Session

Technical Session XI: The Technology Frontier-- Advanced Technologies, Subsystems, and components for Small Satellites: Section II

Abstract

A low cost, robust Attitude Determination System with good accuracy for small satellites can be achieved by using a combination of magnetometer and Coarse Sun Sensor. The Coarse Sun Sensor consists of solar cells placed on each of the six outside surfaces of the satellite. The main measurement error of the Coarse Sun Sensor occurs due to the Earth's albedo which can cause an angular deviation of more than 20 degrees. By modeling the albedo and applying state estimation methods for attitude determination the error can be reduced drastically to less than 1 degree, depending on the conditions. In the _rst part of this paper the development of the albedo model is presented. It is used to correct the measurements of the Coarse Sun Sensor as well as to have a realistic simulation of the sensor. The albedo model of the Earth is created using existing re_ectivity data. The variation of the re_ectivity is analyzed statistically and parametric functions are derived to describe the albedo light vector received by a satellite in its orbit. In the second part an Extended Kalman Filter for attitude determination is presented which is using the two vectors provided by the magnetometer and the Coarse Sun Sensor. A model of the total light vector - the sum of sunlight and Earth albedo light - and the magnetic _eld vector taken from the IGRF model are utilized as the measurement model. The paper will present simulation results showing the accuracy of the attitude determination system for various types of orbits.

SSC03-XI-7.pdf (560 kB)
Presentation Slides

Share

COinS
 
Aug 14th, 12:00 PM

Low Cost, Good Accuracy - Attitude Determination using Magnetometer and Simple Sun Sensor

A low cost, robust Attitude Determination System with good accuracy for small satellites can be achieved by using a combination of magnetometer and Coarse Sun Sensor. The Coarse Sun Sensor consists of solar cells placed on each of the six outside surfaces of the satellite. The main measurement error of the Coarse Sun Sensor occurs due to the Earth's albedo which can cause an angular deviation of more than 20 degrees. By modeling the albedo and applying state estimation methods for attitude determination the error can be reduced drastically to less than 1 degree, depending on the conditions. In the _rst part of this paper the development of the albedo model is presented. It is used to correct the measurements of the Coarse Sun Sensor as well as to have a realistic simulation of the sensor. The albedo model of the Earth is created using existing re_ectivity data. The variation of the re_ectivity is analyzed statistically and parametric functions are derived to describe the albedo light vector received by a satellite in its orbit. In the second part an Extended Kalman Filter for attitude determination is presented which is using the two vectors provided by the magnetometer and the Coarse Sun Sensor. A model of the total light vector - the sum of sunlight and Earth albedo light - and the magnetic _eld vector taken from the IGRF model are utilized as the measurement model. The paper will present simulation results showing the accuracy of the attitude determination system for various types of orbits.