Intelligent air-target tracking and aiming mechanism via visual servoing
Journal
Journal of Control Systems and Technology
Journal Volume
6
Journal Issue
4
Pages
287-294
Date Issued
1998
Author(s)
Abstract
In this paper, a night-and-day operational monocular visual tracking system (VTS) for a gun turret system is proposed. Its functions include detection and tracking an air-target in a realistic noisy environment as well as automatic aiming of the turret gun at the target. For detection purposes, a real-time (about 100 ms) integrated detector is proposed to obtain the 2-D target location. In addition, to help the gun fighter make a better firming decision, we also try to retrieve the other 3-D orientational information of the target via model based pattern recognition. Thus, the task of 5-D target trajectory detection can be achieved in our monocular VTS. However, in this paper, only the first 2-D target location tracking is implemented. Next, a two-layer structure fuzzy controller, called the recurrent fuzzy visual gain controller (RFVGC), based on the usual perspective visual model but with additional consideration of the target velocity, is proposed. Inside this controller, a mixed Kalman filter with an IMM algorithm is also developed to estimate the 2-D target position when the target is temporarily missing and its 2-D velocity; i.e., both kinds of 3-D information are projected into the image plane. This mixed Kalman filter is mainly used to tune the first-layer de-fuzzy membership function. Finally, some experimental results show that this proposed monocular VTS can achieve satisfactory performance in tracking a maneuverable air-target.
Other Subjects
Algorithms; Fuzzy control; Gain control; Kalman filtering; Maneuverability; Mathematical models; Membership functions; Servomechanisms; Tracking (position); Recurrent fuzzy visual gain controller (RFVGC); Visual tracking systems (VTS); Computer vision
Type
journal article