Spinal minimally invasive surgery has been widely used in the medical field,and its characteristics of small recovery after surgery are favored by many doctors and patients.By combining the high precision and stability of the robot,it can effectively avoid the disoperation caused by the fatigue of the doctor during the operation,effectively improve the success rate of the operation,reduce the risk of surgery and improve the postoperative recovery effect.According to the pre-operation procedure and the actual operating environment,the doctor's operation can be transformed into the motion of the end-effector of the robot,which can also effectively avoid the damage to the doctor and patient's body caused by radiation during the operation.Therefore,the minimally invasive spinal surgery robot has become a research focus in medical robot for difficult spinal minimally invasive surgery.A 6-stewart parallel robot platform was designed to build the robot body for the particularity of the work requirements of the minimally invasive spinal surgery robot in this paper.According to the implementation environment and requirements of spinal minimally invasive surgery,the size,quality,load,workspace constraints and precision requirements of the surgical robot were detailed.Considering the difficulties of translating the compiling operation into the movement of the operating mechanism and the requirements of machining and assembly accuracy,the design scheme of Hooke hinge combined with cylindrical linear motion unit is adopted.Aiming at the kinematics problem of minimally invasive spinal surgery robot in this paper,the basic theory of motion description and coordinate transformation of Stewart parallel platform are constructed.And the analytical solution of inverse kinematics is derived,the inverse kinematics solution of the robot is obtained by using the cyclic solution method.After enough theoretical derivation and example analysis,the size and workspace of the surgical robot are verified,which lays a foundation for the forward kinematics solution.In this paper,the forward kinematics of minimally invasive spinal surgery robot is analyzed.According to the precision requirement of the surgical robot,the numerical solution method of approximation method can meet the operation requirements more effectively.The method of autonomous learning based on BP neural network model is adopted.And the kinematics inverse solution is taken as the learning sample to solve the problem,the mapping relationship between the workspace of the parallel robot moving platform and the motion space of the linear driving unit is built.According to the structure and the defects of BP neural network,an improved algorithm based on genetic network and L-M error compensation is designed.By comparing the variety of the sample data and the model of network errors in the data analysis results,it ensure the effectiveness and practicability of forward motion solver,and lays a foundation for clinical trials of minimally invasive spinal surgery robot. |