INFERENCE OF VEHICLE AND ATMOSPHERE PARAMETERS FROM FREE-FLIGHT MOTIONS

Inference of vehicle and atmosphere parameters from measurements of free-flight motions is accomplished by iterative calculations of parameter values which generate or best fit the observed trajectories. Model-matching techniques are used. Predictions made by an analytical differential-equation model are compared with observations. The model parameters are varied until least-squares agreement between predicted and observed data is achieved. Systematic and gradient searches can be used, but more economical processing for a large number of parameters is obtained using the accelerated random search, which (on the average) is faster than conventional procedures if the problem has more than three parameters. This search is also capable of solving multimodal problems in finding global optima. To illustrate the inference procedure, equations of motion of an axially symmetric re-entry vehicle are derived in terms of vehicle drag and normal-force ballistic coefficients. Numerical examples are presented which demonstrate inferences of these coefficients and of the density-altitude profile. Fewer than 100 iterations generally achieve ±5% accuracy in a typical two-parameter problem, indicating the practicality of on-line parameter inferences. © 1969 American Institute of Aeronautics and Astronautics, Inc., All rights reserved.