Inertial Sensor-based Indoor 3D Positioning Research And Implementation

Nowadays,the geographic location has a wealth of information and widely used in daily life.Most of these services are based on the GPS.Because GPS dependent on satellites of positioning so that it is limited to scenarios without cover outdoor,The canyon effect in the city and the different degrees of occlustion in the room has seriously restricted the application of GPS.Now,most of people's activities in the indoor environment,indoor positioning is getting more and more attention.Currently there have been a lot of focus on indoor 2D positioning scheme,but 2D positioning has been unable to meet the needs of indoor positioning.Indoor 3D positioning although has some solutions and eauipment,but due to the price,equipment deployment range limitations and limited by the user's special,so it has not perfect solution of indoor 3D positioning.Under the background of this application,the research of indoor 3D positioning with low cost and low power consumption has great significance.This paper is based on the low cost inertial sensors to complete the indoor 3D positioning.The main work and contribution:First,this paper analyzes the error sources of inertial sensors,and quantitatively analyzes the characteristics of the error propagation,the gyro is the main factor that causes the data drift.By means of an integral of the acceleration data,the velocity of data is obtained from the velocity level,which is a prerequisite for the elimination of the accumulated error.Secondly,the algorithm based on strapdown inertial navigation system and mounted on shoes.According to the gait information,data drift characteristics,using the ZUPT(zero velocity update)to design the cumulative error elimination method.Based on the characteristics of inertial sensor acceleration in static condition,the data is more stable than the gyroscope data,with the acceleration as reference,combined with the gait information to eliminate the data drift of gyroscope.In view of the importance of gait information to eliminate the error.This paper uses the method of short time energy method to extract the gait information.To reduced plane motion perpendicular to the direction of the error.Firstly,this paper according to the characteristics of the data when walking design walking state estimation model for distinguishing plane motion(normal walking,jogging),upstairs,downstairs three states,according to the walking state judging model results to design the vertical distance error elimination strategy,in order to improve the positioning accuracy.Finally,the indoor 3D positioning system based on inertial sensor is designed.And use Unity3 D to draw the floor of the 3D map,In the map mapping shipped 3D vector trajectory which Completed indoor 3D positioning.The experimental results show that it has a better 3D positioning effect,and it has important significance for the study of 3D positioning in the later stage.