Research On Indoor Positioning Algorithm Based On IMU

With the rapid development of China's social economy and technology,Location Based Service(LBS)has played an important role in people's lives,emergency rescue and other fields.At present,although the GNSS(Global Navigation Satellite System)positioning technology provides convenient navigation and positioning services for people's travel and emergency rescue activities,it is still difficult to apply to tall buildings or complex indoor environments,especially in this case.There are emergencies underneath,and how to achieve fast real-time navigation and positioning is a key technical difficulty to be solved urgently.Based on the needs of indoor emergency response decision-making,this paper studies how the positioning infrastructure or the built-in database in the indoor scene is destroyed due to factors such as power outage and high temperature smoke in the case of indoor disaster emergency rescue.Provide location services quickly.Although the Inertial Measurement Unit(IMU)has the advantage of not relying on the indoor positioning infrastructure or the prior database,the existing IMU-based indoor pedestrian positioning algorithm has many disadvantages such as large deviation and low precision.Therefore,this paper proposes a Kalman filter algorithm based on the inertial sensor-based fusion ROF model and an adaptive gait detection algorithm for the dual state machine.The IMU data filtering algorithm and gait detection algorithm are improved,and the indoor pedestrian rapid accurate positioning algorithm is established.The model and the effectiveness of the proposed method are demonstrated by an example.The main work and innovations are as follows:(1)For the pedestrians in the process of exercise,due to body shake and uneven road surface,the IMU data is noisy.The Kalman filter IMU data filtering method combining ROF model is proposed to eliminate the noise problem in the IMU data.The results show that the proposed method has higher filtering smoothing ability and stability,and can effectively avoid the problem that the initial time estimation value and actual deviation of Kalman filtering algorithm are too large.(2)The pedestrian trajectory estimation algorithm based on inertial navigation is used to analyze the step estimation method,gait detection method and direction judgment method,and the indoor pedestrian positioning algorithm is established.Focusing on the pedestrian gait detection algorithm,based on the comprehensive analysis of the current related research,an adaptive gait detection algorithm based on dual state machine is proposed to solve the steps of different motion states and different sensor attitudes in the gait detection process.The applicability problem of state detection shows that the gait detection accuracy of this algorithm is more than 99% under different sensor attitudes and different motion states(running + walking mixed state).(3)Through the example verification,the results show that the Kalman filtering algorithm of the fusion ROF model and the adaptive gait detection algorithm based on the dual state machine proposed in this paper use PDR to achieve the maximum point medium error of pedestrian positioning of 0.36 m,the minimum point medium error is 0.02 m and the average point medium error is 0.23 m.