Analysis of Six Algorithms for Bearings only Ranging in an Air-to-Air Environment
Abstract
Six algorithms for finding the range to an airborne target using noise-corrupted bearing and bearing rate measurements from an airborne sensor were developed and evaluated. The algorithms all employ extended Kalman filters, one using a spherical coordinate system representation, and the other five using Cartesian system representation. The target models include independent, Gauss- Markov acceleration (two algorithms), constant turn rate motion in a single plane (one algorithm), and basically straight and level flight (three algorithms). The algorithms were evaluated in a Monte Carlo analysis. The truth model for this analysis was based upon two trajectories, one for the target and one for the attacker, which were deterministically set before the Monte Carlo runs. The output of the simulation was composed of plots of the error process, including the mean of the errors, the mean of the errors plus and minus one standard deviation, and the positive and negative filter-generated estimate of the standard deviation of the errors. It was found that, due to an observability problem, only the algorithms which modelled target motion as straight and level flight worked satisfactorily. These algorithms worked best in scenarios in which the actual target motion was nearly directly toward or directly away from the attacking aircraft.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1982
- Accession Number
- ADA125405
Entities
People
- Donald R. Stiffler
Organizations
- Air Force Institute of Technology