Most people know something about space ships nowadays, and probably think that navigation in space is quite a simple matter; at any rate, it is a subject that is glossed over very briefly in most books on the subject. In my view, space navigation is not a simple matter, and it has certainly not received the attention it deserves. Navigation on the Earth is easy, because of the one important fact that you are on the surface of the Earth. A couple of sights, measuring the angles from two stars down to the horizon, together with the azimuths of the stars and the distance from the centre of the Earth, will give an exact statement of position. But out in space there is no Earth, no horizon—in fact nothing whatever to use as a basis of measurement. Clearly then, two angles are not enough; a third one is needed, to give a sort of tripod of sights—two of the legs being anchored to two planets (or the Sun and a planet) because their positions in space at any time are known, and the distance between them can be used as a base-line. The solution of all the triangles involved is indeed a difficult problem, but there is also the impossibility of making three simultaneous observations. It might be thought that one could do as at sea and take one sight followed later by others, making allowance for the motion of the ship in the intervals. However, this involves the idea of dead reckoning, which, although a useful concept at sea, is quite impossible to apply in space, as the following example shows.
Development of a Radio-Astrometric Catalog by Means of Very Long Baseline Interferometry Observations