← Back to list of papers of the 2019 EuroGNC conference |
CEAS EuroGNC 2019 |
||||||
Experimental study of an attitude estimator with measurement disturbance rejection |
||||||
|
||||||
Abstract: Attitude estimation is a corner stone of the flight stability or safety for UAVs. Even if a large panorama of efficient solutions exists, it is still difficult to guarantee the accuracy of the attitude filter during common disturbances (large accelerations or local magnetic disturbances). The integrity of the covariance (accuracy estimation) is also a difficult point in both nominal and disturbed case. This paper introduces a fault tolerant architecture for attitude estimation. It is intended to handle sensor malfunctions and unexpected environmental disturbances. The estimation architecture consists of three distinctive parts: a set of sensor models to detect incoherent or corrupted sensor measurements; a sensor data health check which activates or deactivates the state correction of the attitude filter; an attitude filter including a saturated gyroscope bias model and a decoupling between the roll/pitch and yaw angles. Simulation and experimental results show that the proposed architecture handles both inertial acceleration disturbances and magnetic disturbances without the need for speed or position measurements, or drag force models. | ||||||
Keywords: Fault detection, isolation, and reconfiguration; Robust and adaptive filtering; Sensor fusion | ||||||
View PDF CEAS-GNC-2019-035 |
Gustav Öman Lundin, Philippe Mouyon, Augustin Manecy: Experimental study of an attitude estimator with measurement disturbance rejection. Proceedings of the 2019 CEAS EuroGNC conference. Milan, Italy. April 2019. CEAS-GNC-2019-035. |
BibTeX entry: @Incollection{CEAS-GNC-2019-035, authors = {Lundin, Gustav \"{O}man and Mouyon, Philippe and Manecy, Augustin}, title = {Experimental study of an attitude estimator with measurement disturbance rejection}, booktitle = {Proceedings of the 2019 {CEAS EuroGNC} conference}, address = {Milan, Italy}, month = apr, year = {2019}, note = {CEAS-GNC-2019-035} } |