Research On Pedestrian Autonomous Navigation Algorithm

Navigation is a general term for a category of technology.It is a means of monitoring and guiding people or machines,etc.,from one location to another.However,the weaknesses of satellite navigation systems based on GPS are also very obvious.In the indoor areas of urban buildings,tunnels,forests,underground buildings and even general buildings,the signal reception of satellite s is very poor.At this time,active navigation is represented by GPS.It is difficult to play an efficient navigation role.Inertial navigation has many advantages.For example,it has good autonomy and can be undisturbed by unfavorable external conditions.It can continuously update navigation information such as position and posture in a relatively complicated navigation environment.In recent years,the MEMS-based strapdown inertial technology has made great progress.Compared with general inertial components,it has a small size,low cost,high reliability,and is easy to integrate.It can be applied to pedestrian autonomous navigation research.First,based on the basic principles of inertial navigation,the initial alignment,posture update,and navigation solution design of the inertial navigation system are completed.There are detailed derivation for position,velocity,and attitude,and error analysis of the model.Secondly,in the navigation solution process,due to the low precision of MEMS devices,there is a drift error in the sensors,and there is an accumulated error in the results of the navigation solution.In this paper,the zero-speed correction algorithm is used to compensate for the accumulated error.The zero-speed correction can be divided into two parts,zero-speed detection and correction.In this paper,the zero-speed detection is deeply studied,and the adaptive zero-speed detection is realized in the case of multi-gait motion based on the random forest algorithm;the zero-speed moment correction is based on the error estimation of the traditional Kalman filter,and the position of this moment is estimated.Error,speed error,attitude angle error value,thus further error compensation,improve navigation and positioning accuracy.Finally,in the pedestrian autonomous navigation algorithm,due to the larger drift error of the skyward gyro,the heading angle diverges during the navigation and positioning process.Therefore,based on the zero-speed correction,the heading-based heading correction algorithm based on the heuristic heading correction algorithm is also designed and improved to improve the accuracy of the heading estimation.Based on the research of this algorithm,this paper carries out the experimental analysis of the navigation system under the algorithm.Through the experimental verification,under the multi-gait motion,the system error converges and the positioning accuracy is good.