| The subsea cable system is a significant channel for world interconnection and a classical underwater infrastructure,which possesses a pivotal strategic position in the development of the marine economy and naval defence consolidation.Autonomous Underwater Vehicle(AUV),as important marine equipment,have been widely used in resource investigation,marine observation and defence,etc.The AUV-based autonomous detection for subsea cables can improve the efficiency of maintenance,reduce the cost of operation and maintenance,and reduce the economic loss and social impact caused by subsea cable failure.By investigating the research status of subsea cable detection,this thesis summarizes the roadmap of subsea cable detection based on AUV platform.Accordingly,the subsea cable-tracking AUV prototype equipped with long bow wings is developed to carry a triaxial electromagnetic array.Through electromagnetic detection system design,electromagnetic compatibility design,system modelling and maneuverability validation,it provides the basis for subsequent localization and tracking of subsea cables.In response to the low autonomy caused by the lack of autonomous cable localization and online planning of the survey line detection method,and the separation of the segmented localization algorithm and the AUV control system in the electromagnetic array detection method,this thesis conducts intelligent research on above two types of detection approaches.The highlight of the thesis is four-fold:(1)For the characteristics of electromagnetic sequences collected by the AUV on the straight survey line,an autonomous localization method for subsea cable is proposed based on the particle swarm optimization(PSO)algorithm,which reduces the sensitivity of localization to electromagnetic noise.By using the localized subsea cable locations,a planar second-order continuous detection path was designed online by splicing the Fermat spiral sections and the straight survey lines.Then the path-following control system is designed by using the line-of-sight guidance and the adaptive neural network system,which solves the problems of underactuation and uncertainties of the AUV.The smooth detection path and stable cable-tracking performance avoid the mislocating of the cable induced by sudden changes in AUV heading and attitude.Finally,the autonomous detection for subsea cable is realized by integrating PSO localization,online detection path planning and tracking control.(2)The spiral diving process for subsea cable approaching and searching is transformed into a three-dimensional(3D)path-following problem.Based on the relative motion theory,the 3D tracking error and its dynamics are derived concisely.Then the symmetrical hyperbolic-tangent guidance and kinematic controller are proposed,which avoids the potential singularities caused by multiple coordinates rotations and direct differentiating to the tracking error.Subsequently,the first-order AUV kinetic model is reconstructed as a second-order nonlinear system by introducing the actuator hysteresis dynamics and smooth saturation approximation function.The Nussbaum function is applied to achieve the adaptive estimation of an unknown time-varying gain containing the actuator hysteresis coefficients.The adaptive kinetic controller is designed to get rid of the dependence on the accurate AUV hydrodynamic parameters.Consequently,the stable AUV diving process and reliable searching for subsea cable are realized.(3)For the triaxial electromagnetic array-based subsea cable localization algorithm,the three-component electromagnetic sequences are first regularized,which eliminates the influence of AUV attitude on the localization process based on ensuring the in-situ relationship between sensors and the electromagnetic field.Then the relative location between the subsea cable and AUV is analytically derived in 3D space.The localization algorithm includes the deviation angle between the AUV heading and the subsea cable,the lateral offset and the vertical distance between the centre of the electromagnetic array and the subsea cable.Subsequently,the singularity cases and parameter sensitivity of the localization algorithm are analyzed,which provide guidelines for designing the AUVboarded electromagnetic array and planning the cable detection task.Furthermore,the consistency of the localization results under different relative positions and attitudes is demonstrated,which avoids the shaking localization result caused by switching the segmented localization algorithm.The localization algorithm provides support to design the autonomous near-bottom cable-tracking control system.(4)Considering the singularity of the triaxial localization algorithm and the cable detection target,a near-bottom cable detection architecture guided by the localization signal and along the subsea cable is proposed.The vector field guidance and kinematic control law are designed by using the localization signal and altimeter feedback,which get rid of the dependence on the tracking error dynamics and bridge the subsea cable detection system and AUV control system.The kinetic cable-tracking controller is designed by using the command filtering backstepping technique to overcome the electromagnetic noise interference in the guidance signal.Subsequently,the triaxial electromagnetic localization guided near-bottom autonomous detection for subsea cable is realized.Finally,based on the simulation studies,multiple groups of land-based simulation experiments and underwater experiments are conducted in the open water by using the developed cable-tracking AUV prototype.The comparative simulation results and underwater experiments verify the effectiveness and accuracy of the designed survey linebased and the triaxial electromagnetic array-based subsea cable detection methods. |
PDF Full Text Request