Session

Technical Session XI: Advanced Subsystem or Component Developments I

Abstract

This paper proposes a new approach to GPS (Global Positioning System) attitude determination for small satellite application in LEO (low Earth orbit). Prior knowledge of attitude and integer resolution is not required. The methodology of the new approach includes integer ambiguity search, initial estimation of attitude and line bias, attitude initialisation, path difference estimation and fine attitude determination. The observable is the carrier phase difference measurement between two GPS antennas. A dual short baseline (typical baseline length up to 30 cm) is assumed in this research. The key point to initialising attitude is to estimated the attitude of individual baseline vectors with respect to the reference frame. Elimination of integer ambiguity is a simple task. Two set of vectors are required to determine an initial attitude. Once attitude is initialised, an estimation algorithm based on the extended Kalman filter starts to determine the attitude. The integer ambiguities and cycle slips can be resolved properly. The filter now is converged and, fine attitude is estimated. The robustness of the filtering estimator is tested with simulated anomalous conditions.

Share

COinS
 
Aug 26th, 8:15 AM

New Approach to Achieving Stand Alone GPS Attitude Determination using Dual Short Baselines for Small-Satellite

This paper proposes a new approach to GPS (Global Positioning System) attitude determination for small satellite application in LEO (low Earth orbit). Prior knowledge of attitude and integer resolution is not required. The methodology of the new approach includes integer ambiguity search, initial estimation of attitude and line bias, attitude initialisation, path difference estimation and fine attitude determination. The observable is the carrier phase difference measurement between two GPS antennas. A dual short baseline (typical baseline length up to 30 cm) is assumed in this research. The key point to initialising attitude is to estimated the attitude of individual baseline vectors with respect to the reference frame. Elimination of integer ambiguity is a simple task. Two set of vectors are required to determine an initial attitude. Once attitude is initialised, an estimation algorithm based on the extended Kalman filter starts to determine the attitude. The integer ambiguities and cycle slips can be resolved properly. The filter now is converged and, fine attitude is estimated. The robustness of the filtering estimator is tested with simulated anomalous conditions.