[email protected] writes:

> Ok, sounds good. But this means the gyro sensor is not involved in
> apogee computation, right?

Correct. Rockets generally tilt over at apogee, but I have seen many
slide directly backwards, making any sort of angle measurement an
uncertain indication of apogee.

> Speed is probably computed by integrating acceleration, while apogee
> will be at the zero point of the first derivative of the acceleration,
> right?

Effectively yes, but each of these are done in the context of the
physical model which is the predictive part of the Kalman filter. A
Kalman filter is a way of estimating the state of a hidden Markov chain
using noisy sensors along with a model of the Markov system
itself. We're running predictive Newtonian model of the rocket assuming
a constant acceleration (which is a good estimate near apogee), and then
constantly correcting the model based on sensor input.

> I guess I should have another look at the code, but I tried before and
> did not find what I was looking for...I'll have a closer look on it
> over the holidays :-)

There's a lot of math involved in understanding why it works, but only a
very small amount of arithmetic to actually fly the rocket. Kalman
filters were developed for early space flight applications, including
Margaret Hamilton's firmware that landed spacecraft on the moon.

> One other point is that for launch sites at high latitudes GPS is very
> unreliable due to the GPS satellites only having 55° inclination. Just
> saying...

GPS satellites are not in equatorial orbits and offer similar precision
over the entire planet; we've seen poor results everywhere from
Australia to Canada, and great results across the same range of
locations.

-keith

Attachment: signature.asc
Description: PGP signature

_______________________________________________
altusmetrum mailing list
[email protected]
http://lists.gag.com/mailman/listinfo/altusmetrum

Reply via email to