Research On Key Navigation Algorithm Of Dual-Function Deep Sea Vehicle

As a powerful means of ocean exploration,deep-sea vehicles have the advantages of small size,good concealment and wide range of activities,which play an important role in both military and civil fields.In order to meet the engineering requirements of long-term migration and short-term high-precision detection,the deep-sea vehicle should have gliding and cruise working modes.When the deep-sea vehicle is in glide mode,it is in the form of underwater glider.The system uses low-precision MEMS-IMU(Micro Electro Mechanical System-Inertial Measurement Unit),which has very low power consumption and can guarantee long-term endurance;When the deep-sea vehicle is in cruise mode,it is in the form of autonomous vehicle.The system uses high-precision FOG-IMU to ensure the high positioning accuracy of underwater detection.In this paper,the key navigation algorithms in the workflow of deep-sea vehicle are studied.The main work of this paper is as follows:(1)The software flow and algorithm structure of the dual-function deep-sea vehicle are designed based on the analysis of the navigation requirements.The navigation requirements of the actual working process of deep-sea vehicle are analyzed,and corresponding navigation simultaneous interpreting algorithms are designed according to different sensing devices enabled under different working modes..When the deep-sea vehicle is in the low-precision glide mode,the system uses MEMS-IMU and magnetometer,which only needs to reduce the power consumption as much as possible on the basis of ensuring a certain accuracy;when the deep-sea vehicle is in the high-precision cruise mode,the system uses fiber-optic IMU,and cooperates with Doppler velocimeter to achieve high-precision navigation and positioning.(2)On the basis of establishing the alignment error model of large azimuth misalignment angle,the alignment technology based on strict reverse navigation algorithm is studied.Firstly,based on the case of sea wave,a large azimuth error model of moving base is established;secondly,the strict reverse navigation algorithm with process control is improved;Finally,the algorithm is verified by simulation and experiment.The simulation results show that the improved algorithm can reduce the alignment time by nearly 60% on the basis of ensuring the alignment accuracy.(3)Aiming at the high-precision navigation requirements of deep-sea vehicle in complex and changeable underwater environment,the Sage-Husa adaptive filter is improved three times.Firstly,aiming at the problem of too much computation caused by the simultaneous estimation of Q and R,this paper proposes only iterative estimation of R.Secondly,in the process of R updating,the positive qualitative judgment of R is added to avoid the divergence of filtering.Then,aiming at the problem of the decrease of the adaptive degree,the method of periodic updating of the adaptive coefficient is adopted to improve the tracking ability of the measurement.The experimental results show that the improved algorithm can reduce the RMSE of velocity by more than 70% compared with the traditional Sage-Husa adaptive filter on the basis of ensuring the stable convergence of the filter.(4)The improved adaptive federated filter is applied to the mode switching process of deep-sea vehicle.Based on the derivation of the structure of the federated filter,the method of information allocation is improved adaptively.The precision factor is constructed by using the covariance matrix P,the state error of the filter,and the information allocation between the main filter and the sub filter is carried out based on the precision factor;Secondly,the design of each sub filter is completed and its filtering model is also realized.Finally,the simulation results show that the adaptive federated filter is more stable than the traditional Kalman filter,and the position root mean square error is reduced by more than 42%.