Rotation Matrix Reconstruction in Quaternions with an Optimal Observer: A Navigation Algorithm Based on Nonlinear Complementary Observers
Electronic Engineering - Faculty of Engineering - Universidad Mayor de San Andres ́
La Paz, Bolivia
Abstract—This work proposes a variation of an earlier nav- igation algorithm in the literature composed by SO(3) comple- mentary observers. Such modification means the inclusion of a quaternion optimal observer to determine the rotation matrix instead using the original vectorial reconstruction approach. In particular, this proposal involves an experimental comparison between the original and modified method. The results show about 40% increase for estimation quality while 21% more complexity using this new approach. Index Terms—Inertial navigation, Kalman filters, Quaternions, INS, Navigation Systems, Optimal Observers, Nonlinear Filters
The model estimates the position and orientation of a rigid body from the signals
of accelerometers and gyroscopes.
The results compared with the original methods describe the improvements.
The estimation signals are shown below.
For more information check the paper.