The present invention provides improved methods and apparatus that use machine vision techniques to rapidly and automatically guide objects into desired docking configurations. The present invention is based at least in part upon using multi-depth, rotationally symmetric targets that are observed from two or more observation perspectives. By taking advantage of parallax effects associated with the multi-depth topography of the target, the apparent positions of target features in captured image information encodes position and angular alignment of the objects relative to each other in three dimensional space. The practice of the present invention provides a fast, accurate, reliable and automatic approach to achieve hard or soft docking configurations in the electron beam therapies as well as to implement real-time gating and tracking during the course of a treatment.