| Underwater high-precision and high-reliability navigation and positioning is the core foundation for autonomous underwater vehicle(AUV)to complete intelligent and diversified operations.However,in the face of the underwater due to electromagnetic signal attenuation,the need for concealment and the complex unknown seafloor topography,ocean currents,magnetic field environment and unpredictable aquatic organisms and other complex environmental interference,resulting in fewer available information sources in the water,the sensor observation information due to the influence of the environment and large error problems,so that the current navigation algorithm is still difficult to meet the AUV underwater required navigation performance requirements.Based on this,the following research is carried out around multi-source information fusion algorithm and fault detection and self-organization reconstruction algorithm:(1)To address the problem that the error characteristics of underwater sensors caused by interference in the current underwater environment have not been clarified,the error characteristics of each sensor are extracted from the measurement principle and error sources of the sensors,and the error characteristics are analyzed from multiple perspectives,such as underwater applicability,probability distribution characteristics,fault tolerance and stability,etc.The results of the simulation experiments show that using only the Inertial Navigation System(INS)can lead to the error in the navigation.Navigation System(INS)will lead to navigation error will be accumulated over time,so other auxiliary navigation correction inertial error is needed.The Beidou Navigation Satellite System(BDS)has high accuracy but it is only applicable when the AUV is located at the surface or shallow water depth;the Doppler Velocity Log(DVL)acoustic information is easy to propagate underwater but it is susceptible to interference from the complex underwater environment and The acoustic information of Doppler Velocity Log(DVL)can be easily disseminated underwater but it is susceptible to interference from the complex underwater environment and poor concealment;Magnetic Compass(MCP)is independent and autonomous but it has low accuracy and is susceptible to interference from other magnetic fields.The above mentioned,independent sensors can hardly meet the actual AUV navigation requirements,so multi-source fusion is needed to achieve complementary advantages in order to achieve the expected AUV navigation performance requirements.(2)In response to the traditional multi-source fusion filtering model is tight and leads to more difficult identification and isolation of fault systems and poor fault tolerance,an inertial-based underwater multi-source fusion algorithm based on federal filtering is studied,with INS as the core and deep fusion of BDS,DVL and MCP to form a multi-source fusion navigation complementary advantages.For the variability of accuracy and error characteristics of each sensor under underwater complexity,the dynamic adjustment strategy of adaptive information allocation coefficients is constructed,and the allocation coefficients are readjusted according to the navigation performance of each subsystem to guarantee the optimal estimation of multi-source fusion navigation model and provide technical support for high accuracy of underwater navigation.(3)To address the problem that the underwater complex environment interferes with the navigation performance of each sensor to reduce the reliability of integrated navigation,the underwater multi-source fusion fault detection and self-organizational reconstruction algorithm is proposed,and the fault detection method based on the observation residuals of vectorized detection is constructed based on the statistical characteristics of each sensor error.At the same time,the system self-organized reconstruction algorithm is formed based on the fault detection results,and the optimal reconstruction is implemented for different sensor weights,so as to improve the system accuracy and fault tolerance performance and provide technical support for underwater navigation reliability.In response to the problem that the navigation system accuracy and reliability are reduced due to the complex underwater environment,the research of error characteristics analysis and modeling of each underwater navigation equipment,multisource information fusion algorithm based on adaptive federal filtering and vectorized fault detection and self-organized reconstruction algorithm provides high accuracy and high reliability navigation and positioning information for AUVs,thus favorably guaranteeing the intelligent and safe operation of AUVs underwater. |
