Research On The Multi-sensor Asynchronous Fusion Methods Of Dynamic Positioning System

Because of exploiting and using the marine resource is the vital important to human society,more and more project will be set including undersea tunnel engineering,the bottom of the sea oil ang gas and cable laying connect with land.So for resources development and the project operation of ship equipped with the technology and equipment are also put forward higher request.Application in ship dynamic positioning technique makes for more effective development and utilization of Marine resources.From multiple sensors asynchronous measurement ship location information is a key technology in the study of ship dynamic positioning.In this paper,we use interactive multiple model based on covariance matching the volume of kalman filter(IMM-CMASRCKF)algorithm to solve the problem of different time sampling multiple sensor fusion,first of all,the signal using volume kalman filter based on covariance matching filtering(CMASRCKF)to deal with the use of IMM fusion method of fusion processing.Accounting to the movement of ship dynamic position,it built ship six degrees of freedom model including north-east coordinate system,geocentric coordinate system,and fixed coordinate system and ship-body coordinate system.Meanwhile,it also built the model of measurement model to DGPS and underwater acoustic system even the tension rope measurement system model.It abstracts the equation of measurement and observation,this will pave below the ship sensor to capture the signal.It also built the model of wind,wave and flow.Then it study about the sensor pre-process the measurement data and fault detect.Sensor measurement pre-process data include time temporal registration,space registration,even the quality test of sensor measurement.And then study the way of quality test of sensor,this include outliers detection,detection of variance,range detection and freezing detection and predicted detection even difference predict and the median detection.In the study of nolinear filter in ship coordinate system,CKF and SRCKF is the better filter in the filter.Because of the problem in statistical characterizations of system noise,filter will be weaken.Thus,bringing in the covariance matching technique,the innovation and residuel sequences are used to determine the covariance matrices of the process and measurement noises.It detrives CMASRCKF.The proposed algorithm can estimate and adjust the system noise statistic online,and thus enhance the adaptive capability of the SRCKF.Simulation and experimental results demonstrate that the performance of the proposed algorithm.Finally,this paper use IMM fusion algorithm to deal with the problem of asynchronous sensors.Using the construction of asynchronous sensors and make use of theory of IMM fusion,it combine CMASRCKF and get the IMM-CMASRCKF fusion algorithm.Then,compareing it with IMM-SRCKF,it get the different of these filter.