Research On Indoor Location Algorithm Based On Pedestrian Dead Reckoning And MEMS Inertial Sensor

Indoor positioning technology has been the focus of research in recent years, due to the blockade of the building can not receive the GNSS signal,it is hard to accurately obtain the staff location information. At present, the realization of indoor positioning can be divided into two kinds: one is dependent on the wireless positioning network to receive the signal strength indicator (Received Signal Strength Indication, RSSI)wireless indoor positioning mode, the other way is entirely dependent on their own sensor independent indoor positioning. Although the current RSSI-based wireless indoor positioning is more used by people, but it needs to be deployed in advance indoor signal transmission device and the characteristics of the fingerprint library led to its application is very limited. Such as the city firefighters in the face of the complex structure of the building's fire, because before the fire there is always have no sign and the fire environment is bad, any need to lay out in front of the indoor signal facilities to assist the positioning of the way are not suitable for firefighters positioning, and this similar to the underground mine personnel positioning.In this paper, indoor positioning research is based on the premise of independent indoor navigation and positioning, based on the current common Android mobile phone GNSS chip and MEMS inertial sensors, combined with the traditional strapdown inertial navigation and positioning algorithm based on pedestrian dead reckoning (Pedestrian Dead Reckoning, PDR) and zero velocity update(Zero Velocity Update, ZUPT) for pedestrian inertial navigation and positioning, the main research content includes the following aspects:1. Real-time transmission of sensor data between the Android side and the PC side.In this paper, the use of the transfer program is to call the way the Android phone gyroscope, accelerometer and GNSS chip and obtain the data, the WLAN network is used for real-time transmission to the PC side, and analysis date for solution.2. The preparation and analysis of traditional strapdown inertial navigation algorithm. Unlike the inertial navigation of the platform, the strapdown inertial navigation can not be used without the help of the mathematical platform. For the solution, it needs the real-time correction, the quaternion correction algorithm and the direction cosine matrix correction algorithm of coordinate transformation, attitude matrix and position matrix correction algorithm and so on. In this paper, according to the traditional theory of strapdown inertial navigation,we learn the corresponding inertial navigation algorithm theory, the overall derivation, algorithm writing,program implementation, precision analysis and so on.3. Peopposing a new inertial navigation algorithm based on pedestrian dead reckoning and zero velocity update. Because of the large error caused by the traditional strapdown inertial navigation algorithm in the indoor positioning application, this paper introduces the pedestrian dead reckoning algorithm and the zero velocity update to make the velocity error of the inertial sensor not accumulate,Greatly reducing the impact of the algorithm error.4. Based on the three-dimensional model of the experimental space, the indoor navigation and positioning algorithm is tested and analyzed. This paper chooses the computer reading room of Tunxi Road campus of Hefei University of Technology as the experimental site. Establishes the three-dimensional model based on the actual situation and import the experimental data. During thar time we do a number of straight-line walking test and rectangular route walking test, the final test results of the error in the total length is less than 5%.