Research On The Indoor Location Algorithm Based On Multi-sensor Fusion And Map Information

With the continuous development of the information age,location-based services play an indispensable role in people's entertainment times,medical rescue,and life services.Global positioning systems(GPS)has a good performance in outdoor positioning.However,due to the complex indoor environment,satellite signals suffers from severe attenuated,which making indoor positioning unable to use GPS,so indoor positioning has become a new hot area of research.Smart terminals with inertial devices are gradually being widely used,the application of Pedestrian Dead Reckoning(PDR)based on inertial navigation in indoor positioning has attracted widespread attention.However,the position calculation of the PDR is based on the positioning information at the previous moment,so that the errors introduced in the calculation process are continuously accumulated,which leads to trajectory deviation.Therefore,other methods need to be introduced to merge with the PDR algorithm to improve the positioning accuracy.The related research work of this thesis mainly focuses on the problems such as the deviation of the direction data collected by the direction sensor of the intelligent terminal and the accumulation of PDR positioning error.According to the research and development of indoor positioning technology,the relevant theories,advantages and disadvantages of PDR positioning are cleared.Fusion fuzzy logic is proposed to improve the accuracy of direction angle calculation,and the WiFi landmarks,map information and particle filtering are fused to improve PDR positioning.The main work contents are as follows:(1)The direction angle calculation in pedestrian track estimation has been improved,and the adaptive fusion algorithm of direction angle based on fuzzy logic is proposed:first,determine whether to use the fusion algorithm or simply use a certain optimal state direction Angle data:The degree of deviation of the direction angle is a fuzzy proposition.It can be used as a decision criterion to set the relevant theorem,determine the fuzzy logic inference rules and the output area,and calculate the membership on the output area to obtain a suitable algorithm to calculate the walking direction.Experimental results show that the algorithm can increase the adaptability of the algorithm to improve the calculation accuracy of the orientation,further improve the positioning accuracy.(2)For the problem of PDR error accumulation,this thesis proposes a particle filtering algorithm that fuses map information and WiFi landmarks:while using the existing map information to limit the position of the particles to a valid position,the signal peak is obtained according to the change trend of the WiFi signal received during walking to obtain limited landmark points,and then the landmarks are used to calibrate the PDR trajectory as the filtered observation information,and the landmark will participate in updating particle weights and position,finally obtain the positioning result.Experimental results show that the algorithm can effectively mitigating the cumulative error of PDR.