Heading angle and offset determination using principal component analysis
F. A. C. Pinto, John F. Reid · 1998
Summary: The main goal of the vision part of a guidance system is to output the heading angle and the offset of the vehicle. In this research, this task was addressed as a pose recognition problem where each combination of heading angle and offset was considered a pose. A set of poses (images) was collected and used as a training set. The training stage of the algorithm used the principal component analysis (Hotelling Transform) to output a low-dimensional eigenspace on which each pose was represented by its projections. Given a new image, the pose (heading angle and offset) recognition was done by projecting the image onto the eigenspace and determining the closest training image projection.