Extended Kalman Filtering and Maximum-Likelihood Estimation for Point Target Localisation

摘要

Estimation of the location of point targets from radar observations will be investigated. The location can be determined from the echo's delay and the aspect angles to the target, which can be inferred by e.g. an maximum-likelihood (ML) estimator. For rapidly moving targets delay and aspect angles undergo fast variations. Therefore, short time observations are only considerable for the estimator application. Hence, the estimation effort becomes high for a long time observation of the targets. Target tracking by extended Kalman filter (EKF) will be considered in order to reduce the computational effort. A simulation based comparison of ML and EKF based target localisation accuracy will be drawn.