Research And Implementation Of The Indoor Pedestrian Autonomous Location Algorithm

The fields of military pedestrian system,firefighting and large venues have a very large demand of indoor positioning and navigation.As the GNSS navigation will appear attenuation,multipath and other issues in a complex indoor environment,and the indoor location based on the base station has the status of high cost and poor autonomy,the Personal Dead-reckoning navigation system based on low-cost MEMS inertial sensors become a hot topic in the field.After the research of this field and analyzing the indoor positioning experiment,in this paper,the algorithm of pedestrian autonomous positioning is designed and the algorithm is used to solve pedestrian location in real time based on the strapdown inertial navigation system.The system module is worn on the pedestrian foot,and can complete the pedestrian location in the indoor and outdoor environment.In particular,the main work done in this paper are the following aspects.First of all,according to the requirement of precision,real-time and size of the indoor pedestrian location system,this paper introduces a complete design of the pedestrian autonomous positioning system.As low-cost MEMS inertial sensors have deterministic bias and random noise,we analysed the error sources of the accelerometer and gyroscope,which the corresponding error model was established.The parameters in the model were determined by the angular velocity experiment,six-position static calibration experiment,temperature calibration experiment.The error compensation method of the sensor is proposed.When the sensor is in different temperature,using the temperature calibration results to compensate the bias and scale factor of the gyroscope,and the drift of the yaw angle is restrained to a certain extent.Secondly,by analyzing the gait features of pedestrians,the pedestrian walking stage is divided into static interval and non-stationary interval,the generalized likelihood ratio test is proposed to detect the pedestrian zero-speed interval,which can improve the accuracy of detection and lay a good foundation for the zero velocity correction algorithm.Then,the attitude updating algorithm based on gradient descent is adopted to integrate the information of accelerometer,gyroscope and magnetometer to update the pedestrian attitude.Compared with the Kalman filter fusion multi-sensor information method,thisalgorithm obtains better fusion performance with smaller computation amount and lower sampling frequency,which provides a powerful condition for realizing the real-time system.The system's strength lies in the use of a technique known as “Zero Velocity Update”(ZUPT)that virtually eliminates the ill-effects of drift in the accelerometers.It works very well with different gaits,such as normal walking,running,up and down stairs.Finally,through the design of several experiments in the indoor environment,the correctness and feasibility of the pedestrian autonomous positioning algorithm are analyzed and validated.And we completed the verification of pedestrian autonomous positioning system and achieved real-time acquisition,calculation of sensor information and sent positioning results to the display terminal.