Three-Axis Attitude Determination Using Global Positioning System Signal Strength Measurements

An alternative method to the standard carrier-phase algorithm for deriving three-axis attitude solutions from global positioning system (GPS) signal sources is investigated. This new method uses signal-to-noise ratio (SNR) measurements from two or more canted antennas and a knowledge of each receiving antenna's gain pattern to generate pointing vector solutions. These vector solutions are then converted to a three-axis attitude solution. The method has the advantage of requiring no complicated initializing procedures such as integer ambiguity resolution, and it can generate three-axis solutions with as few as two antennas. A solution is produced whenever a minimum of three GPS satellites are in view, regardless of the vehicle's orientation at that time or at any time previously. The performance of this SNR approach is investigated using a Kalman filter to derive solutions on a satellite whose attitude generally remains fixed in the local frame. The performance of the algorithm is evaluated while the canting angles between the antennas and the filter values are varied. In addition, self-calibrating and self-scaling options are explored to reduce the algorithm's dependence on any outside knowledge of the vehicle's attitude during an in-flight calibration process. The study also examines the performance of the solution under the expected error conditions of an inaccurate calibration and sky blockage. The results demonstrate that the estimate is relatively insensitive to these expected errors. Some general conclusions are drawn about the performance and sensitivity of the developed algorithm.