Learning Air-Data Parameters for Flush Air Data Sensing Systems

An adaptive scattered data approximation scheme was developed to calibrate the flush air data system of a surface vessel. An array of pressure sensors were mounted flush with the deckhouse periphery, and the air data parameters were extracted from the pressure measurements. Forward and inverse prediction approaches were devised with a self-adaptive function-approximation scheme that gave reliable and robust surrogates for predicting wind speed and direction. The empirical convergence rate matched with theoretical learning rates of a greedy approximation with radial basis functions. The resulting surrogates were also used to evaluate the sensitivity to each flush-mounted pressure sensor. Fault tolerance of the proposed surrogates was also studied with respect to pressure sensor failure. Fault tolerance of the proposed surrogates was also studied with respect to the complete failure of individual pressure sensors.