Uav coning

https://groups.google.com/forum/?fromgroups=#!topic/uavdevboard-dev/dgVENaLiexg

eam, (and especially Mark Whitehorn and Paul Bizard)

Part of the reason that I have been quiet lately is that I have been reading the attached papers, and discussing them with Doug Weibel. Doug sent them to me. (Thank you, Doug.)

They are very interesting reading on the subject of "strapdown". Basically, they present the foundations of using rate gyros and accelerometers to implement an IMU. They go into a lot more depth than the Mahoney papers.

In particular, one of them covers the topic of "coning", which is a rotation motion in which the aircraft is spinning around an axis that itself is moving. The point is, a coning motion introduces second order errors. The present version of MatrixPilot has a correction for third order errors caused by what I call "spinning" (rotation about a fixed axis), but it ignores the second order errors caused by coning. So, there is an opportunity to further improve the accuracy of MatrixPilot under extreme flight conditions.

In most flights, coning does not arise. But there are some conditions where it does. (Pay attention Mark Whitehorn and Ric Kuebler.) For example:

1. During the flight of a quad in which it is caused to yaw rapidly while making a turn. (This shows up in some of Mark's videos) 2. During the flight of a fixed-wing in which the plane is rolling rapidly while making a turn. (This shows up in Ric's videos) 3. Some of the tests I conducted with a record player, in which I changed the orientation of the spin axis while the turntable was spinning at 78 revolutions per second.

Best regards, Bill Premerlani