A Simulation Study of Adaptive Force Controller for Medical Robotic Liver Ultrasound Guidance

Compensation for the respiratory motion is a major challenge in the control design of medical robot ultrasound for accurately scanning the liver. Therefore, this paper presents an adaptive fuzzy proportional–integral–derivative (PID) force control of robot-assisted ultrasound to improve the guidance performance for anatomical or pathological structures of the liver under free breathing. A six degree-of-freedom robotic arm equipped with a 2-D ultrasound probe and a force sensor is simulated to perform the guidance procedure of the liver. The respiratory motion is also modeled and synchronized with retrospective liver ultrasound images. Without a priori knowledge of the nonlinear dynamic characteristics or mathematical modeling of the robot, the developed force controller exploits the advantages of fuzzy logic to directly auto-tune the PID controller gains for manipulating the robotic ultrasound probe on the patient’s abdomen at desired force levels in real time. A simulation framework has been developed to test the robotic force controller using four ultrasound image sequences of the liver. Compared to conventional PID controller with fixed gains, the adaptive fuzzy PID controller showed significantly better performance by resulting minimum tracked force errors of approximately 0.3 and 0.01 N with and without force sensor noise, respectively, for all tested image datasets. This simulation study presents potentially a good solution to accomplish the tracking task of desired probe forces during the robotic liver ultrasound guidance based on the developed fuzzy PID controller.