Custom GPS

[PeteS 0]

One of my neighbors is doing some awesome things with GPS, custom designed for his company's flight projects. He wants to know if there is enough interest in building wingsuit specific GPS/ datalogger systems. His questions are, what update rate, 1-10 hz ? Do you want accuracy from +/- 3m to 4 cm in x,y, and z axis? data logging? Video capacity? and for the best accuracy, inertial motion sensor to smooth the plot of the flight. He is going to build a prototype for me to use, and we will work on optimal mounting for the antenna. I am by no means any type of expert on this techie type of stuff and will try to get this guy on here asap.

[KrisFlyZ 0]

I think every second logging is good enough.

If the end result is that it is

a) easy to mount
b) has WAAS/ENOS capability and
c)acquires and maintains a satelite fix like the newer Garmin units with a SiRF chip

I want one.

Kris.

[yuri_base 1]

An accurate GPS combined with Attitude and Heading Reference System (AHRS) would make a great tool for wingsuit, especially for WS BASE, where accurate acceleration and velocity data allows one to "scan" the Lift/Drag ratio for various angles of attacks during the initial ~30 seconds of flight. This will allow you to find the angle of attack that gives the maximum L/D.

I researched this a bit lately and found that existing AHRS are too expensive:

Crossbow AHRS board equipped with GPS - $1500
Xsens MTi - more than $2K

It sounds like an interesting project!
For what it's worth, here is an exchange with a Xsens rep, perhaps, you will find it useful.

Quote

I am evaluating the possibility of using MTi sensor in skydiving, in particular, integrating the acceleration/orientation data to obtain velocity components Vx, Vy, Vz, and displacements X, Y, Z. What kind of accuracy in velocity and displacement can I expect over the period of integration of about 1 minute? The typical speeds in freefall are ~120mph, displacements are of order ~2 miles.


Thank you very much for your enquiry. To know the position of the MTi using the accelerometers, you will have to double-integrate this acceleration. The MTi also measures gravity, so you have to subtract gravity first. To do this, you will have to know the orientation, which is 0.5 deg accurate (this is excellent when you do orientation measurements). This means that you will have a residue from the gravity of 10cm/s2. You will double-integrate this 10 cm/s2, making the position very inaccurate after several seconds. If you would reach the ground after 180 seconds, the measured position can be back in the plain again (2 miles inaccuracy). It seems that you can better measure this with a barometer and GPS (please see attached leaflet of the MTi with GPS).

For the speeds, it will be impossible as well. It will not go quadratically, but after 180 seconds, the speed can be up 40mph off.

The MTi with GPS is a good product for this purpose however and you will measure the orientation and the velocity/position of the skydiver.


Thank you for your excellent explanation.

I already use GPS with barometric altimeter in skydiving and its accuracy is not sufficient. It only samples once per second and the speed data is very noisy.

Actually, for my purposes, the orientation and absolute position are not that important. Important are horizontal and vertical components of speed. But as you mentioned, 10cm/s2 error in acceleration gives 10m/s error in speed after 100s flight.

Do you think the accuracy in speed can be improved if we know in advance that the motion is 2-dimensional (that is, the gravity vector is always in the plane of motion)?

Also, what is the sampling frequency of GPS in the new MTi and what is its accuracy?


You are right that when you are falling down and make assumptions on the direction, you can estimate the velocity better. There is always a starting speed however which is unknown (since when you are jumping out of the plane, you are accelerating in the downwards direction as a whole, but the sensor can rotate as you yourself are tumbling).

You will also still have the noise, which is approximately 2 cm/s2. This is certainly less than the 10 cm/s2, though.

The sampling frequency of the GPS is low (approximately 4 or 5 Hz), since it is a low cost solution. The accuracy in altitude will also be estimated by the barometer, all these data go into a filter to calculate the most accurate altitude. What this will be exactly, will have to be tested.

Yuri

Android+Wear/iOS/Windows apps:
L/D Vario, Smart Altimeter, Rockdrop Pro, Wingsuit FAP
iOS only: L/D Magic
Windows only: WS Studio

[yuri_base 1]

Yo!

More input.

If horizontal & vertical accelerations Ax, Ay and velocities Vx, Vy are known for a given moment in time, one can calculate the momentary L/D ratio (which is not equal to glide ratio in non-sustained flight) using this formula:

L/D = [Ax + Vx/Vy*(g - Ay)]/[g - Ay - Vx/Vy*Ax]

(here g is the acceleration of gravity)

This is the precise formula based on Newton dynamics; it does not depend on actual aerodynamics of the flying body.

Acceleration can be measured using AHRS with accuracy 0.01g or better, making the integrated velocity reasonably accurate during the first few seconds of WS BASE flight. This will give an accurate L/D curve as a function of time. If you combine this data with the body pitch angle data from AHRS, you get the L/D curve as a function of the angle of attack (AoA = glide angle - pitch).

This is a very valuable piece of data: only folks with horizontal windtunnels have a luxury of having it! L/D(AoA) graph may show, for example, that your sustained AoA is 10 degrees and corresponding L/D = 2.4; but at 16 degrees, L/D = 2.8. You'll be then like, "hmmmm...."

Yuri

Android+Wear/iOS/Windows apps:
L/D Vario, Smart Altimeter, Rockdrop Pro, Wingsuit FAP
iOS only: L/D Magic
Windows only: WS Studio

[KrisFlyZ 0]

Quote

This is a very valuable piece of data: only folks with horizontal windtunnels have a luxury of having it! L/D(AoA) graph may show, for example, that your sustained AoA is 10 degrees and corresponding L/D = 2.4; but at 16 degrees, L/D = 2.8. You'll be then like, "hmmmm...."

Yuri

How do you know when you are actually flying the next time that you are at 16 degrees?

Kris.

[yuri_base 1]

Quote

How do you know when you are actually flying the next time that you are at 16 degrees?

That's what you want the device to tell you. Two things: L/D and AoA.

P.S. Of cource, L/D(AoA) curve is only useful when body position is frozen throughout the flight and you can reproduce it reliably on subsequent flights. Otherwise, you may achieve 16 degrees AoA by bending your knees, but the aerodynamics will change and L/D will become 2.0. If you have the same body position from flight to flight, the purest way to change AoA is by using weights to move center of mass.

Android+Wear/iOS/Windows apps:
L/D Vario, Smart Altimeter, Rockdrop Pro, Wingsuit FAP
iOS only: L/D Magic
Windows only: WS Studio

[KrisFlyZ 0]

I thought the unit recorded AoA and other such parameters...do you intend to also have some kind of beeping or HUD as well?

Kris.