Visual Servo Application Using Model Predictive Control (MPC) Method on Pan-tilt Camera Platform
Saragih C.F.D.a, Kinasih F.M.T.R.a, MacHbub C.a, Rusmin P.H.a, Rohman A.S.a
a School of Electrical Engineering and Informatics, Institut Teknologi Bandung, Bandung, Indonesia
Abstract
© 2019 IEEE.Visual servoing is a technique which use feedback information extracted from a visual sensor to control the motion of a robot. Combined with an image processing algorithm, visual information from a camera becomes a visual sensor which provides feedback or measurement value i.e. center point coordinate value from detected face. The face tracking system with pan-tilt camera can be utilized for many applications. The 2 DOF (degree of freedom) camera platform being used consists of 2 servo motors which moves according to control signal calculated using a control method such that the system will be able to keep the tracked face at the center of the camera frame. In this research, Model Predictive Control (MPC) method is implemented to build a controller, in which each movement of pan and tilt servo camera is controlled by control signals calculated with MPC method. MPC application is chosen so the system can produce an optimal control signal while considering input signal saturation aspect. Optimal control signal provides by the MPC controller is produced through optimizing the system performance index. The result showed that MPC controller successfully implemented in pan-tilt camera platform. The MPC control signals deliver a good control pan and tilt system response with rise time < 1 second and overshoot < 15%.
"Author
DOF (degree of Freedom),Face tracking systems,Feed back information,Image processing algorithm,Object Tracking,Pan-tilt camera,Performance indices,Visual information
"Indexed
face detection,MPC,object tracking,pan-tilt camera,Visual servoing
"Funding
ACKNOWLEDGMENT This research is a part of Research, Community Service, and Innovation Program (Program Penelitian, Pengabdian kepada Masyarakat, dan Inovasi/ P3MI) funded by Institute for Research and Community Service at the Institut Teknologi Bandung (LPPM ITB).