Research On The Performance Optimization And Simulation Platform Establishment For The Multi-arm Surgical Robot

The multi-arm surgical robot is well-performed during the surgery. The structure of the surgical robot is usually complex. Preoperative planning is vital for the safety of the surgery and the performance of the robot. Therefore, one key aspect of the Minimally Invasive Surgery(MIS) is preoperative planning for the multi-arm surgical robot. Most recent research on preoperative planning relies on surgeons’ experiences. Such methods require trained surgeons and the results vary according to some subjective factors, such as surgeons’ preference and clinical studies, which bring risks to the surgery. Therefore, a preoperative planning method for the multi-arm surgical robot based on the performance optimization is proposed. A simulation platform is established. Experiments of preoperative planning are carried out with the simulation platform and the mixed reality technique.The parameters needed to be optimized for preoperative planning are chosen through the structure analyze of the multi-arm surgical robot. In order to present the internal difference of the surgical workspace, the surgical workspace is divided and the subspaces are assigned with different weights according to the distribution of the hotspots. Three performance metrics for the multi-arm surgical robot are introduced. Based on the performance metrics, three objective functions are proposed. Global Isotropy Index(GII) is designed to present the dexterity of a single robotic arm. Cooperation Capability Index(CCI) is proposed to present the multi-arm cooperation capability during the surgery. Maximum Distance Index(MDI) is presented for the external-cavity robotic arm collision avoidance.The problem of preoperative planning in this paper can be formalised in the form of a constrained optimisation problem. An algorithm which combines Particle Swarm Optimization(PSO) and Gaussian Process(GP) is proposed. PSO is used to carry out each iteration of optimization. At each iteration of PSO, GP is used to predict the globally optimum. The optimal state for the multi-arm surgical is obtained after iterations of optimization.Kinematics analysis of the instrument arm and the endoscope arm is carried out.The position and the orientation of the grasping instrument are given by the master haptic device. The kinematics model of the robotic arm and the instrument is established.Then the simulation platform for the multi-arm surgical robot is built and function analysis of the simulation platform is carried out. Mixed reality technique is further used tomix the surgical robot in the real-world environment and the patient in the virtual-world environment.Experiments are carried out to verify our proposed method. Experiments on the established simulation platform are carried out to validate the kinematics of the virtual surgical robot. Experiments based on the established simulation platform are proposed to validate the proposed method of preoperative planning for the multi-arm surgical robot based on PSO-GP. Experiments based on the surgical robot in the real-world environment are carried out to present the optimization results of preoperative planning and validate the optimization results based on the multi-arm cooperation performance. Experiments based on mixed reality are carried out. The real robot is used to manipulate and insert the virtual instruments into the virtual human model to present the optimized robot positioning and port placements.