Prediction based control for telesurgical applications

OData support
Dr. Kovács Levente
Department of Control Engineering and Information Technology

The goals of my project work were to design and simulate controllers for telesurgery systems, built up by static and dynamic models, using different control methods and to investigate the critical factors of the systems through simulations, then draw conclusions about the usability of the control method. Additional goals included, evaluation of the quality characteristics, exploration of the new opportunities and delimitations of the possibilities.

The following control methods were applied in my work: empiric PID controller designed by Kessler method, a PID controller tuning and a model predictive controller provided by Matlab MPC Toolbox, centralized MPC algorithm, state–space predictive controls (Rawlings–Muske method and LQ–based predictive control).

The primary aim was to carry out a case study through simulations, where I can test the system's response and quality characteristics (mostly overshoot and settling time) by different values of parameters of control methods. The secondary aim was to design a modular structure to the controlled telesurgery system.

By testing the system it can be concluded that predictive methods may well be used in the telesurgery systems, because they can handle the modeled telesurgery system in the communication lag time range given, where classical control methods cannot provide stable response.

In the first instance the ability of human to the delay time, and the computing resources (calculation speed and capacity) may limit the usability of model predictive controls which goes towards the currently analyzed research problem.


Please sign in to download the files of this thesis.