基于模糊神经网络的水下机械手的运动轨迹规划

运动轨迹规划对水下机械手的研究和设计具有重要的影响,本文提出了一种基于模糊神经网络的水下机械手的运动轨迹规划方法.文章首先简要叙述了水下机械手的发展现状和存在问题,然后通过分析抓取任务及环境信息的获取,对水下机械手进行了数学建模.接着重点阐述了规划器的结构及工作原理.最后通过仿真对轨迹规划方法进行了验证.     

面向水下抓取作业的复合腔体仿生软体手设计

水下机械手在水生物采样、考古打捞等水下作业中起着关键作用,而现有水下机械手存在耦合安全性差、适应性弱及抓取不稳定等问题,使得作业效果并不理想。针对以上问题,本文开展了新型水下软体手的设计与研究。首先提出了具有密封管结构的复合腔体仿生软体驱动器,基于该驱动器设计出一种三指包络型水下软体手;硅胶材料的弹性和流体的可压缩性能够保证目标物的无损抓取,密封管结构可提高深水高压适应能力,保持腔和弯曲腔的复合结构及指纹、指甲的仿生结构提升了软体手耦合适应能力。利用基于Yeoh模型的有限元分析方法及注塑、3D打印等加工技术对软体手进行了结构优化和制作。设计了一套具有较高精度的水压驱动系统来提高软体手耦合稳定性。最后,针对影响机械手水下作业能力的重要因素,模拟相应实验场景对软体手进行了实验测试,实验结果显示软体手纵向抓取力达到26 N,驱动深度可达3000 m,较现有水下软体手均有较大提升;水下目标物抓取测试及与传统水下刚性手的对比实验证明了所提出软体手具有稳定的抓取性能和更好的适应性及安全性,适用于水下目标物的无损抓取作业。     

绳驱动纱筒抓取仿生机械手设计

针对纱筒上下料对人力过度依赖的问题,在研究仿生学手指基础上,构建面向智能制造的纱筒抓取仿生机械手。首先,采取模块化设计思想,设计适合纱筒抓取的仿生机械手结构模型,并选择绳索传动作为驱动方式;其次,详细分析仿生机械手的组成及其抓取原理,运用D-H坐标法,实现机械手指坐标系和手指基座坐标系之间变换,推导机械手末端位置方程,得到最优抓取姿态;最后,利用有限元软件,建立三维欠驱动仿生机械手模型并对其进行虚拟装配与运动仿真分析,以验证机械手抓取纱筒的可行性和稳定性,形成机器人智能抓取仿生机械手的关键技术。     

水下智能识别与自主抓取机器人设计与实现

设计了一款面向海珍品捕捞的水下智能识别与自主抓取机器人. 首先通过YOLOv4-tiny网络对海珍品图像 离线训练, 设计单双目自适应切换与多目标选择算法以实现海珍品在线识别与持续定位. 进一步, 采用声呐与深度 传感器融合策略获取水下机器人深度信息, 设计基于模糊比例–积分–微分控制的定深抓取控制器, 以确保目标定位 与抓取过程中深度信息的有效反馈. 所提目标识别算法, 具有实时性强、复杂度低优点; 同时, 定深与抓取控制器, 不依赖于系统复杂模型, 可适应不同海况下的精确抓取. 最后, 通过试验验证了方法的有效性.     

基于行星齿轮机构的牵引式欠驱动机械手设计

为实现夹持力调节和目标物体抓取功能,设计了一种基于行星齿轮机构的牵引式欠驱动机械手。该机械手采用不固定输出轴和内齿圈的行星齿轮机构来分配2种互斥的运动,一种用于手指的转动,实现机械手的夹持功能;另一种用于皮带轮的无限转动,通过皮带实现目标物体的拉入功能。相对于腱绳式和连杆式欠驱动机械手,这种单输入双输出形式不仅能够保持机械手的自适应性,同时还能降低机械手的耦合程度。根据行星齿轮机构的这种传动特点,设计了阻力矩调节机构,实现了机械手的夹持力调节功能。夹持力测试实验表明,在阻力矩调节机构的作用下,机械手能有效调节夹持力。抓取实验结果表明,机械手能够实现对刚性和柔性目标物体的抓取操作,验证了机械手设计的有效性。     

基于肌力信号与电刺激感觉反馈的多自由度机械手人机交互控制

为使操作者能够灵活控制多自由度机械手并能感受到机械手的抓取力,提出了一种具有双向信息传输能力的可穿戴式人机交互系统及控制方法.该系统利用压力传感器(FSR)阵列采集与操作者手部动作对应的前臂肌力信号,基于SVM(支持向量机)多类分类器算法实现对手部动作的识别,通过发送动作模式码控制机械手动作.另外,基于经皮神经电刺激(TENS)原理,将机械手抓取力信号转变为电刺激信号刺激体表皮肤,实现机械手抓握力向人体的感觉反馈.实验表明,基于肌力信号和SVM分类器的动作模式识别方法可实现对10种手部动作的识别,成功率不低于95%;电刺激感觉反馈可向人体准确反馈抓取力感并实现盲抓取.     

基于ARM9的机械手控制系统的设计

设计的机械手控制系统以ARM9微处理器作为控制核心,采用Linux操作系统作为软件平台.在Qtopia-2.2.0开发平台上开发人机交互界面,ARM9作为上位机,单片机作为下位机,结合相关的外围电路实现对机械手运动轨迹和抓取动作的控制.为了提高机械手和被搬运物体的安全性,系统采用了拉压力传感器和行程开关装置结合控制,另外通过光栅反馈控制可以进一步提高机械手的控制精度.相对于传统以工业控制机或工业计算机为核心的机械手控制系统,该系统具有功耗低、成本低、体积小等优点.     

面向3维物体的三指机械手"包笼抓取"方法

研究了面向3维多面体和类多面体的三指机械手包笼抓取策略.为了减小计算量,将高维包笼抓取构形的搜索问题转换成低维空间中包笼抓取集合的搜索问题,提出了利用机械手手指在构形空间中形成的约束域搜索包笼抓取构形的方法.这种方法搜索到的包笼抓取构形能够收敛到不完全形封闭抓取构形.最后,通过Barrett手的抓取实验验证了算法的有效性.     

水下机器人手爪力感知系统研究

提出了一种水下机器人手爪力感知系统的组成结构,该力感知系统由一个六维力/力矩传感器和三指夹持器指端的三个指力传感器组成,本文介绍了上述两种传感器的设计和标定,并对利用该手爪系统抓取物体进行了实际测试和分析,实验结果表明;所设计的力感知系统能够实时地感知腕力和夹持力信息,可以满足机械手力控制的需要.     

一种新型欠驱动机械手爪的抓取分析和优化设计

传统欠驱动机械手的运动和功能单一，难以实现对不同尺寸物体的稳定抓取．为此，提出了一种新型欠驱动手爪结构，并进行抓取分析和优化．首先，介绍了欠驱动机械手爪的整体机构设计，并对手指进行静力学分析，针对手爪包络抓取物体时可能发生弹射的不稳定情况，进行手指结构优化．然后，基于刚度矩阵的势能模型，确定指尖合理的尺寸范围并建立指尖最佳形状．通过几何约束中的数学公式，表达了指尖抓取时手指位姿和物体尺寸的关系．最后，完成手爪样机的搭建，并对常见家用物品进行了指尖抓取和包络抓取实验．实验结果表明，该机械手爪能够对各种尺寸大小的物体进行稳定抓取．     

A nonregressor nonlinear disturbance observer-based adaptive control scheme for an underwater manipulator

A new nonlinear disturbance observer-based tracking control scheme for an underwater manipulator is presented in this paper. This observer overcomes the disadvantages of existing disturbance observers, which are designed or analyzed by the linear system techniques. It can be applied in underwater manipulator systems for various purposes such as payload compensation, interaction effects compensation, underwater current or external disturbance compensation, and independent system control. The performance of the proposed tracking control scheme is demonstrated numerically by the payload compensation and interaction effects compensation for a two degrees of freedom vertical underwater manipulator.     

Kinematic Design and Analysis of a 7 Degree-of-Freedom Dual-Stage Inspection Manipulator for Dexterous Subsea Applications

This paper describes the design of a 7 degree-of-freedom (d.o.f) manipulator for underwater inspection applications. The functional requirements of an underwater manipulator for subsea inspection are discussed and the desired performance requirements identified. The inspection process of a weld joint using a manipulator is described and the desirable attributes of a 5 d.o.f manipulator for the inspection process established. A novel kinematic structure, for Underwater Robotic Vehicle (URV) operation, having a 2 d.o.f launching stages and a 5 d.o.f inspection stage is proposed for the manipulator. This configuration increases the dexterity, without compromising on the total reach of the manipulator. The kinematic structure of the 7 d.o.f, 2 stage, manipulator is presented. A hybrid power actuation is proposed for the manipulator to exploit the benefits of both hydraulic as well as electric actuators. Kinematic analysis of the manipulator is presented. The link dimensions of the inspection stage manipulator is done on the basis of kinematic performance indices of the manipulator. The novel kinematic structure and the hybrid power actuation strategy results in a power efficient, dexterous manipulator for underwater applications.     

Digital RAC with a disturbance observer for underwater vehicle-manipulator systems

Most control methods of underwater vehiclemanipulator systems (UVMS) are based on the computed torque method that is used for underwater robotic vehicles. We have proposed a resolved acceleration control (RAC) method for UVMS. In this article, we propose a disturbance compensation control method for both vehicle and manipulator based on the RAC method. Experimental results using an underwater robot with a vertical planar 2-link manipulator show that the proposed control method has good control performance.     

A coordinated control of an underwater vehicle and robotic manipulator

Remotely operated underwater robotic vehicles (URVs) have been used for various tasks: inspection, recovery, construction, etc. With the increased utilization of remotely operated vehicles in subsea applications, the development of autonomous vehicles becomes highly desirable to enhance operator efficiency. However, engineering problems associated with the high density, nonuniform and unstructured seawater environment, and the nonlinear response of the vehicle make a high degree of autonomy difficult to achieve. The vehicles are usually equipped with mechanical manipulators that are utilized during the working mode. The accurate performance of the vehicle during the working mode can be achieved by controlling the vehicle and manipulator at the same time and compensating the end-effector error due to the vehicle motion. This article describes an adaptive control strategy for the coordinated control of an underwater vehicle and its robotic manipulator. The effectiveness of the control system is investigated by case study. The results show that the presented control system can provide the high performance of the vehicle and manipulator in the presence of unpredictable changes in the dynamics of the vehicle and its environment.     

Experiments on a floating underwater robot with a two-link manipulator

This article concerns experiments with a free-floating underwater robot with a two-dimensional, horizontal planar, two-link manipulator. Some dynamic models of underwater manipulators have been proposed, but only a few experiments have been carried out. Here, we derive a dynamic model for a free-floating underwater robot with a two-link manipulator, including the hydrodynamic forces, and validate the effectiveness of the model by simulation and experiment. We also show an experimental result using a resolved acceleration control method. These experimental results show the effectiveness the model and the control method. This work was presented in part at the 5th International Symposium on Artificial Life and Robotics, Oita, Japan, January 26–28, 2000     

基于神经网络和模糊补偿的水下机械臂控制

针对水下机械臂动力学模型建模复杂且滑模控制的抖振问题,利用Lagrange法和Morison方程精准建立二连杆串联水下机械臂的动力学模型,对模型中参数的不确定项使用4个RBF神经网络分别进行逼近,并且对摩擦项使用模糊控制进行补偿的方法,精准迅速地实现了对水下机械臂控制系统跟踪控制。通过进行仿真分析,基于神经网络和模糊补偿控制的方法与滑模控制、整体RBF神经网络控制和分块RBF神经网络控制相比,控制系统的平均误差分别降低了85.5%、71.8%、93.1%。结果表明,此方法有效降低了控制系统的跟踪误差,并同时提高了稳态性和抗干扰性。     

Trajectory Tracking Control of an Underactuated Underwater Vehicle Redundant Manipulator System

The purpose of this study is to control the position of an underactuated underwater vehicle manipulator system (U‐UVMS). It is possible to control the end‐effector using a regular 6‐DOF manipulator despite the undesired displacements of the underactuated vehicle within a certain range. However, in this study an 8‐DOF redundant manipulator is used in order to increase the positioning accuracy of the end‐effector. The redundancy is resolved according to the criterion of minimal vehicle and joint motions. The underactuated underwater vehicle redundant manipulator system is modeled including the hydrodynamic forces for the manipulator in addition to those for the autonomous underwater vehicle (AUV). The shadowing effects of the bodies on each other are also taken into account when computing the hydrodynamic forces. The Newton‐Euler formulation is used to derive the system equations of motion including the thruster dynamics. In order to establish the end‐effector trajectory tracking control of the system, an inverse dynamics control law is formulated. The effectiveness of the control law even in the presence of parameter uncertainties and disturbing ocean currents is illustrated by simulations.     

Adaptive nonsingular fast terminal sliding mode control for underwater manipulator robotics with asymmetric saturation actuators

In this paper, an adaptive nonsingular fast terminal sliding mode control (ANFTSMC) is proposed for underwater manipulator robotics with asymmetric actuator saturations and unknown time-varying (TV) external disturbances. Firstly, the nonsingular fast terminal sliding mode (NFTSM) control scheme is conducted for the underwater manipulator robotics, which guarantees the boundedness of all the signals in the control system. Secondly, the adaptive method and the smooth hyperbolic tangent (tanh) function are introduced to address the unknown TV external disturbances and the input saturation errors. Thus the prior knowledge about the upper bounds of the system uncertainties is not needed in this paper. To deal with the nonlinear asymmetric input saturation issue, a Gaussian error function is employed in the asymmetric saturation module so that the discontinuous input signals can be transformed into smooth forms. Thirdly, the rigorous mathematical verification is conducted to demonstrate the stability and finite-time convergence of the closed-loop control system via the Lyapunov theory. Finally, numerical simulations are performed on a two-link underwater manipulator robotic system to illustrate the effectiveness of the proposed controller.     

Investigations on the dynamic coupling in AUV-manipulator system and the manipulator trajectory errors using bond graph method

This article presents the modelling and simulation of the dynamic coupling in an autonomous underwater vehicle (AUV)-manipulator system, used for subsea intervention tasks. Bond graph, a powerful tool in multi-domain dynamic system modelling, is used for the first time to model the coupled dynamics of the AUV-manipulator system. This method enables the development of the system model in a modular form by creating sub-system models and connecting these models together at energy interactions ports, thus overcoming many of the computational difficulties encountered in conventional modelling methods. The effects of gravity, buoyancy, added mass and fluid drag on the dynamics of a 3 degrees of freedom (DoF) manipulator mounted on a 6 DoF AUV are analysed. The manipulator trajectory errors due to the interaction forces and moments between the vehicle and the manipulator have also been investigated and the results are presented. The dynamic model predicts the reaction forces on the vehicle under various operating conditions of the manipulator and their influence on the manipulator trajectory. The percentage errors of manipulator tip trajectory for different initial configurations and operating conditions are analysed. The estimation of resulting errors in the manipulator path due to dynamic coupling effect on the manipulator trajectory helps in the design of suitable trajectory controller for the system. Cartesian space transpose Jacobian controller for trajectory control of manipulator has been implemented and results are presented.     

蠕动爬行攻泥机构工作特性的有限元分析

针对水下攻泥机器人蠕动爬行攻泥机构的结构与工作机理，建立了适当的有限元计算 模型，进而通过弹塑性有限元计算，分析了攻泥机构直行攻泥时与土相互作用的工作特性， 阐明了攻泥机构在前进过程中限位块导致的土体局部弱化区对其工作特性的影响程度．综合 分析计算与模型试验结果，建议了攻泥机构直行攻泥所必须的条件．