Research On Force Dragging And Master-slave Force Control Of Ultrasonic Probe’s Holding Manipulator

In the traditional ultrasound-guided kidney puncture operation,the ultrasound probe needs to be controlled,positioned,and detected by a dedicated person,which causes a series of problems such as fatigue of the surgical staff and a decline in the quality of the operation.In this paper,aiming at the whole procedure of renal puncture surgery under ultrasound guidance,a pre-operative human-machine collaboration mode is designed for pre-positioning of ultrasound probes.Intraoperative masterslave mode,for master-slave hand position to position,position to force mapping.Firstly,according to the structure of the manipulator,a forward and inverse kinematics model is established,and multiple solutions are selected.For the robot end pose,a forward and inverse representation method based on the RPY pose angle is established.The geometric Jacobian matrix of the joint velocity to the end velocity mapping is solved.Furthermore,the analytical Jacobian matrix of the velocity mapping from the joint velocity to the end pose representation is solved.Finally,under the matlab robot toolbox,the correctness of the kinematics is verified through spatial motion tracking,and the error characteristics of using inverse kinematics and Jacobian matrix for motion tracking are analyzed.Secondly,for the consideration of accuracy and anti-interference ability,the admittance controller based on position output is adopted to realize the humancomputer interaction positioning control.In view of the fact that the center of gravity of the tool does not coincide with the measurement center of the sensor,the gravity and the moment of gravity are compensated.Since the positioning center does not coincide with the sensor measurement center,the torque interference caused by the position positioning on the attitude setting is compensated.The sensor data is analyzed and the filter is designed.The human-computer interaction closed-loop control system is established and its stability is analyzed.Using the pole placement method based on root locus,the stable controller parameters of the system are designed.Finally,fuzzy reasoning is used to identify the human body’s positioning intention,and dynamically adjust the controller parameters to further improve the response ability and stability of the positioning.Thirdly,starting from the tracking accuracy,a master-slave mapping strategy based on inverse kinematics for position tracking is selected.Aiming at the situation of shaking hands,a closed-loop integration strategy is adopted to smooth the trajectory.Under the master-slave control,the admittance controller with a stiffness coefficient of not zero is used to realize the mapping from the master-slave position to the position when there is no contact,and the mapping from position to force when there is contact.Aiming at the situation that when the admittance controller is in contact with the human body,the change in contact stiffness leads to instability.In the Cardan root-finding formula of the one-dimensional cubic equation of the system pole,a method for numerically solving the parameters of the system’s critically stable controller is found.Finally,through real-time identification of contact stiffness,the damping parameters of the admittance controller are dynamically adjusted while ensuring system stability and dynamic response capabilities.Finally,on the robotic arm,the human-computer interaction and master-slave control algorithms designed in this paper are experimentally verified.Through the analysis of contact force and robot movement position,the effectiveness,flexibility and stability of human-computer interaction are verified;through the comparison of the master-slave hand trajectory,the effectiveness of the strategy of eliminating shaking from the hand and master-slave tracking in this paper is verified.Accuracy:By comparing the master-slave position and contact force under fixed and dynamic parameters,the stability and dynamics of the dynamic parameter adjustment method in this paper are verified in the master-slave force mapping.