measuring vertical movement with accelerometer

In paragliding, it is important to be informed as soon as a vertical speed increases. This indicates a thermal (warm air climbing), so the location to stay to climb.

The standard way of measuring vertical speed is of course the barometer, but some devices to be more reactive, use the accelerometer.

This sounds simple if the device would be attached to a fix object. I would then just need to measure the vertical g. But on the wrist, it is necessary to develop an algorithm with the different g axis.

So, how to detect vertical movement (reference earth gravity) with a multiple axis g-meter?

Any idea/recommendation on how to do that ?

thanks all.

0 li2niu🏃🏻 over 3 years ago

IMHO，it is not accurate. Cuz g won't change a lot when the distance an object away from the ground is far less than the radius of the earth.

0 Nick over 3 years ago

I looked into this, and without the GYRO data from Garmin it basically cannot be done....

0 jeromej_ over 3 years ago in reply to Nick

Yaw looks to be missing, But the sensor provides roll and pitch? I assume these are giro in degree, no?

0 Nick over 3 years ago in reply to jeromej_

No, it provides acceleration, not roll or pitch.

There are no Gyro outputs at all

0 jeromej_ over 3 years ago in reply to Nick

Thanks Nick, interesting... I thought that the roll and pitch described in the doc were gyro values.

developer.garmin.com/.../AccelerometerData.html

0 Nick over 3 years ago in reply to jeromej_

As you can see, they are under " Accelerometer Data" and the formula used atan2(y, sqrt(x^2 + z^2))., will only ever work if you are completely stationary.

They do have a gyoro in most watches I believe, but as you can see there is no Gyrodata feed, only accelerometer.