Research On Indoor Localization Technology Of UWB And QR Code Fusion Based On Unscented Kalman Filter Algorithm

With the continuous growth of enterprise demand for intelligent manufacturing upgrading,intelligent manufacturing is attracted unprecedented development opportunities,and the navigation and positioning of Automatic Guided Vehicle(AGV)is become a hot topic of concern to relevant researchers.UWB positioning with the concealment of the signal characteristics such as good and fast transfer rate,the orientation of QR code with accurate and cheap in price,but due to the nature of each sensor is different,have different applicable scenario,pure rely on a sensor for positioning,difficult to adapt to the complex environment,to ensure the accuracy of positioning.Using sensor fusion algorithm to fuse UWB and QR code location data can improve accuracy and reliability while reducing cost.This paper studies the key technologies of UWB and QR code positioning.The main research contents are as follows.Study UWB and QR code localization method.In order to complete the positioning of the AGV,the positioning method of UWB and QR code is studied,the influencing factors of UWB positioning are analyzed and the ranging error model is established.At the same time,the positioning accuracy calculation method is studied,and the positioning accuracy calculation method is verified by testing.The method of two-dimensional code positioning is designed,and the influencing factors of two-dimensional code positioning are analyzed.After the camera is calibrated,the positioning accuracy of QR code is tested and analyzed.The test results show that the positioning characteristic area method can be used to judge the positioning status of two-dimensional code.Research UWB and QR code fusion localization algorithm.In order to improve the positioning accuracy of the UWB and QR code fusion localization algorithm.Firstly,the kinematics model of the experimental vehicle is established according to the wheel train structure,and the Unscented Kalman filter algorithm based on fuzzy reasoning system is designed according to the kinematics model of the vehicle.The output of fuzzy inference system is used to dynamically adjust the noise of untracked Kalman filter,so as to realize the joint positioning of UWB and QR code and get more accurate positioning results.Through the simulation of the sensor data,the experiment verifies that the algorithm can get the optimal positioning results under the condition of large sensor positioning errors.Based on UWB and QR code combined positioning system and real vehicle experiment.Structures,the AGV indoor positioning system,the hardware part of the selection and the design of PC interface,relevant real vehicle experiment was designed based on the positioning system,through the fixed track experiment,verify the effectiveness of the algorithm is designed in this paper can effectively improve positioning accuracy,fixed point navigation experiment was carried out at the same time,the results show that the algorithm designed in this paper can reach the fixed point accurately.