An image based targeting method to guide a tentacle-like curvilinear concentric tube robot

Abstract

Concentric tube robot (CTR) comprises several precurved concentric tubes which can rotate and translate independently with respect to each other. Each pre-curved tube of a CTR can be controlled to rotate axially and translate along its axis, which enables the tip to bypass obstacles and finally reach the target position. The concentric tube robots enable surgeons to perform delicate procedures by combining the flexibility of elastic catheters with the use of operation tools, such as needles, forceps and grippers. Hence, many clinical applications of concentric tube robots are promising, such as transnasal robotic neurosurgery. Concentric tube robots are suitable for such minimally invasive surgeries because they are able to present various flexible shapes with great dexterity, which strengthens the ability of collision avoidance and enlarges the reachability of operation tools. This paper describes an image based visual servoing method to control the concentric tube robot so that it can reach a desired position. A model-less control algorithm is developed by first estimating and adjusting the Jacobian matrix based on measurement of each incremental movement traced by a single camera, and then generating the next motion using the estimated Jacobian matrix. Therefore, no kinematic model is required to control the CTR during experiments and the experimental results reveal the feasibility and accuracy of the proposed method. To the best of our knowledge, this is the first work that realizes model-less feedback control of concentric tube robots based on visual servoing.