3:30 PM - 4:30 PM
[G05-P-04] Multi GNSS attitude estimation of UAVs during landing
Sensor fusion methods are becoming increasingly important in all aspects of commercial aviation, especially in the challenging phases of flight like landing, where ILS, GNSS and vision based sensor sources can be fused. If we would like to enhance the safety of air transport, then it is necessary to improve the navigation system and built -in fault detection of these functions. The attitude estimation has a prominent role of this improvement. The inertial measurement units (IMU) provide data necessary to all navigation tasks, but the drift of the gyroscope and accelerometer sensors decreases the reliability of these measurements, particularly at the low-cost devices.
The main objective of the H2020 VISION project (Validation of Integrated Safety-enhanced Intelligent flight cONtrol) is to validate smarter technologies for aircraft Guidance, Navigation and Control. Within the framework of the project we would like to present an algorithm which estimates the attitude of the aerial vehicle based on single frequency, single baseline, multi GNSS code and phase measurements, to provide a redundant alternative for IMU based attitude data.
The baseline coordinates between the two GNSS antennas and the single-differenced integer ambiguities are estimated by an Extended Kalman Filter (EKF). This EKF algorithm is supplemented with a Baseline Constrained Least-Squares Estimator of the Integer Ambiguities. The heading and the elevation angles of the aircraft are estimated with this method within one degree precision, significantly enhancing the inertial based attitude determination. The method also has the advantage, that the solution is drift free unlike the attitude of the IMU.
This study discusses the computational algorithm and the validation experiment.
Based on the real flight test data, it is examined how single frequency GNSS attitude solution can improve the Guidance and Navigation during landing of the aircraft, which is one of the most hazardous phase of the flight.
The main objective of the H2020 VISION project (Validation of Integrated Safety-enhanced Intelligent flight cONtrol) is to validate smarter technologies for aircraft Guidance, Navigation and Control. Within the framework of the project we would like to present an algorithm which estimates the attitude of the aerial vehicle based on single frequency, single baseline, multi GNSS code and phase measurements, to provide a redundant alternative for IMU based attitude data.
The baseline coordinates between the two GNSS antennas and the single-differenced integer ambiguities are estimated by an Extended Kalman Filter (EKF). This EKF algorithm is supplemented with a Baseline Constrained Least-Squares Estimator of the Integer Ambiguities. The heading and the elevation angles of the aircraft are estimated with this method within one degree precision, significantly enhancing the inertial based attitude determination. The method also has the advantage, that the solution is drift free unlike the attitude of the IMU.
This study discusses the computational algorithm and the validation experiment.
Based on the real flight test data, it is examined how single frequency GNSS attitude solution can improve the Guidance and Navigation during landing of the aircraft, which is one of the most hazardous phase of the flight.