Highest density gates for target tracking

The problem of forming validation regions or gates for new sensor mea surements obtained when tracking targets in clutter is considered Target dynamics and measurement characteristics are modeled with possible non Gaussianities or nonlinearities so that some degree of approximation is usually required in the computation of the ltering densities for the target position and predictive densities for future measurements Highest den sity gates are proposed as summaries of the predictive densities These gates are constructed numerically via simulation from the ltering den sity approximation The algorithm results in gates that are exact up to numerical accuracy regardless of the approximation used for the ltering density That is given an approximation to the ltering density the gat ing procedure accounts for all further nonlinearities and non Gaussianities Numerical examples show that when the predictive density is markedly non Gaussian highest density gates o er advantages over the more com mon rectangular and ellipsoidal gates