Assessment of the Kinematic Parameters of Distant Noncooperative Objects

Abstract

The paper presents a method and a system for assessing the kinematic parameters of distant noncooperative objects. An experimental measuring system was developed using a long distance laser rangefinder, a GPS receiver, an electronic inclinometer, and a CCD camera mounted on a motorized pan-tilt unit. During the measurement process, the system first establishes its position and orientation in a global coordinate system. Particle filtering approach based on adaptive template matching is used to track a moving object of interest in the acquired digital image sequence. The tracking and rangefinder data is employed to determine in real time the relative position of the object, thus obtaining its 3D trajectory and velocity. To enable repetitive range measurements, the tracking data is also used to actuate the pan-tilt unit directing the rangefinder towards the object. Experiments are presented which demonstrate the performance of the system for characterization of movement of vehicles and people at kilometer-range distances.