The Research On Intelligent Perception Method Of Moving Target Based On Motion Sensor

With the continuous innovation of sensor technology and the wide demand for location services in society,many service industries have increasingly higher requirements for mobile target perception,especially in the field of urban traffic perception.Data fusion technology based on motion sensors has become one of the indispensable technologies for scientific management and rational planning of urban transportation.Location service is one of the key components in intelligent transportation.The GNSS system represented by GPS,BDS,GLONASS,Galileo,and the motion sensor series equipment represented by inertial devices provide important technical support for the intelligent perception and intelligent manipulation of urban traffic.Among them,the perception system represented by the combination of motion sensor and GNSS system can not only implement a large number of applications,but also drive the development of some social pillar industries.For the purpose of manage urban traffic effectively and reduce the cost of urban traffic operation and maintenance,the development of motion sensors and the application of data fusion technology have become the focus of research.Thesis has conducted related research from four aspects.Data noise reduction processing method based on motion sensor.Target posture perception method based on motion sensor.Target orientation speed measurement method based on dual-antenna GNSS auxiliary motion sensor.Target positioning method based on low-cost vehicle-mounted motion sensor combined GNSS system.The main research results of dissertation are:(1)A data noise reduction method of MEMS gyroscope motion sensor is designed.Introduce the wavelet analysis method to analyze the error characteristics of the MEMS gyro signal.According to the analysis result of the error characteristics,a processing method of wavelet analysis combined with singular values is designed to eliminate the weak noise signal.(2)Aiming at the problem of attitude perception of moving targets,a filter structure for quaternion fusion between accelerometer and gyroscope is designed,and an improved weighted least squares algorithm is introduced to reduce gyroscope drift and improve the accuracy of attitude perception.(3)Aiming at the problem of directional speed measurement of moving targets,a GNSS/IMU directional speed measurement method based on dual-antenna assistance is studied to complete static alignment and alignment between targets.In addition,more in-depth research has been done to suppress azimuth divergence.(4)Aiming at the problem of combined positioning of GPS and low-cost on-board motion sensors,a nonlinear filtering fusion method to quantify the uncertainty of the system is proposed.When there is initial attitude error,the concept of system uncertainty is introduced according to the meaning and statistical characteristics of innovation,so that the working state of the system filter can be quantified,and the online adaptive adjustment system model can be realized.Finally,an improved adaptive filtering method is embedded in the unscented Kalman filter framework to form a fuzzy IAE-UKF filter suitable for processing low-cost integrated navigation systems.Finally,simulation test and semi-physical simulation test are designed,including static and dynamic turntable test,cart test and vehicle test.The effectiveness and superiority of the proposed method are verified,which provides a certain reference value for the intelligent sensing method of moving targets based on motion sensors.