Bayesian Neural Network Modeling and Hierarchical MPC for a Tendon-Driven Surgical Robot With Uncertainty Minimization

In order to guarantee precision and safety in robotic surgery, accurate models of the robot proper control strategies are needed. Bayesian Neural Networks (BNN) capable learning complex provide information about uncertainties learned system. Model Predictive Control (MPC) is a reliable strategy ensure optimality satisfaction constraints. this work we propose use BNN build highly nonlinear kinematic dynamic tendon-driven surgical robot, exploit epistemic by means Hierarchical MPC (Hi-MPC) strategy. Simulation real world experiments show that method ensuring tip positioning, while satisfying imposed bounds on kinematics dynamics robot.