Development of auto tracking and target locking on static defence based on machine vision

Auto tracking and target locking are an automated weapon system. The system works by tracking and locking target automatically against targets selected. This system can replace the human role in a defence point. The existing technologies utilize radar and opto-electrical technology. These technologies are vulnerable against jamming and have a high degree of difficulty and the cost of procurement that is expensive. This research proposes an automatic system for target tracking and locking based on visual perception using camera. The camera on the system aims to acquire the visual information as image data. With the target reference in image selected by the operator, the image data is processed further to estimate and to track the target position in image representation. The estimated target position is used to generate motion command for the gun platform so that the gun direction aims at and lock the target. Then, the operator can take a decision on whether the target should be shot or ignored. Results of the experiment conducted in this research are the system is able to track target of the human body within moving or stationary targets chosen by the operator. The difference of gun prototypes does not affect performance of the system. The shooting accuracy has been obtained from the firing test with various distance and the height of the target is approximately 100 percent.