Abstract

In this research, a learning-based force control schema for tendon-driven steerable catheters with the application in robot-assisted tissue ablation procedures was proposed and validated. To this end, initially a displacement-based model for estimating the contact force between the catheter and tissue was developed. Afterward, a tendon-driven catheter was designed and developed. Next, a software-hardware-integrated robotic system for controlling and monitoring the pose of the catheter was designed and developed. Also, a force control schema was developed based on the developed contact force model as a priori knowledge. Furthermore, the position control of the tip of the catheter was performed using a learning-based inverse kinematic approach. By combining
the position control and the contact model, the force control schema was developed and validated. Validation studies were performed on phantom tissue as well as excised porcine tissue. Results of the validation studies showed that the proposed displacement-based model was 91.5% accurate in contact force prediction. Also, the system was capable of following a set of desired trajectories with an average root-mean-square error of less than 5%. Further validation studies revealed that the system could fairly generate desired static and dynamic force profiles on the phantom tissue. In summary, the proposed force control system did not necessitate the utilization of force sensors and could fairly contribute in automatizing the ablation task for robotic tissue ablation procedures.