Dynamics Analysis on Underwater Articulated Structure without Parameter Estimation

在这项研究中，我们提出了一种新的无参数估计的水下环境下人类手臂和机器人关节力矩的动力学分析方法。研究成果总结如下：[1]水下机器人机械手的迭代学习控制从理论和实验上验证了迭代学习控制的有效性。实验中使用了单自由度和三自由度水下机械手。[2]基于学习控制和时间尺度变换的水下机器人机械手动力学分析在该方法中，不使用参数辨识，而是对附加质量、阻力和浮力等水动力项进行了定量计算。实验实现了单自由度和三自由度机械手的动力学分析。机械手控制的前馈输入为…[3]利用平行绳索驱动系统测量人体运动的关节扭矩我们研制了一种用于测量人体关节扭矩的平行绳索传动系统。利用该系统实现了关节扭矩的精确测量，最大误差在0.2[Nm]以内。[4]并联绳索系统的迭代学习控制从理论和实验上验证了迭代学习控制的有效性。数值仿真和实验结果表明了控制性能的影响。[5]任意人体运动关节力矩的测量基于并联绳索驱动系统和迭代学习控制的方法实现了对任意人体运动关节力矩的测量。通过实验验证了该方法的有效性。[6]基于并联绳索驱动系统的水下人体关节力矩测量采用基于并联绳索驱动系统和迭代学习控制的方法实现了水下人体关节力矩的测量。在该方法中，对作用在人体手臂上的附加质量、阻力和浮力等流体动力矩进行了定量评估。较少

In this research, we propose a new method of dynamics analysis on joint torques of human arms and robot manipulators in underwater environments without parameter estimation. The research results are summarized as follow:[1] Iterative learning control for underwater robot manipulatorsThe effectiveness of iterative learning control was verified theoretically and experimentally. The single and three degrees of freedom underwater manipulators were utilized in the experiments. As a result, the, motion of the robots in the water was realized as precisely as that in the air.[2] Dynamics analysis on underwater robot manipulators based on learning control and time-scale transformationIn the proposed method, hydrodynamic terms such as added mass and drag and buoyancy were quantitatively evaluated without using parameter identification. Dynamics analysis on the single and three degrees of freedom manipulators was realized experimentally. The feedforward inputs for the manipulator control were for … More med from the method.[3] Measurement of joint torques of human motions using parallel wire driven systemsWe developed a parallel wire driven system for the measurements of joint torque of human motions. By using the system, the joint torque was precisely measured and the maximum error is within 0.2[Nm].[4] Iterative learning control for parallel wire driven systemsThe effectiveness of iterative learning control for parallel wire driven systems was verified theoretically and experimentally. Numerical simulations and experimental results showed the effects of the control performance.[5] Measurements of joint torques for arbitrary human motionsMeasurements of joint torques for arbitrary human motions were realized by the proposed method based on the parallel wire driven system and the iterative learning control. The validity was verified by some experiments.[6] Measurement of joint torques of human motions in underwater environment using parallel wire driven systemsMeasurements of joint torques for human motions in underwater environments were realized by the method based on the parallel-wire driven system and iterative learning control. In the method, hydrodynamic torques such as added mass and drag and buoyancy acting on human arms were quantitatively evaluated. Less