Variation of an inertial navigation system

your right AWOL.

as an update for everyone too we are also using an absolute pressure sensor + 9 DOF IMU. currently we are using mems but will be switching out to more accurate sensors shortly. we are not looking to spend millions to do this and we are not looking to spend 100 bucks to do this either. I think one mind set with arduino is to do everything on the cheap. Where I use it as a prototyping platform as a quick processor to use. In the end I rarely use it as my final choice in processors. I figured I'd post here initially because I figured a bunch of arduino guys are using cheap IMU's in there RC projects and would have been able to give me a jump start on some of my research.

stealthtransam:
being an engineer for over 10 years has taught me that anyone saying it can't be done just isn't trying.

That's great news! Let's get started on that time machine project...

1 hour 100 ft. That's 0.5m / minute. Much less than 5m. While on somebody's back? Walking thru a cave? Bouncing and bumping into things? Come on. Maybe if you had a million dollars I'd bet in your favor. I'm looking forward to your results. Please do post here when you achieve them, even if it's 2m/min error. It helps to have realistic expectations before you begin. Nobody else can do this. You will be the first in the world! I'm usually the last to be be the pessimist. Go for it!

Alright, it seems that the calculations are in hand, though I'm not sure what you mean about "8 MHZ sampling" - surely not 8 million measurements a second! Similarly "the propogation of error" - a numerical integration can have too many steps and even more degrades accuracy further. I'd also recommend second-order numerical integration rather than first, and the state variables must be held to high precision. But without details of your calculation, I'm guessing.
Just a warning about magnetometers in the field: in recent news in New Zealand a tramper was lost in fog in part because the magnetic field of some of the fancy electronics he was carrying messed with the geomagnetic field other fancy electronics was attempting to measure. And caves can be surrounded by all manner of odd rock.

Yeah I do have some concerns about erroneous readings do to finite material around me but that concern needs to wait until. Like any project this won't be an one product fits all solution. For those interested Analog Devices sells some very nice high precision MEMS Gyros as COTS parts. The specs are similar to optical gyro's but still use MEMS technology.

Press Release

Data Sheet
http://www.analog.com/static/imported-files/data_sheets/ADIS16136.pdf