Statistical analysis Techniques for Error Propagation in Large-Scale Missile Systems
Abstract
The statistical covariance technique is proposed as a computer analysis tool for missile systems error propagation studies. It is shown that on representative examples the statistical covariance technique performs better, based on speed and accuracy, than the traditional Monte Carlo approach of averaging a large number of random simulation runs. Although exact only for linear, time-varying systems, the covariance technique gives attractive approximate results for wide range of nonlinear missile systems. The extension to nonlinear systems utilizes incremental equations about the noise-free solution. Except in the extreme cases of very harsh nonlinearities and high input noise levels, the statistical covariance technique has been shown to yield, for the systems considered, results comparable to 1000 Monte Carlo runs.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 20, 1971
- Accession Number
- AD0739513
Entities
People
- James R. Rowland
- Willard M. Holmes