Integer Ambiguity Estimation and Validation in Attitude Determination Environments

Integer ambiguity resolution is a critical procedure for GPSbased attitude determination. Especially in real-time applications, fast and reliable integer ambiguity estimation and validation are the key issues for robust attitude solutions. We address integer ambiguity estimation and validation in attitude determination environments assuming rigid body platforms. In those cases, we can exploit the baseline length constraint to improve the performance of integer ambiguity estimation. For integer ambiguity validation, we investigate the IA ratio test in comparison with the conventional ratio test. In contrast to the conventional ratio test, the critical value of the IA ratio test is adjusted to the fixed failure rate. In this paper we will show that using integer least-squares with baseline constraint results in a tremendous performance improvement even in single epoch, single frequency data processing environments, allowing reliable ambiguity resolution in most attitude determination environments. We also show the advantage of the IA ratio test above the conventional ratio test in terms of the possibility of the control over the failure rate.