| With the increasing development of modern building technology,more and more high-rise buildings have been built,and the service demand of indoor positioning,such as large-scale place navigation,fire rescue,individual combat and other fields has also increased.At present,the wireless network indoor positioning technology is affected by the signal weakness and multipath effect in the indoor environment,and there is a positioning blind area.In order to make up for the lack of short-range signal,a low-cost and small volume MEMS Inertial Measurement Unit came into being.By studying the application of IMU in indoor positioning,the effects of errors in the process are explored,effective optimisation methods are proposed,and a hardware and software platform is designed.The main contents are as follows:(1)The shortcomings in the mainstream positioning techniques are pointed out and an improved solution for indoor positioning based on inertial measurement units is proposed.By summarizing the relevant research at home and abroad on reducing the error of the IMU and improving the positioning accuracy,the key research direction and feasible processing methods are made clear.(2)In the data pre-processing stage,an algorithm based on fractional Fourier filtering is used for noise reduction for the random errors still present in the original data after the initial alignment.Because of the complexity and randomness of noise in inertial data,signal and noise can not be effectively separated in time domain or frequency domain.This method determines the amount of rotation by parameter estimation,and carries out FRFT transform to separate energy in fractional Fourier domain.The experimental results show that this method can effectively suppress noise and recover acceleration and angular velocity signals.(3)In the data fusion stage,to address the problem that the quality of the UKF state estimation is limited by the scale parameters,a UKF algorithm based on adaptive tuning is used to improve the quality of the state estimation by numerical techniques to select the best scale parameters in the system at that time before filtering at each moment,thus providing effective error compensation for the heading angle information.(4)The built hardware and software system,using three-dimensional positioning map,has a more intuitive positioning effect,while ensuring the stability of the system and the accuracy of the positioning accuracy.In terms of hardware,STM32F103 is selected as the core processing unit of the whole system.According to the functions needed in the system,strict selection and related circuit design are carried out.In the aspect of software,the method of data processing is integrated into the positioning and navigation program to provide a program design scheme for platform driver and data transmission. |
