Research On Indoor Positioning Method For Mobile Robot Based On Magnetic Field

With the unceasing development of the robotics industry,the intelligent mobile robot has gradually come into the indoor environment and provided life services.The robots generally carry various sensors,and perceive the surrounding environment by analyzing and processing the sensor data that collected by the sensors.The mobile robot achieves accurate indoor positioning through various sensors is not only with great theoretical value and practical significance,but also lays the foundation for completing follow-up mission planning and improving intelligence level and reliability of robot.There are some mobile robot positioning methods such as the position estimation method,based on the distance sensor positioning method,based on computer vision positioning method,WiFi and Bluetooth beacon positioning method.But each method has its own advantages and disadvantages.Recently,the indoor positioning method mainly uses WiFi,Bluetooth,RFID as beacon to achieve location.The beacon is scanned by the airborne sensor,and then make use of triangulation to estimate the position.Although the positioning method that uses Bluetooth or RFID as beacon can obtain relatively high positioning accuracy,the method usually requires additional infrastructure and with interference from interior layout and external signal,thus limiting the large-scale use of positioning technology.As for the WiFi network,the large-scale deployment has reduced the cost of the indoor location system,but WiFi is susceptible to factors such as low reception sensitivity,multipath effects,signal variability,and device differentiation,so the location is barely satisfactory.At present,the indoor location method that use indoor geomagnetic positioning is increasingly concerned by the researchers.Because indoor geomagnetism has been distorted under the influence of reinforced concrete,and its local differences and stability provide the possibility for indoor positioning.In this paper,the mobile robot achieve positioning in the indoor environment with geomagnetic sensors to collect the indoor magnetic field.The main contents of paper as follows:(1)Do a feasibility analysis for indoor positioning with indoor geomagnetic.In this paper,the distribution characteristics of indoor magnetic field,the stability of time and the difference of different equipment are analyzed by experiment.At the same time,put forward a solution to achieve indoor positioning.(2)Research on indoor location algorithm of improved particle filter.In this paper,the basic particle filter algorithm is optimized and improved,including geomagnetic sequence fingerprint model,motion model,observation model,resampling model and location failure detection model.The observation model is used to filter the raw data,and the similarity is calculated by the appropriate fingerprint matching algorithm.Finally,the influence factors of the accuracy of the localization algorithm are analyzed by experimental analysis:the number of particles,the frequency of acquisition and the length of magnetic matching,and the validity of the algorithm is verified.(3)Indoor positioning system and simulation experiment platform.This paper designs and builds a set of mobile robot indoor positioning system,which provides robust and real-time indoor location service for mobile robots.At the same time,a simulation experiment platform is proposed to validate the validity of the localization algorithm and improve the accuracy of the localization algorithm.