Comparison of Pre-Smoothing, Exponential Smoothing, and Kalman Filtering Applied to Several Methods of Target Motion Analysis
Abstract
Comparisons are made of the bias and standard deviation in the miss distance produced by random bearing errors when pre-smoothing, exponential smoothing, and Kalman filtering are used in several methods of target motion analysis. All three techniques, alone and in combination, are used to estimate the bearing and bearing rate that are needed for two new TMA methods. The effects of pre-smoothing on the CHURN TMA also are included in the comparison. Good results are obtained with the two new TMA methods when the bearing and bearing rate are obtained by exponential smoothing with no pre-smoothing, exponential smoothing with moderate pre-smoothing, maximum pre-smoothing (two groups), and Kalman filtering with no pre-smoothing. Significantly larger biases and somewhat smaller standard deviations are obtained with CHURN when optimal pre-smoothing is used.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1972
- Accession Number
- AD0890894
Entities
People
- James M. Dobbie
Organizations
- Arthur D. Little