Study On Force-position Signal Perception And Human-machine Cooperative Strategy For A Vascular Interventional Surgery Robot

The development of vascular interventional surgery(VIS)robot makes up for the disadvantages of traditional interventional surgery and reduces the doctor’s body burden and radiation risk.However,the existing VIS robots have problems such as incomplete acquisition of surgical information,low intelligence of VIS robot and human-machine collaboration,which have a great impact on the surgical effect of doctors using interventional surgery robots.This dissertation focuses on the above three key issues,and three aspects of research work are mainly conducted in this dissertation as follows:(1)A master-salve VIS robot is designed and built.The indexes of the robot system are comprehensively evaluated.Real-time collection of force position information and master-slave remote operation during surgery are realized.A bionics-based flexible guide wire clamping mechanism is designed by analyzing the principle of doctor’s finger kneading.It realizes the flexible and precise force clamping of the guide wire by constructing a closed loop of kneading force.Based on the principle of torque balance,a guide wire delivery force measuring mechanism is designed to enhance the weak force collection capability.A master-slave wireless remote-control system is designed and built based on wireless transmission technology.The doctor’s real sense of force is constructed in remote operation by a master-slave bidirectional force closed-loop control algorithm.With the addition of multiple sensors and visual interfaces,the doctor can perceive multiple real-time information of the remote operation robot and remotely control the robot during the operation.(2)The force and position information collected in the vascular model were trained based on softmax regression machine learning algorithm.The training results is applied to the surgical system to realize the real-time prediction of the guide wire delivery state,assist the doctor to perform surgical operations and adjust the surgical plan,and improve the safety performance of surgery using VIS robots.The experimental results show that,the guide wire delivery state identification model has high confidence and meets the requirements of vascular interventional surgery.(3)The human-machine trust evaluation model based on the guide wire delivery state identification model is established by analyzing the actual doctors’ surgical scenarios and the meaning of surgical actions.It realizes the real-time and stable evaluation of the level of human-machine operation coordination.Based on the real-time evaluation results of the human-machine trust in the interventional surgical robot system,a human-machine trust evaluation model is proposed.A human-machine collaborative surgical robot system is constructed that is based on doctor’s decision-making and assisted by robot decision-making.The experimental results show that,the human-machine collaborative model based on human-machine trust can adjust the ratio of master-slave mapping dynamically in real time.It realizes three master-slave operation modes that are adaptive to different surgical conditions,thereby improving the safety and efficiency of human-machine cooperative vascular interventional surgery,and improving the surgical effect.