Localization And Autonomous Tracking Of Indoor Robot Based On UWB

With the introduction of wireless positioning technology into the lives of the general public,indoor positioning has gradually become a research hotspot.Due to the complexity and variability of the indoor environment,indoor positioning errors are often relatively large,especially when non-line-of-sight(Non-Line of sight,NLOS)errors exist When,the positioning error becomes more serious.The thesis takes the Ultra-Wide Band(Ultra-Wide Band,UWB)positioning technology as the starting point,combined with the C-T(Chan-Taylor,referred to as C-T)collaborative algorithm,and proposes a CTK(Chan-Taylor-Kalman,referred to as CTK)combined positioning algorithm in a line-of-sight(Line of Sight,LOS)environment.Improve the CTK combination algorithm under the environment,and propose a N-CTK(NLOS Chan-Taylor-Kalman,referred to as N-CTK)combination positioning algorithm that is resistant to NLOS errors.Finally,a three-base-one-label robot following model based on the UWB module is designed,and a mobile robot experimental platform is built.The main work is as follows.(1)The paper first simulates the positioning performance of the three commonly used Fang algorithm,Chan algorithm,and Taylor algorithm based on UWB positioning.In order to solve the problem of poor positioning accuracy of a single algorithm,the Chan algorithm and Taylor algorithm with higher positioning accuracy among the three algorithms are used.Based on the algorithm,a C-T collaborative positioning algorithm is designed to realize the static and accurate positioning of the robot.The simulation results show that the probability of the root mean square error of 2cm under static positioning is 95%.(2)The paper addresses the problems of large random errors and poor stability of mobile robot positioning in indoor LOS environments,and proposes a CTK combination algorithm.Based on the CT collaborative algorithm,this paper introduces weight coefficients and incorporates an improved Kalman filter algorithm for outlier processing,And carried out dynamic and static positioning experiments.The test results show that the root mean square error of the CTK combination algorithm under dynamic positioning is concentrated at 9cm,and the average root mean square error of the CKT combination algorithm under static environment is 7.54cm,and the positioning accuracy is significantly improved.(3)This thesis has the problem of large NLOS errors in the positioning of mobile robots in indoor NLOS environments.This paper proposes an N-CTK combined algorithm against NLOS errors.Firstly,the NLOS error model is established,and the NLOS error conversion factor is introduced,and then the extended Kalman filter is merged to perform NLOS error processing,reconstruct the LOS data,and finally combine the CTK combination algorithm to obtain the final estimated value of the positioning target.The dynamic positioning experiment results show that the cumulative distribution function of the CTK combination algorithm is 0.51 when the positioning error is 20cm in the NLOS environment,and the cumulative distribution function of the N-CTK combination algorithm is 0.92,which has high positioning accuracy.(4)At the end of the thesis,an experimental test was carried out on the robot platform.Based on the UWB module,this paper designs a robot follower model with three base stations and one label,builds a mobile robot experimental platform,realizes the combined positioning algorithm of CTK and N-CTK,and conducts autonomous follow-up tests on the mobile robot.The results show that the positioning accuracy is high.The following error is less than 10cm.