Development of a Precise Control Method for a Medical Robot Working with Stiff Tissues during Hip-Joint Surgery

摘要

Many medical robots are used in orthopedic surgery. Treatment via a small incision is known as minimally invasive surgery. The restricted applicability of current robots in minimally invasive surgery led us to explore the development of a robotic assisting tool. In orthopedic surgery, the operating space is full of muscle tissues, ligaments, and tendons. These stiff tissues must be scraped and retracted to create a clear operating space and must be treated precisely so as not to damage the nerves and blood vessels they contain. However, many conventional surgical robots cannot be controlled precisely enough to work with stiff tissues because they are driven by wire-transmission. In our research, the final target is development of a muscle-scraping robot for surgical applications. As part of our previous work, we designed and built a prototype robot and then evaluated it in regard to its capability to peel living tissues and its range of motion in regard to a human model. In this paper, we describe the precise control method we developed for the robot to work with stiff tissues. The extended lengths of wires or belts were estimated from their tensions, based on a spring model and the Maxwell model. The control method was integrated into the feedback controller of the prototype robot. The control gain can be set to the desired specification using the motion equation model and our control method. The control method was tested for accuracy on a phantom with properties similar to the stiff tissues of humans. The method was tested with and without feedback from the extended belt length. With feedback, the control accuracy of the bending angle was 34% higher that without feedback. Stability and high frequency response were simulated by both measured and calculated parameters. In the future, high frequency response will be improved by phase-lead compensation. Furthermore, our method can be recomposed and used to control robotic multi-joints motion in which the robot contacts soft tissues of more complicated property in human body. Finally, in future our muscle-scraping robot will be evaluated while working on stiff living tissues.