Cerebellum-Inspired Model-Free Tracking Control and Visual Servoing of a Rigid-Flexible Hybrid Robotic Endoscope With RCM Constraints

Abstract

This article designs a model-free control scheme based on a cerebellum-inspired network model for tracking control and visual servoing of a surgical rigid-flexible hybrid robotic endoscope. The hybrid robot is composed of a rigid robot manipulator and a flexible endoscope as the end-effector. Inspired by the cerebellum associated with human motion control, we proposed a cerebellum-inspired model-free control scheme for the visual servoing and tracking control of the robotic endoscope with RCM (remote center of motion) constraints, which does not need to know the kinematic model of the robot. This scheme combines liquid state machines (LSM) and zeroing neural network (ZNN), where the LSM is able to generate control signals and the ZNN is used as the training signal for the LSM. Simulations and physical experiments demonstrate that the proposed cerebellum-inspired control scheme is effective to handle the tracking task under RCM constraints as well as the visual servoing task.