Extremely Precise Vertical Motion Sensing

Hello,

I had an idea for a gyroscopic remote capable of controlling a UAV in the pitch, roll, and yaw axes along with throttle. All gyroscopic dimensions regarding the X, Y, and Z axes controlling pitch, roll, and yaw are fairly basic and nothing is to be said for that. The question I am having is of the feasibility of adjusting the throttle or position of the aircraft in the up/down direction using very precise readings from a barometer and accelerometer. I could use any arbitrary barometer (let's say with a sensitivity of 300 to 1,100hPa) to give rudimentary readings of altitude using the hypsometric formula -->

h = (((P0 / P)^1/5.257 - 1) * (T + 273.15)) / 0.0065
where h = height in meters, P = standard pressure at sea level (1013.25), P0 = pressure at current location, and T = temperature at current location in Celsius

and combine that with readings from an accelerometer. These two inputs would be fed into some kind of Kalman filter to reduce noise from the barometer and use all input data to estimate where exactly the remote is in relation to height. I am curious as to the feasibility of this project as well as how accurate of readings I could obtain using this method (preferably I would like an accuracy of no worse than ±1 cm).

Thank you!

Do a survey of the available sensors, and from the data sheets (noise and sensitivity figures), you can determine the accuracy to expect.

Barometers used as altimeters must be calibrated very frequently to take into account weather induced pressure changes.

Here is a for-instance on using a barometer to determine altitude.

One of my former customers product is an aircraft oxygen system. They changed their barometer a few years ago to one made in Korea. No part number that I remember. But the engineer said they could detect a 15 ft. change in altitude. Doubt if you can find a hobby device that can match that.

Paul