Research On Indoor Location System Based On Multi-Sensor Fusion

With the rapid development of information technology,indoor positioning technology has been widely used in various scenarios such as logistics,commerce,and healthcare.However,due to the obstruction of buildings and the complexity and diversity of application scenarios,global satellite positioning services are difficult to continuously provide accurate location information indoors.Therefore,it is particularly necessary to study indoor positioning technologies with high accuracy,strong penetration,and good robustness.When relying solely on a single sensor for positioning,even using the Kalman filter algorithm as an error elimination method,there will still be non-line-of-sight positioning errors.Therefore,the positioning method based on multi-sensor fusion has important value in practical applications.This article uses UWB,odometer,and inertial measurement unit as sensors,and focuses on the research of mobile robot positioning in complex indoor environments.The main research content is as follows:Firstly,this study discusses the UWB positioning measurement and solution methods,and adopts bilateral two-way ranging as the UWB ranging method.In order to improve the accuracy of indoor mobile vehicle positioning,a fusion positioning method based on UWB,odometer,and inertial measurement unit was adopted.Aiming at the measurement error of UWB in non line-of-sight environments,an unscented Kalman filtering algorithm is proposed to reduce the measurement error of fusion positioning in non line-of-sight environments.In this paper,an adaptive fusion localization algorithm is proposed,which calculates the odometer range residual and UWB range difference each time to obtain the adaptive weight in the non line-of-sight state.In addition,the impact of inertial measurement units on the positioning accuracy of multisensor fusion is also analyzed.Finally,simulation results show that the proposed algorithm can effectively improve the accuracy of indoor mobile vehicle positioning.Secondly,a multi-sensor fusion indoor positioning system is designed and built,including both hardware and software components.The hardware components mainly include: an STM32F103-based microcontroller module,a DW1000 UWB communication module,an STM32 vehicle with an odometer,an HC05 Bluetooth communication module,an MPU6050 IMU module,and an LCD display.The software components include: implementing UWB base station and tag communication,measurement,and calculation in the Keil u Vision5 development environment,reading MPU6050 angles and odometer data,and fusing the adaptive unscented Kalman filter algorithm.Communication between the mobile phone and Bluetooth module,using the mobile phone to control the vehicle,and displaying the environment on the LCD screen,which can display the real-time position of the mobile vehicle according to the positioning calculation results.Finally,the indoor positioning mobile car introduced in this paper is tested,and the test results are analyzed.Testing is divided into line-of-sight test and non-line-of-sight test.Under the condition of line-of-sight distance,the maximum positioning error of the mobile car is about16.3cm,and the measurement result is stable.Under the condition of non-line-of-sight,the maximum measurement error is 19.7cm.The test results show that the indoor positioning system with multi-sensor fusion can reduce the non-line-of-sight error of UWB and meet the positioning requirements of mobile cars in complex indoor environments.