Size/Position Optimization And Dynamic Path Planning Of Multi-arm Minimally Invasive Surgical Robot

Laparoscopic surgery robot has the advantages of filtering tremor and preventing leverage effect in minimally invasive surgery.at present,it is the most successful commercial surgical robot,and it is used more and more widely,but at this stage,there are also the following problems:(1)in terms of robot size optimization,multi-arm robot has poor in vitro cooperative obstacle avoidance performance and global accessibility.And in the traditional size optimization,the attitude ball method has the problem that it has been designed and the weight can not be accurate to each attitude.(2)in the aspect of preoperative positioning,the artificial preoperative positioning based on experience is not automated,and there are some problems,such as poor coordination,poor accessibility,easy interference and time-consuming of the manipulator;(3)the current motion control methods of surgical robots based on teleoperation are lack of automatic trajectory planning and dynamic obstacle avoidance,which leads to the fatigue of doctors and the risk of collision of the manipulator.In order to improve the accessibility,obstacle avoidance,cooperation performance and automation of minimally invasive surgery,a scheme of size optimization,automatic positioning and dynamic path planning of multi-arm minimally invasive surgical robot is proposed in this paper.the main research work is as follows:1.Kinematic modeling of multi-arm surgical robot based on spinor.The kinematics model of the da Vinci-like multi-arm surgical robot is established based on the spinor method,and the experimental results show that the kinematics model can accurately realize the forward and inverse mapping between the joint angle and the terminal pose matrix of the surgical robot.The spinor-based kinematics modeling method only needs the base coordinate system and the terminal coordinate system,which solves the disadvantage of the traditional Dmurh modeling method that each joint coordinate system needs to be defined one by one,and it is more transferable when the configuration of the robot changes.2.Size optimization of multi-arm surgical robot.The size optimization scheme of multi-arm surgical robot based on multi-objective differential evolution algorithm is designed,and the calculation schemes of global reachability objective function and obstacle avoidance performance objective function in size optimization are proposed.According to the human body parameters and 8 kinds of surgical target space,the abdominal cavity model and the three-dimensional envelope model of 8 target surgical areas are established,and the calculation scheme of the attitude range and weight of the terminal instrument is put forward.it can solve the problem that the index calculation based on the traditional attitude ball is overdesigned and the weight can not be specific to the attitude.The size optimization experiment of surgical robot based on dual-objective differential evolution algorithm is carried out.Compared with the design parameters without size optimization,the obstacle avoidance performance index GOAP is increased by 21.82%,and the global reachability index GA＿IM is improved by 22.84%.3.Positioning optimization of multi-arm surgical robot.In this paper,the multi-objective differential evolution algorithm is improved,and the adaptive strategies of crossover rate and mutation rate are proposed,so that the algorithm can converge quickly while obtaining reasonable optimization results.The positioning optimization index of multi-arm surgical robot is designed,including local accessibility index,obstacle avoidance performance index and cooperation performance index,and the collision detection scheme between manipulator and human body is put forward.The positioning optimization experiment is carried out based on the three-objective differential evolution algorithm.compared with the empirical setting parameters,the local obstacle avoidance performance index GA＿P is increased by 69.61%,the local reachability GOAP＿P is doubled,and the local cooperation performance GOP＿P is improved by 67.90%.After the improvement of the algorithm,the average number of iterations required for convergence is reduced by 40%.4.Dynamic path planning and obstacle avoidance of multi-arm surgical robot.The robot motion is modeled based on the Gaussian mixture model,the globally asymptotically stable nonlinear dynamic system is established,the robot dynamic path planning based on the dynamic system is realized,and the dynamic obstacle avoidance strategy of surgical robot based on repulsive velocity is proposed.Through the comparative experiments of 9 groups of teaching trajectories and generalization trajectories,it is proved that the generalization trajectories can better learn the characteristics of the teaching trajectories.Through the simulation experiments that the generalization trajectory can still converge to the target point after the starting point change and a certain point mutation in the path,it is proved that the trained dynamic system model is robust to external disturbances.The feasibility of the proposed dynamic obstacle avoidance strategy based on repulsive speed is proved by the experiment of robot obstacle avoidance based on repulsive speed.