【作者】 王翔宇；

【导师】 刘凯；

【作者基本信息】 大连理工大学 ， 船舶工程（专业学位）， 2023， 硕士

【摘要】 自主式水下航行器(Autonomous Underwater Vehicle,AUV)是一种具备在水下环境自主运行、执行任务功能的机器人。AUV具备收集海洋中的数据、勘探海底矿产资源、执行水下侦察任务、检查海底管线的安全状况和执行海洋污染监测等功能,因此研究AUV有着重要意义。由于海底工作环境复杂,航行过程中舵面故障等问题难以避免,为了面对岩石和海草区域等极具挑战的环境,AUV在恶劣条件下的控制能力和品质需要进一步提升。其中姿态控制是指控制AUV的方向、姿态、位置等状态,以实现预定的任务目标,对AUV的性能和功能具有非常重要的影响,因此,为了满足自主式水下航行器故障情况下的姿态控制需求,本文将采用递推最小二乘辨识方法来进行水动参数的辨识研究。在此基础上,将开展自适应干扰抑制姿态控制方法的研究,以提升自主式水下航行器在复杂环境下的控制能力。最后,本文将通过仿真实验来验证所提出方法在自主式水下航行器舵面损伤故障情况下的应用效果。具体分为以下几个部分:(1)在水动参数在线辨识过程中,为了满足对水动参数在线辨识过程中精确性和实时性的高要求,首先,需要设计恰当的参数选择方法,以确保所获得的水动数据具有高精度和足够的信息量。其次,在数据处理方面,使用巴特沃斯滤波器来降低噪声干扰,从而提高数据的可靠性和准确性。最后,为了保证水动参数的在线辨识结果具有时效性和精度,并提高计算效率和简化计算复杂度,采用递推最小二乘方法进行水动辨识的在线应用。通过数值仿真实验的验证,验证了该方法在常规状态和舵面损伤状态下的计算精度和效率,并且其辨识精度和实时性能够满足姿态控制的需求。(2)在自主式水下航行器的舵面损伤容错控制环节,若舵面发生结构缺失故障,流体动力学特性会发生显著变化,这将对预设标称控制器的跟踪性能造成严重影响。为了解决这个问题,设计了一种基于递推最小二乘在线辨识的自适应干扰抑制控制方法。该方法将在线实时模型辨识、非线性动态逆和误差观测器相结合,利用在线辨识计算出的无因次力和力矩模型来控制自主式水下航行器。通过与标称控制器闭环仿真的比较,结果表明该控制方法可以提高自主式水下航行器的控制品质、操纵稳定性。(3)在硬件在环仿真环节,利用NI仿真机,将本文将研究方法应用于舵面损伤后的水下航行器控制中,针对不同程度的舵面缺失故障,进行自适应干扰抑制姿态容错控制。通过实验验证,本文研究方法证明了其有效性,并展示了其在实际工程应用中的可行性。更多还原

【Abstract】 The development of Autonomous Underwater Vehicle(AUV)has a significant research value as they are an integration of energy and propulsion technology,sensors and signal processing,communication and navigation,and automatic control technology.AUV are standardized,modularized,and have autonomous capabilities,making them a unique underwater platform with high safety,low cost,small size,lightweight,high flexibility,and a broad range of activities.They are useful for long-term observation,detection,and salvage operations,with applications in many fields,such as military,scientific,and economic sectors.However,the complex underwater environment poses significant challenges to the control of AUVs.When navigating through rock or seaweed areas,the control faces severe challenges,and rudder failures are inevitable during the navigation process.Therefore,enhancing the control quality and capabilities of AUV under harsh conditions is of great engineering significance.In order to fulfill the attitude control needs of Autonomous Underwater Vehicles(AUV),this article focuses on the study of hydrodynamic parameter identification using recursive least squares method.Additionally,an adaptive disturbance suppression attitude control method will be developed based on the identified model and parameters.Finally,the effectiveness of the proposed method will be demonstrated through simulation of an autonomous AUV with damaged rudders.Overall,this study is composed of the following main parts:(1)In the online hydrodynamic parameter identification process,to ensure the high accuracy and timeliness of hydrodynamic data and sufficient information,a suitable parameter selection method will be designed.In the data processing,a Butterworth filter will be used to reduce noise interference and improve the reliability and accuracy of data.Finally,a recursive method will be used for online hydrodynamic identification to ensure the timeliness and accuracy of the identification results,improve the calculation efficiency,and simplify the computational complexity.The weighted processing method will be used to obtain the global model and the nonlinear combination of segment models.The simulation results will demonstrate that the method has high computational accuracy and efficiency under regular conditions and damaged rudder conditions,and the identification accuracy meets the control requirements,while ensuring real-time online use.(2)In the rudder damage fault-tolerant control process,when the AUV rudder encounters structural failure,the fluid dynamic characteristics will significantly change,and the preset nominal controller tracking performance will be severely affected.Therefore,a non-linear recursive least squares identification-based adaptive disturbance suppression control method will be designed to control AUVs with structural rudder failures.The method combines realtime model identification and non-linear dynamic inverse control design,using the dimensionless force and moment models calculated from the identification model to control the AUV.The simulation results show that the adaptive disturbance suppression controller improves the control quality,operational stability,and system response speed of AUVs and reduces overshoot during the control process compared to the nominal controller.(3)In the hardware-in-the-loop simulation process,the NI simulation machine will be used to apply the methods developed in this article to control the AUV after rudder damage failure.The adaptive disturbance suppression attitude fault-tolerant control method will be applied to simulate the AUV under different degrees of rudder failures,demonstrating the effectiveness of the developed method in terms of real-time identification and control.更多还原