Predicting drag or drag coefficient? Would it be possible to mathematically predict what the drag on free-falling object will be in 10 seconds (for example), if you know the current altitude, velocity and pressure? I know you can calculate "future" pressure by using atmospheric scale height, and from there you can calculate density too (have to use spreadsheet for average atmospheric temperature at certain altitudes).
But in order to calculate "future" drag force, you also need future drag coefficient, and vice versa. So how would you do it? I feel like there has to be a way if you know all other parameters, but couldn't find anything so far.
 A: The problem of predicting drag coefficients of a rigid body involves building a map with the inputs being the geometry of the body, it's translational and attitudinal velocities and positions which provide the flow field including the pressure field, and, the outputs being drag coefficient scalar values.
If the phenomenon is structured, that is, identical w.r.t. these inputs and outputs, a good idea is to collect experimental data and use machine learning techniques (nonlinear least squares, Gaussian Process, single (or low) layer artificial neural networks). In the event that large amounts of data are not available or cannot be collected, one must rely on physics based models. In this case, dimensionality reduction is critical. Particularly, one may formulate the drag coefficients as dependent on the aggregate flow velocity vector and populate tables of coefficient values as a function of flow speed and angles obtained by experiment. Such a table will enable simple (such as linear) curve fitting. Further, simulating the dynamics of the rigid body coupled with the fluid dynamics is now possible and may therefore be used to predict the drag coefficients at a certain time in the future.
