity relative to that initial position). You can’t replace something you depend upon.

      • Eager Eagle@lemmy.worldOP
        link
        fedilink
        English
        arrow-up
        12
        arrow-down
        1
        ·
        8 days ago

        the initial location doesn’t need to be GPS, just a known anchor location. Which is trivial to implement in the case of trains, since stations don’t move that drastically.

        • catloaf@lemm.ee
          link
          fedilink
          English
          arrow-up
          6
          arrow-down
          1
          ·
          8 days ago

          “Fixed” ground points move a surprising amount. The local ground can shift, and of course whole continents are constantly drifting.

          • Eager Eagle@lemmy.worldOP
            link
            fedilink
            English
            arrow-up
            5
            ·
            8 days ago

            surely these are things that should be considered, but they move in relation to what? And is this surprising amount of any significance for tens or hundreds of miles of rail?

            • catloaf@lemm.ee
              link
              fedilink
              English
              arrow-up
              5
              ·
              8 days ago

              In relation to all other points of interest, which are themselves all moving.

              It’s not really relevant for rail, no, but not because of inaccuracy and drift, but because the trains are on fixed paths already. Inertial navigation and dead reckoning are accurate enough to get from station to station, and each station can have local markers, even something as simple as a reflector at the end of the platform.

              But they’re not developing it just for rail. It would be incredibly valuable for submarines and mining, for example.

        • hoshikarakitaridia@lemmy.world
          link
          fedilink
          English
          arrow-up
          3
          ·
          8 days ago

          But wouldn’t you scramble the precision with that? Stations can be quite big and anchoring to the station location means you already start with an offset to your location.

          Depending on the accuracy over time, they could pinpoint a location while the user is sleeping and than use that as an anchor for the day.

          But everything about that is speculative; let’s see where this goes first.

          • Eager Eagle@lemmy.worldOP
            link
            fedilink
            English
            arrow-up
            7
            ·
            edit-2
            8 days ago

            you’re thinking anywhere on the platform, I’m suggesting a known place near a station by which the train passes and its location - at that moment - is known.

            All the system needs is a ground-truth location after a certain amount of time. GPS is just a cheap and convenient way to do it almost anywhere, but this location correction doesn’t need to be satellite-based at all.

            • hoshikarakitaridia@lemmy.world
              link
              fedilink
              English
              arrow-up
              1
              ·
              7 days ago

              Yeah maybe that could work. I definitely agree that there’s ways to get good anchor points. Maybe through cross-check with wireless networks even.

    • Justin@lemmy.jlh.name
      link
      fedilink
      English
      arrow-up
      2
      ·
      8 days ago

      Yeah, more accurate dead reckoning is always useful, but you’ll still need some sort of of ground-based or satellite based navigation system if you’re using this for any system that requires any reasonable amount of accuracy.