Abstract
Images are an important source of information for spacecraft navigation andfor three-dimensional reconstruction of observed space objects. Both of theseapplications take the form of a triangulation problem when the camera has aknown attitude and the measurements extracted from the image are line of sight(LOS) directions. This work provides a comprehensive review of the history andtheoretical foundations of triangulation. A variety of classical triangulationalgorithms are reviewed, including a number of suboptimal linear methods (manyLOS measurements) and the optimal method of Hartley and Sturm (only two LOSmeasurements). It is shown that the optimal many-measurement case may be solvedwithout iteration as a linear system using the new Linear Optimal SineTriangulation (LOST) method. Both LOST and the polynomial method of Hartley andSturm provide the same result in the case of only two measurements. The varioustriangulation algorithms are assessed with a few numerical examples, includingplanetary terrain relative navigation, angles-only optical navigation atUranus, 3-D reconstruction of Notre-Dame de Paris, and angles-only relativenavigation.