Hi all,


I'm about to start a project on making my first autopilot! From what I've read so far, thermopiles are fairly accurate and not susceptible to noise such as vibration whereas the IMU is. However, from most autopilots that I've seen, neither are used together for attitude estimation (i.e. they seem to be only used separately). I thought it would be great if the sensor drift from the IMU could be fused with somewhat reliable data from the thermopiles. Am I looking at it wrongly or is there a reason why this isn't commonly implemented?





Views: 437

Reply to This

Replies to This Discussion

A good IMU is pretty reliable. The primary input is the gyro, the drift is eliminated using accelerometers and the gps or optional magnetic sensors. Smart algorithm required, but fortunately there are great papers (start with the very clear and outstanding Premerlani descriptions) and open source code as well.

A thermopile cant't work well in many situations, for example close to trees, buildings, or inside clouds. Correcting (overriding) the IMU upon the thermopile doesn't seem reasonable.

I would consider more as a back up to an IMU rather than a using it for sensor drift. The sensor drift problem has pretty much been solved with the DCM method. It could be useful to recover after an in flight reset, but I would think the system would use one or the other at any one time.
Hi Joe ,

There are two type of IMU.One is thermopiles that it can use easy.But accuracy is alittle low.IMU is the best,But it is very difficut to write own program and build CKT.We use X-sene IMU http://www.xsens.com/en/general/mti-g.But we suggest ,you should use V2 deve:.After that change one.Good luck.
Aero hawk
Aero Space University
To me IR is superior for simple wing leveling applications. It's 10 times less complex / more robust. The IMU's really shine when used for coordinated turns and closed in spaces.

Reply to Discussion


© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service