Analysis Of Indoor Positioning Method Combining Indoor Map And Inertial Sensor Information

With the rapid development of information technology and the popularity of smart phones,Location Based Service(LBS)plays an important role in more and more applications,and indoor positioning technology has gradually become a hot spot of research.At present,in the indoor positioning technology,the inertial sensors can be used to calculate the path of the traveling person by calculating the step and direction of each step in the pedestrian walking process.The PDR method has cumulative error,and the location scheme that is fused with WiFi can eliminate the cumulative error to a certain extent,but WiFi and other wireless location dependent infrastructure still lack such infrastructure in many indoor environments.The existing method of map constraint has the disadvantages of low location accuracy and high complexity.In view of the above problems,from the perspective of indoor space information and its model,this thesis studies the indoor location method based on particle filtering without infrastructure.Firstly,based on the indoor landmark perception method of smart phone inertial sensors,an enhanced particle filter location method,which combines PDR,landmark and map information,is proposed to bind the particles through the map information,and to correct the cumulative error by identifying the landmarks in the environment,and to make it different by refining the parameters of the step length model.Then,based on the requirements of mobile devices such as smart phones for low power and low complexity,a PDR indoor pedestrian location method based on the map model is proposed.At the same time,a particle filter propagation model based on graph model is established,and the error of location estimation caused by PDR accumulation error is further eliminated by the direction calibration method based on the graph model and the particle backtracking method.The experimental results show that the enhanced particle filter positioning method can reach the positioning precision of 2.8m under the action of 200 particles in 80% cases,and the positioning accuracy is improved by 20% compared with the existing particle filter positioning method only using map constraint.The function of the PDR chamber localization method based on the map model is used in the 50 particles.Under 80% conditions,the positioning accuracy of 2.7m is achieved.Compared with the traditional particle filter fusion positioning method,the number of particles is reduced to the original 1/4,which greatly reduces the number of particles required in the pedestrian movement model,reduces the complexity of the algorithm,and the precision and complexity are all required for the design index.