Data Compression in Recursive Estimation with Applications to Navigation Systems
Anil N. Joglekar, J. David Powell · Journal of Aircraft · 1975
The goal of data compression or preprocessing is to reduce computational requirements in a Kalman filter while retaining adequate estimation accuracy. A technique which averages batches of data is considered here. Guidelines are given for averaging, such as, how often to sample the data, when to average, and how many data points to average. The averaging technique is applied to aircraft navigation using two DMEs and either air data or inertial data. The results demonstrate that significant computation time reductions are possible with some covariance degradation. VOR/DME flight data of aircraft landing approaches were used to verify the analysis and show agreement with the analytical predictions.