Abstract
This paper describes a set of computer programs devel oped for Monte Carlo analysis of a complete space mis sion. The results of such an analysis are used in mission planning (for example, in the determination of how various errors affect the trajectory, midcourse-correction fuel re quirements, and other parameters).
In Monte Carlo analysis, a sample mission is simulated a large number of times, with the various stochastic pa rameters assuming different random values for each run. The resulting data is then processed statistically to yield useful probabilistic statements regarding output param eters of interest.
Three programs are described. The first derives, edits, and formats the input data required for simulation of a particular mission. The second is the simulation proper, generating a set of output parameters for each run. The third program processes the data from the simulation to derive the required statistical error-analysis data. The first and third programs are relatively straightforward and are briefly described; the simulation is described in more detail.
In general, the simulation operates by ordering all the events of the mission and using a set of standard mathe matical descriptions called PROP boxes to compute the state vector at the end of each event from its state at the beginning of the event. It uses the deviations of the pa rameters from those of a reference trajectory rather than the values of the parameters themselves, thereby simplify ing the mathematical treatment and reducing the comput ing time by orders of magnitude.
Get full access to this article
View all access options for this article.