RFID And Laser Data Fusion For Dynamic Objects Identification And Localization

In the field of object identification and localization,Radio frequency identification(RFID)has a unique tag ID,which can quickly identify the object by RF signal.It can not only save a lot of computing resources,but also solve the problems of object occlusion and environmental factors.However,due to its own hardware defects,it cannot achieve accurate positioning.Laser sensors are widely used in the field of object localization due to its high precision,wide range,and fast transmission speed.However,the laser sensor can only get sparse environmental information,and it is difficult to distinguish similar objects.Therefore,there is a certain singularity in the object identification.In practical use,it is necessary to extract effective features from the sparse information for identification.The algorithm is very complex.Therefore,this thesis integrates RFID and laser sensor information,and studies the dynamic object positioning problem.To address the above issues,this thesis fuses RFID and laser sensor data.The main work content is as follows:(1)A radial velocity similarity matching method based on sliding time window is designed to fuse the information of two sensors.The specific method is to use RFID and a two-dimensional laser rangefinder to continuously collect the tag ID and phase information returned by the RFID tag carried by the target,as well as the distance and angle of each frame of the laser point to the environment.On this basis,use the DBSCAN algorithm to cluster the discrete laser points,use the center position of the cluster as the coordinates of each object in the environment,and estimate the mobile tag based on the RFID phase based on the RFID phase information and laser ranging information.The poor radial velocity and the radial velocity of all objects in the environment based on laser clustering.Take a fixed-size time series as the "sliding window" during the entire experimental period and compare the radial velocity in each window with the similarity matching algorithm.(2)For the case when there are multiple obstacles and dynamic objects in the environment,the center position of the clusters cannot be accurately used as the estimated position of the objects because the objects may be frequently obscured,this thesis combines Pearson correlation coefficients with particle filtering and estimates the motion trajectory of each laser cluster,and then obtains the radial velocities of the laser clusters.This thesis has successively conducted experimental validation for the proposed methods.The experimental results illustrate that the proposed radial velocity similarity matching method can effectively fuse the data from RFID and laser sensors and realize the identification and localization of multiple dynamic objects.At the same time,the proposed method of combining Pearson correlation coefficient and particle filter can effectively constrain the particles,make the particles closer to the true state of the object,realize the trajectory estimation of the moving object in a complex environment.