Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.



Smooth GPS

Increases accuracy and improves GPS design
  [vote for,

It seems to me that adding 3-D accelerometers to GPS devices would be easy and useful.

Accelerometers are devices that measure acceleration. These days there are very small ones that are used in airbag components and handheld 3-d mice (I'll look for a link). Knowing initial position and velocity, you can use the data from a set of accelerometers to continuously determine your location and velocity. This can be used to provide location information that is independent of your GPS data.

Here's how it would work. Say you get a GPS data point each second. This data point lets you know your position at that instant and your velocity since the last second. While waiting for the next data point, your location and velocity is derived from the accelerometers.

The result is smooth continuous knowledge of your position and velocity, rather than the current jerky once-per-second system. The accelerometer becomes your data source, and the GPS just a method of calibrating that data and correcting for drift.

Benefits of this system:
1. User interface. Assuming you include processor power that can smoothly relay this data, you have yourself a system that is much more satisfying for the user.
2. Sky independence. Assuming you don't make movements that will confuse your system (sharp acceleration like shaking or dropping the device will cause data drift - this can mitigated or removed as computing power increases), you will be able to travel a significant period of time away from a GPS source without losing your data source. You'll be able to go under a tunnel, in a basement, inside buildings, etc., which is currently not possible with GPS.

This device will be high-end and expensive at first, due to the increased computing power required, but the price should drop with time.

Worldgineer, Apr 23 2004

Off-the-shelf accelerometers http://www.vernier....p&template=acc.html
Sounds like rapid motion won't be as large of a problem as I thought: "Use it for studying the complex motion of an amusement park ride, a bungee jumper, or simply a toss in the air." [Worldgineer, Oct 05 2004, last modified Oct 17 2004]


       While not generally available to the public, a system very similar to this is used by jumbo jets for navigation.   

       For years before the advent of GPS, airliners used inertial navigation units (INU) which had to be carefully calibrated before each flight. They were extremely sensitive but slight miscalibrations would add up to unacceptable deviations at end-course, and so they could not be used for navigation down to minimum altitudes in inclement weather.   

       With the introduction of GPS, the INU could be calibrated in-flight. While still not accurate enough to land the plane, it was now good enough to get the plane close enough to intercept the instrument landing signal (ILS glideslope) and switch autopilot control over to the autoland system.   

       Now, with the recent availability of more accurate GPS signals (the removal of the intentional error and scrambled correction factors) as well as additional differential GPS stations at many airports, the GPS system is becoming accurate enough to allow an aircraft to autoland purely on GPS guidance.   

       Many new small aircraft are coming equipped with full "glass cockpit" displays, and many of these include a GPS navigation system. I'm sure some of them also include accelerometers.   

       I like your idea, but I'm afraid I have to cry baked.
Freefall, Apr 23 2004

       Alas, baked...as part of a patent of mine :-)   

       Anyhow, who says GPS is once per second? This is only a limitation on some retail units. I was running my RTK GPS at 0.1s intervals and also used on- demand location retrieval. Pro systems use carrier phase differential which constantly analyses the harmonics of the carriers across the L1 and L2 frequencies. Retail are often single frequency timecode jobbies - "Rough hairy beasts, with eight channels!"
timbeau, Apr 23 2004

       [ff] Don't be afraid to call baked, as I'd be surprised if it wasn't. I imagine they use this exact technology in military applications in case of GPS frequency jamming.   

       But my idea is more aimed at retail-level GPS devices, such as in vehicles or handheld systems. True, small plane use is approaching my target market, but I'd be a bit surprised if they do currently have accelerometers.   

       [tim] I was just guessing on the once per second based on GPS units I've seen. Even if the available frequency is an artificial restriction, losing contact with the GPS signal is a real restriction that this device will solve.
Worldgineer, Apr 23 2004


back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle