Antenna boresight calibration using optimal estimation techniques

As part of the Canadian Department of National Defence Spotlight SAR program, the Defence Research Establishment Ottawa (DREO) has been developing a Synthetic Aperture Radar Motion Compensation System (SARMCS). The SARMCS processes accelerometer and gyro measurements from an inertial measurement unit (IMU) to estimate and compensate for radar antenna motion. Uncompensated azimuth misalignment of the antenna boresight with respect to the IMU is a significant contributor to motion compensation error. Currently, an awkward to perform, laser based antenna boresight calibration procedure is employed. This thesis investigates an alternative approach in which the antenna azimuth misalignment is estimated in-flight by means of a Kalman filter. The filter integrates data from an inertial navigation system (INS), a Global Positioning System receiver (GPS) and the IMU to maintain position, velocity and orientation information. During calibration, measurements of the antenna azimuth angle to a fixed radar target are provided by a Target Acquisition Unit (TAU). A measurement of antenna azimuth misalignment is obtained by comparing the TAU-determined antenna azimuth with a computed value based on the IMU position and heading and the known target location. Several of these measurements are processed by the Kalman filter to obtain an accurate estimate of the antenna azimuth misalignment. The current SARMCS Kalman filter integrates INS, GPS and IMU data for navigation and motion compensation purposes. Formulation of an antenna boresight calibration Kalman filter is discussed with the objective of augmenting the SARMCS filter for antenna calibration. Much of the investigation involves simulations to determine a flight trajectory and aircraft/target geometry that will allow the filter to accurately estimate antenna misalignment in a minimum amount of time. A computer simulation software package, developed to support this effort, is described and the analysis methods and results of the simulations are presented. It is shown that two horizontal manoeuvres a few minutes apart followed by a period of TAU measurements while flying directly toward the target is effective. A performance evaluation using synthesized sensor data provided by DREO suggests that calibration to an accuracy of better than 4 arc minutes rms within a 20 minute period is feasible.

Name thesis:degree_name

Master of Applied Science - MASc

Level thesis:degree_level

master's

Discipline thesis:degree_discipline

Electrical and Computer Engineering

Grantor dc:publisher

University of British Columbia

Year dc:date

1993

Author dc:creator

• Christianson, John G.

Statement dc:rights

• For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

Language dc:language

eng

Handle dc:identifier

http://hdl.handle.net/2429/2627

OAI identifier oai:identifier

oai:circle.library.ubc.ca:2429/2627