Research On Indoor Positioning System Of Nursing Robot Based On UWB

Since the 21 st century,the problem of aging society has become more and more serious.The lives of many elderly people have not been guaranteed,which has continuously stimulated the development of the nursing industry.With the continuous progress of science and technology,traditional nursing methods have been continuously replaced by robots.Researchers have invented many nursing robots.However,since the nursing robot works indoors most of the time,the indoor environment is more complicated than the outdoor environment,and the positioning environment is extremely harsh,so it needs to have sufficient positioning accuracy.In order to solve this problem,this paper proposes and implements a UWB-based nursing robot positioning system by analyzing and comparing several common positioning technologies and UWB technologies.The simulation and experimental analysis of the system verify the UWB-based nursing robot.The positioning system has significantly improved positioning accuracy.The UWB-based nursing robot positioning system uses three base stations and positioning tags on the nursing robot to locate the nursing robot.The positioning scheme uses a positioning model based on the time difference of signal arrival(TDOA),and does not require clock synchronization between the tag and the base station.It only needs the clock synchronization of the three base stations to obtain the position information of the tag on the care robot relative to each base station.The collected position information is transmitted to the upper computer system through the indoor local area network.Partial least squares linear regression is performed in the upper computer system,and the data is further optimized by genetic algorithm and unscented Kalman filtering to obtain accurate position information and displayed in the nursing.Robot positioning interface,while the host computer also has user management and data storage functions.In this paper,several common de-labeling algorithms and the PLS-GA-UKF algorithm proposed in this paper are simulated and analyzed.The simulation analysis shows that the PLS-GA-UKF algorithm improves the positioning accuracy of the nursing robot compared with other de-labeling algorithms.Finally,the base part of the nursing robot is designed,and the upper computer system is built.The system test is performed in the actual environment.The accuracy of the test results is lower than the simulation accuracy,but the decline is within the allowable error.The feasibility and effectiveness of the nursing robot positioning system based on UWB proposed in this paper,and provides a new positioning idea for the development of nursing robots.