Research And Implementation Of Intelligent Blood Collection Robot Elbow Detection And Dual-Arm Collaborative Control System

Venous blood collection is an essential part of today’s medical diagnosis,but there has long been a shortage of highly qualified blood collection nurses,making it difficult to guarantee a successful collection rate and often a poor patient experience.In addition,errors in routine blood collection by healthcare workers can lead to accidental self-inflicted injuries,and healthcare workers are often required to take on other medical duties in addition to blood collection.The development of intelligent blood collection robots will considerably minimize the threat of infection,enhance the success rate of blood collection,and lessen the burden on hospital operations.In this paper,we focus on the problems in the "seeing" and "moving" of the intelligent blood collection robot,and the research work is as follows.First,an elbow movement detection algorithm incorporating YOLOX and an improved frame difference method is studied and implemented.Based on the frame difference method,this paper optimizes the image noise and elbow contour detection effect to produce a clear and noise-free detection image.According to the elbow blood collection scene,we introduce target detection to determine the detection range of frame difference method,and design and implement an elbow movement detection algorithm for intelligent blood collection.Then,an improved RRT path planning algorithm with adaptive multivariate Gaussian distribution and dynamic variable step size is designed and implemented.This part of the research revolves around the model of the intelligent blood collection robotic arm,and establishes a spatial position relationship model based on the Craig D-H method,and performs forward and inverse kinematic analysis on it,and further studies and designs a user coordinate system establishment method.Based on the blood collection robot arm model and its coordinate system,the intelligent blood collection robot motion planning algorithm is designed from the working domain,trajectory planning and path planning respectively.Focusing on the path planning algorithm,the improved robotic arm path planning algorithm is designed and implemented in this paper.The path planning efficiency of this algorithm is improved by nearly 8.9 times,which is a significant improvement compared with the traditional method.Finally,based on the clinical blood collection guideline specification,the intelligent blood collection robot dual-arm cooperative system is researched and implemented.And the system pilot was completed in Hunan University Hospital,the experimental results show that the robotic arm cooperative control model and algorithm designed in the thesis have good application effect.