The Application on Tracking Moving Face by Combining Efficient Second-Order Minimization and PD Neural Controller

This paper presents the visual feedback control for tracking the face of the jogger on the treadmill by combining the existing techniques, efficient second-order minimization algorithm (ESM) with the intelligent control PD neural network. This study aims to develop an efficient algorithm for controlling the camera installed at the end of a 3 DOF robot arm to fix its orientation to a 2D moving face. Firstly, the ESM algorithm is derived for feedback on the orientation of the 2D human face. Then, the intelligent control PD neural network algorithm is applied for the tracking system to synchronize the motion of the camera with the movement of the face. To improve the adaptation of the system and assure the real time implementation of the application, the fuzzy switching controller and Kalman filter are also designed to support the output gain of the PD controller and predict the orientation of the face. The visual feedback control scheme was tested and proven effectively in experiments involving the tracking system, the industrial robot, and the camera attached to the end-effecter.