The resurrection of Laplace's method of initial orbit determination

This report deals with a number of interrelated topics. The common thread is Laplace's method of initial orbit determination based on passively acquired optical data. We discuss this method's principal competitor (that of Gauss), the difficulties of Gauss's technique, and the traditional reasons the Gaussian method is preferred to the Laplacian. We reject this hegemony for a variety of reasons and concentrate on Laplace's method in an era of a surfeit of high quality data. This leads us into a discussion of data smoothing. Once one leaves the raw observatorial data the possibility of combining observations from multiple observers comes to mind and hence the determination of parallax by trigonometrical means. All of this may be applied to two different classes of objects-astroids and artificial satellites. Our immediate interests are in fast moving asteroids (greater than 0.5/day or an abnormally fast ecliptic latitude rate) and high altitude artificial satellites (P greater than 6h). In both instances it is the high inclination and high eccentricity subset which is of special concern.