Design And Implementation Of Enhanced Particle Filterlocalization Subsystem Based On Attitude And Heading Reference System

With the continuous development of the Internet and communication technology,people’s demand for shopping and travel is getting higher and higher.In unknown and large indoor space,location information is a great advantage for people.Pedestrians search for specific places,products and services every day.However,in the indoor environment,the satellite navigation signal is blocked by buildings,so it cannot provide positioning services for indoor users.Among many indoor positioning technologies,pedestrian dead reckoning(PDR)is one of the most commonly used.However,due to environmental reasons,there will be deviation in the heading of PDR,which will reduce the positioning accuracy.At the same time,the operation of the positioning algorithm is transparent to the algorithm developers in the positioning process,which makes it very difficult to debug the positioning algorithm in the positioning process and the efficiency of repeated field positioning is low.In this paper,an improved particle filter positioning algorithm based on the distorted is proposed to improve the positioning accuracy.By constructing map information and using particle filter to eliminate invalid particles.Due to the magnetic field is distorted due to indoor environmental factors,the user’s movement direction is difficult to be effectively estimated.At present,the treatment of this problem at home and abroad usually adopts an empirical value or predicts a fixed positioning deviation according to the real positioning direction in the data training stage,but this usually leads to the loss of positioning accuracy.In this paper,IPF algorithm are used to correct the deviation in real time so as to improve the positioning accuracy.This paper follows the relevant principles of software engineering,investigates the existing related indoor positioning technology,and carries out a more comprehensive demand analysis of the system,and carries out a summary design of the system based on the relevant requirements,and then carries out a technical survey to determine the technical stack required by the system development.And according to the function of the system module division,design and improve the database;At the same time through the front-end and back-end separation,data cache,Operation Support Systems to optimize the system;In this paper,the system related functions are discussed through time sequence diagram,class diagram,use case diagram,etc.Finally,this paper writes relevant test cases to test the functional and non-functional of the system in detail.In this paper,the positioning algorithm and system have been tested for many times.The results show that the positioning accuracy is 70%of the positioning error in the non-hollow area is within 4 meters,and 70%of the positioning error in the hollow area is within 5 meters.Meanwhile,the algorithm debugging and monitoring platform runs smoothly during the testing process,providing the algorithm developers with relevant algorithm operation data.It has good application value.