Design And Implementation Of A Suspended Rail-based Ward Inspection Robot System

Manual inspection is a common inspection method,but due to staff shortage and the high labour intensity of inspection work,the work is often not carried out properly,and omissions and mis-inspections often occur,resulting in serious consequences.For infectious disease wards,manual inspection is also a safety hazard.The use of robotic inspection can greatly reduce the workload of health care workers and improve efficiency,so this research is geared towards the needs of unmanned nursing operations.After extensive research into the current situation of inspection robots at home and abroad,the system design requirements were analysed and the mechanical body and hardware and software control system of the orbital robot were designed for the ward environment,using a quadruple S-curve acceleration and deceleration motor control strategy and an A~* algorithm-based path planning strategy for single-latitude connected area ward rooms,so that the robot has the ability to plan paths and conduct stable and fast inspections independently.AdaBoost was used to iterate through the training features and Fisherface was used to recognise the faces of patients wearing masks.The LSSVM algorithm was optimised using the PSO ion swarm algorithm for recognising patient sign monitor data.In the case of the novel coronavirus,monitoring and analysis of information on the patient’s physical signs is used as an example.A robot backend management system was developed,where information collected by the robot body is transmitted via a communication module to the backend management system for registration and analysis,guiding the robot to make a corresponding response strategy.Finally,a self-developed ward inspection robot prototype was used as the research object to build an experimental test platform to verify the basic performance of the robot.Combined with the clinical characteristics of novel coronavirus pneumonia,relevant tests were conducted in a laboratory environment to fully test and validate key functions such as automatic inspection,face recognition and sign detection.