Prediction Control for Brachytherapy Robotic System

In contemporary brachytherapy procedure, needle placement at desired location is challenging due to a variety of reasons. We have designed and fabricated an image-guided robot-assisted brachytherapy system to improve the needle placement and seed delivery. In this article we have used two different predictive control strategies in order to investigate the needle insertion efficacy and system dynamics during prostate brachytherapy. First, we used neural network predictive control (NNPC) to predict an insertion force. The NNPC uses the linearized state-space model of the robotic system to predict future system performances. Second, we used feedforward model predictive control (MPC) which allows the controller to compensate the influence of a measured disturbance's impact immediately rather than waiting until the effect appears in the system. Feedback control problem for the contact force regulation is considered. The simulation results and experiments for both cases are presented and compared.