The Study of Continuum Robot Design and Control for Minimally Invasive Surgery

Abstract

Conventional surgery requires a large incision to allow surgical tools access to the operation site. Recently, minimally invasive surgery (MIS) has been implemented to reduce the post-operative pain and shorten the recovery time of patients. A continuum robot is a snake-like robot that has unique advantages of flexibility and maneuverability in MIS. Given that the continuum surgery robot works in a confined and unconstructed environment, the configuration of each robot section must be regulated to minimize collision between the robot and the environment while still controlling the tip to reach the desired position. The continuum robot also needs to manipulate soft tissues even when it contacts with obstacles or the environment. To address these challenging problems, this study was performed from three perspectives.

First, a continuum robotic surgical system was developed for MIS. Cylindrical rolling disks to pass through a flexible tube were used to improve the stiffness of the continuum robot manipulator. A unique actuation mechanism was developed to deploy the tendons and decrease friction. To ensure that the continuum robot is provided enough workspace for the surgical task, an approximate boundary of the reachable workspace was calculated. The length of each continuum robot section was optimized based on the proposed approximate boundary.

Second, an interactive control approach was proposed to assist users in controlling the position of the end effector while regulating the shape of the entire robot. Two operational modes that work simultaneously were used. The first operational mode was for the position control of the robot end effector along a desired path. Here, a challenging problem was that the desired position provided by users might not be feasible to the end effector because of the workspace limitation and/or kinematic constraints of the robot. To address this problem, we proposed the construction of a reachable workspace by using the so-called approximate boundary solution. This solution ensures that all the finalized desired positions of the end effector can be reached. The second operational mode is for the shape control of the continuum robot. On the basis of the desired robot shape provided by users, a shape correspondence approach was proposed to determine the robot shape base on kinematics. The appropriate regulation of the robot shape ensures minimized collision between the robot and the environment and benefits the position control of the robot end effector when the robot avoids obstacles.

Third, the active deformation control of soft objects by a flexible continuum robot was investigated. A robust model predictive control algorithm was designed to shape the deformation of a soft object to the desired form. Models of both the continuum robot and the soft object were assumed to be unknown. A linear approximation model from the actuation space of the continuum robot to the deformation space of the soft object was established. A state observer was designed to estimate the model uncertainty. The Jacobian matrix was estimated online by a robust Geman–McClure estimator. A prediction horizon-based controller, which was exponentially weighted for model uncertainty, was proposed. Experiments demonstrated that the proposed model-less controller was effective in manipulating the soft object when the robot interacted with obstacles.

This study provides a new design for an automatic control of continuum robots with two major contributions. The first contribution is the design of an interactive control system with two operational modes to assist users in controlling both the position of the continuum robot end effector and the shape of the entire robot. The second contribution is the design of an uncalibrated visual servoing to control the deformation of a soft object when the continuum robot encounters obstacles. The findings of the study will be potentially used for MIS.

Date of Award	11 May 2018
Original language	English
Awarding Institution	City University of Hong Kong
Supervisor	Dong SUN (Supervisor)