Control Concepts for Image-Guided Object Movements
The testing stand for 3D-inverted pendulum’s stabilization was constructed at the Institute of Measurement and Automatic Control. The reference point’s and top’s position must be detected for the identification of the pendulum. The position of the top – the white ball – will be measured with help of the high-speed camera. This CMOS-camera makes 148 frames per second. The camera’s calibration will be primarily done for the definition of the relationship between the 2D-image coordinates and the 3D-world coordinates. The position of the reference point or the cart will be measured with the stepper motor’s counter.
In the control of 3D inverted pendulum two problems occure, the regulation problem and the tracking problem. The aim of the regulation problem is to stabilize the pendulum and maintain the cart at the middle of the table and the other is to stabilize the pendulum while the cart is tracking a circular path. The control schemes for the problems are PID, state feedback control, model reference adaptive controler (MRAC) using full state feedback and non-linear controler. The controlling was technically realized using an xPC-Target Toolbox from MATLAB. The control algorithms are execute on the Target-PC. The image processing is done on the Host-PC and its results are send to the Target-PC through the Ethernet.
These control concepts can be used for the stabilization of the patient's table in the radiation theraphy.
Contact person: Dipl.-Ing. Alexej Janz