A new approach to image processing and interpretation in computer-based autonomous guidance is described. This approach makes use of a wide-angle view technique for the purpose of safety path optimization. Image interpretation is enhanced by a study on the geometrical aspects of the problem, based on experimental observations.
