中国科学院自动化研究所机构知识库

随着我国建设“海洋强国”战略的不断推进，水下作业机器人在海底矿藏采集、水下设备维护、水中物品打捞、海产品捕捞、水下救援等应用领域扮演着越来越重要的角色。本文从仿生推进水下作业机器人的系统设计、动力学建模、路径跟踪控制以及水下自主拼装作业等方面展开研究，论文的主要研究成果包括：

一、通过在水下作业机器人两侧安装仿生波动鳍推进器以及在尾部安装两个仿生蹼推进器，将鱼类的身体—尾鳍推进模式和中间鳍—对鳍推进模式引入到水下作业机器人中，给出了一种仿生推进水下作业机器人的设计方案，并研制了实验样机。此外，构建了仿生推进水下作业机器人的控制与通信系统，设计了人机交互界面。最后，通过水池实验验证了仿生推进水下作业机器人样机的有效性和可靠性。

二、针对仿生推进水下作业机器人的动力学建模问题展开研究。首先，提出了一种基于计算流体力学数据驱动的水动力系数辨识方法，建立了仿生推进水下作业机器人的水动力模型。其次，构建了仿生波动鳍推进器的动力学模型，并基于计算流体力学研究了其推进机理和运动特性。再次，结合弹性体的柔性变形理论和微元分析法，建立了柔性仿生蹼推进器的动力学模型，分析了仿生蹼运动过程中的流固耦合现象，探究了其推进机理、参数规律以及推力效率。最后，通过仿真和实验验证了所建模型的准确性和所提数值计算方法的有效性。

三、针对水下管道跟踪问题展开研究，提出了一种基于视觉伺服的水下作业机器人路径跟踪控制方法。首先，利用ArUco识别码，给出了一种基于视觉伺服的水下管道定位方法。其次，提出了一种带有自适应权重系数的跟踪控制算法。再次，构建了控制量与仿生推进器控制参数之间的模糊规则映射模型。最后，通过水下管道跟踪控制实验验证了所提方法的有效性。

四、针对水下物品抓取及搭建问题展开研究，提出了一种仿生机器人本体与水下机械臂的协调控制方法。首先，提出了一种带中值滤波的微分跟踪器来获取机器人运动状态，基于动态面控制方法设计了仿生机器人本体位姿控制器。其次，基于计算流体力学得出水下机械臂的水动力学特性，建立了水下机械臂的动力学模型，将水下机械臂运动所产生的扰动作为控制系统的前馈补偿项。再次，分析了对撞波的推进机理，构建了波动鳍在以对撞波运动时的模糊规则映射模型。最后，通过水下自主拼装作业实验验证了所提方法的有效性。

With the continuous advancement of China's strategy of building a "marine power", the underwater vehicle-manipulator systems (UVMSs) are playing an increasingly significant role in applications such as seabed mining, underwater equipment maintenance, underwater object salvage, seafood fishing and underwater rescue. This thesis focuses on the system design of a bionic propelling UVMS (BP-UVMS), fluid mechanics modeling, path following control and underwater autonomous assembly operation. The technical contributions of this thesis are as follows:

Firstly, by installing biomimetic undulatory fin propulsors on the UVMS's two sides and two biomimetic flipper propulsors at the UVMS's tail, the fish's body/caudal fin propulsion and the median/paired fin propulsion are incorporated into the design of the BP-UVMS. The design method of the BP-UVMS and the development of a prototype are presented. Furthermore, the control system and communication system of the BP-UVMS are developed, and the human-computer interaction interface is also designed. Finally, the reliability and validity of the BP-UVMS are verified by the swimming experiments.

Secondly, the dynamics modeling problem of the BP-UVMS is researched. An identification method of hydrodynamic coefficients driven by computational fluid dynamics (CFD) data is proposed, and a hydrodynamic model of the BP-UVMS is established. In addition, a dynamic model of the biomimetic undulatory fin is constructed, and its propulsion mechanism and motion characteristics are studied based on CFD. Meanwhile, combining the flexible deformation theory of the elastic body and the micro-element analysis method, the dynamic model of the flexible biomimetic flipper is established. The fluid-solid coupling phenomenon in the motion of the flexible biomimetic flipper is analyzed, and the propulsion mechanism, parameters' law and thrust efficient are also explored based on CFD. Simulation and experimental results verify the accuracy of the presented models and the effectiveness of the proposed numerical calculation method.

Thirdly, a vision-based path tracking control method for the BP-UVMS is proposed to address the underwater pipeline tracking problem. A visual-based underwater pipeline positioning method is proposed by detecting the ArUco identification codes, and a tracking control algorithm with adaptive weight coefficients is presented. Then, the cooperative control of the multiple bionic propulsors is studied, and the fuzzy rule mapping models between the control variables and the control parameters of the bionic propulsors are constructed. The effectiveness of the proposed method is verified by the underwater pipeline tracking control experiments.

Fourthly, aiming at performing underwater autonomous grabbing and building tasks, a vehicle-manipulator coordinated control method is proposed. A tracking differentiator with median filter is proposed to obtain the BP-UVMS's motion states, and the pose controller of the biomimetic robot body is designed based on the dynamic surface control method. Meanwhile, the hydrodynamic characteristics of the underwater manipulator are obtained based on CFD, and the dynamic model of the underwater manipulator is established, then the disturbance generated by the motion of the underwater manipulator is applied to the feedforward compensation of the control system. In addition, the propulsion mechanism of the inward counter-propagating sinusoidal wave is studied, and the fuzzy rule mapping models of biomimetic undulatory fin propulsors are constructed. The effectiveness of the proposed method is verified by the underwater autonomous assembly operation experiments.