An Anthropomorphic Robot Hand Developed Based on Underactuated Mechanism and Controlled by EMG Signals

When developing a humanoid myo-control hand,not only the mechanical structure should be considered to afford a high dexterity,but also the myoelectric (electromyography,EMG) control capability should be taken into account to fully accomplish the actuation tasks.This paper presents a novel humanoid robotic myocontrol hand (AR hand Ⅲ) which adopted an underac- tuated mechanism and a forearm myocontrol EMG method.The AR hand Ⅲ has five fingers and 15 joints,and actuated by three embedded motors.Underactuation can be found within each finger and between the rest three fingers (the middle finger,the ring finger and the little finger) when the hand is grasping objects.For the EMG control,two specific methods are proposed:the three-fingered hand gesture configuration of the AR hand Ⅲ and a pattern classification method of EMG signals based on a statistical learning algorithm-Support Vector Machine (SVM).Eighteen active hand gestures of a testee are recognized ef- fectively,which can be directly mapped into the motions of AR hand Ⅲ.An on-line EMG control scheme is established based on two different decision functions:one is for the discrimination between the idle and active modes,the other is for the recog- nition of the active modes.As a result,the AR hand Ⅲ can swiftly follow the gesture instructions of the testee with a time delay less than 100 ms.

HIT prosthetic hand based on tendon-driven mechanism

An underactuated finger structure actuated by tendon-driven system is presented.Kinematics and static analysis of the finger is done,and the results indicate that the prosthetic finger structure is effective and feasible.Based on the design of finger,a prosthetic hand is designed.The hand is composed of 5 independent fingers and it looks more like humanoid.Its size is about 85% of an adult's hand and weights about 350 g.Except the thumb finger,each finger is actuated by one DC motor,gear head and a tendon,and has three curling/extension joints.The thumb finger which is different from other existing prostheses is a novel design scheme.The thumb finger has four joints including three curling/extension joints and a joint which is used to realize the motion of the thumb related to the palm,and these joints are also driven by one DC motor,harmonic drive and a tendon.The underactuation and adaptive curling/extension motion of the finger are realized by joint torsion springs.A high-powered chip of digital signal processing(DSP)is the main part of the electrical system which is used for the motors control,data collection,communication with external controlling source,and so on.To improve the reliability of the hand,structures and sensors are designed and made as simply as possible.The hand has strong manipulation capabilities that have been verified by finger motion and grasping tests and it can satisfy the daily operational needs and psychological needs of deformities.

A Hand Prosthesis with an Under-Actuated and Self-Adaptive Finger Mechanism

One of the major problems faced by hand amputees is the unavailability of a lightweight and powered multi-functional hand prosthesis. Under-actuated finger designs play a key role to make the hand prosthesis lightweight. In this paper, a hand prosthesis with an under-actuated and self-adaptive finger mechanism is proposed. The proposed finger is capable to generate passively different flexion/extension angles for a proximal interphalangeal (PIP) joint and a distal interphalangeal (DIP) joint for each flexion angle of metacarpophalangeal (MCP) joint. In addition, DIP joint is capable to generate different angles for the same angle of PIP joint. Hand prosthesis is built on the proposed finger mechanism. The hand prosthesis enables user to grasp objects with various geometries by performing five grasping patterns. Thumb of the hand prosthesis includes opposition/apposition in addition to flexion/extension of MCP and interphalangeal (IP) joint. Kinematic analysis of the proposed finger has been carried out to verify the movable range of the joints. Simulations and experiments are carried out to verify the effectiveness of the proposed finger mechanism and the hand prosthesis.

Development of Five-Finger Multi-DoF Myoelectric Hands with a Power Allocation Mechanism

To be used as five-fingered myoelectric hands in daily living, robotic hands must be lightweight with the size of human hands. In addition, they must possess the DoFs （degrees of freedom） and high grip force similar to those of human hands. Balancing these requirements involves a trade-off; ideal robotic hands have yet to sufficiently satisfy both requirements. Herein, a power allocation mechanism is proposed to improve the grip force without increasing the size or weight of robotic hands by using redundant DoFs during pinching motions. Additionally, this mechanism is applied to an actual five-fingered myoelectric hand, which produces seven types of motions necessary for activities of daily living and realizes a -60% improvement in fingertip force, allowing three fingers to pinch objects exceeding 1 kg.

Influence of the Process of Worker on the Mechanical Properties of Carbon Fiber Reinforced Plastic Molded by Autoclave

The present study aims to investigate the difference of the operation process and the significant mechanical properties between expert workers and beginner workers clearly. The influence of the performance of the carbon fiber reinforced plastic (CFRP) molds made by subjects with difference skill level was discussed. Subjects were allowed to choose their process and molding tools. Subject A had 13 years,and subject B had 1 year of professional experience.The time spent usage of tools and process was recorded by a video camera used for analysis and comparison. Mechanical properties assessed in this study include tensile,compressive and Izod impact properties. The working time analysis shows that subject A needs short time for each process. On the other hand,subject B takes twice time to work for each process compared with subject A. From the surface and cross-section observation, it was found that the molding made by subject B had wrinkles all over and the disorderly inter-layer direction; on the contrary,the wrinkles on the molding surface of subject A are barely,and the inter-layer direction of which are more orderly. The results of tensile and Izod impact tests show that the surface winkles have little influence on the mechanical properties of the molding. According to the compression test and failure structure analysis in edge corner part,it was found that it had a great influence on the compression mechanical properties of the sample belonging to different subjects with different experience.

HIT／DLR HandⅡ类人形五指灵巧手机构的研究

为了使灵巧手的外观和结构功能更加类人化,研制了新一代类人形机器人五指灵巧手——HIT/DLR HandⅡ.该灵巧手由一个独立的手掌和5个完全相同的具有4个关节3个自由度的模块化手指构成.机电集成的设计思想保证了所有驱动器,传动系统和电路系统都集成在手指和手掌的内部.钢丝耦合机构的采用,保证了手指末端关节在整个运动范围内都能够保证精确定1:1传动比.包装件同结构功能件融合的设计思想不但进一步缩减的灵巧手的外形尺寸,而且使灵巧手外形更加类人化.所有的创新型设计使灵巧手的手指外形尺寸仅是上一代手指的1/3.此外,为了提高灵巧手的多感知能力,重新设计的力/力拒传感器,位置传感器及温度传感器通巧妙的集成在手指的内部.试验表明,所设计的灵巧手不但达到了设计目标,也表现出了很强的抓握能力.

The anthropomorphic design and experiments of HIT/DLR five-fingered dexterous hand

This paper presents a new developed anthropomorphic robot dexterous hand： HIT/DLR Hand II. The hand is composed of an independent palm and five identical modular fingers, and each finger has three degree of freedom （DOFs） and four joints. All the actuators and electronics are integrated in the finger body and the palm. Owing to using a new actuator, drivers and a novel arrangement, both the length and width of the finger is about two third of its formner version. By using the wire coupling mecha- nism, the distal phalanx transmission ratio is kept exactly 1 ： 1 in the whole movement range. The packing mechanism which is implemented directly in the finger body and palm not only reduces the size of whole hand but also make it more anthropomorphic. Additionally, the new designed force/torque and position sensors are integrated in the hand for increasing muhisensory capability. To evaluate the performances of the finger mechanism, the position and impedance control experiments are conducted.

Development of Multi-Fingered Prosthetic Hand Using Shape Memory Alloy Type Artificial Muscle

A new prosthetic hand with a fixed thumb and four fingers actuated by Shape Memory Alloy （SMA） type artificial muscle has been developed in this paper. Different from typical geared motor, SMA actuator is lightweight and silent, however shows a little short stroke and small attracting force per each unit. In order to achieve enough output force and motion range of each finger, multiple SMA type artificial muscles with special device which facilitates enough length are equipped in the hand. The fundamental properties of the SMA type artificial muscle including output force and electrical response were determined experimentally and considered for the design of hand mechanism. Besides, the structure of each finger and whole system has been designed based on observation of human hand. The electrical hardware to control multiple shape memory alloy type artificial muscles has been also developed. Finally, the usefulness of the prosthetic hand has been investigated through experiments for grasping several types of objects.

人体隐喻词族的多义路向推演——“手(hand)”抉隐索微

语言多义系统严格遵循隐喻路径及其逻辑进行生成、演化和拓展。人体词族的衍生与泛化是该系统的典型代表与展示舞台。从基本意象到测量意象,“手(hand)”词群扩张的隐喻路径有七种。该词族在汉英两种语言中共享大部分的认知投射,整体上呈现“大同小异”的趋势。两种语言中的“手(hand)”隐喻因两个民族具有的相同生理基础和具身感知呈现诸多相似之处,又因两者不同的文化认知观而具有一定差异性。汉英语言对“手(hand)”的认知路径存在分野,其原因是两民族文化的趋异性。人类的认知模式大致相同,但个体之间因为地理环境和生活条件的影响,必然存在异质性。

Computer Assisted Designing System for Hands and Formability of Worsted Fabrics

Equations that can predict worsted fabrics’ properties such as bending, shearing, compression, surface and tension, were achieved by means of theoretical and experimental studies. By combining these equations with Kawabata’s hand and silhouette evaluation methods, a software system was established. Then the mechanical properties, hand and silhouette of a fabric can be predicted quickly and accurately in terms of fiber configurations, yarn and fabric structures. The predictive result if unsatisfied can be revised by the function of “Help for designing modification”.

Evaluation of Fabric Hand with Grey Element Model

A premium composite grey element model is established and used for objective evaluation of fabric hand. Fabric hand is regarded as a grey system and the model is composed of fabric mechanical properties, which are primary hand attributes. Based on comparison with a standard model, fabric hand can be objectively evaluated.

基于标准模块的Handling技术

本文介绍了FESTO公司在Handling技术上的发展状况 ,模块化拼装技术以及专用选型软件PAA。

机器人灵巧手研究综述

机器人灵巧手可以模仿人类动作,完成复杂场景下的多种任务。机器人灵巧手的研究是开发智能人形机器人的关键技术之一。以机器人灵巧手的研究历程为核心,阐述了多指灵巧手的研究现状,分析了灵巧手的演化过程和研究进展。从自由度数目、驱动方式、机械传动方式和传感技术等方面探讨了灵巧手分类和特征,展望了灵巧手未来的发展趋势。

基于深度学习车顶焊缝涂胶机械臂研究

为提高汽车工艺涂胶质量及机械臂作业效率,针对基于深度学习的双目视觉车顶焊缝涂胶机械臂系统,提出了一种SEmYOLOv5算法,在主干网络上增加SE(squeeze and excitation)注意力机制,同时在颈部网络上增加一组采样模块,提高焊缝的识别能力。对提取到的图像进行图像处理,使得更好的提取车顶焊缝的特征信息从而得到特征点坐标,采用B样条曲线法对机械臂进行轨迹规划。改进后的算法相较原YOLOv5算法的mAP值提升了6.76%,针对该系统进行实验并验证了提出的基于深度学习的双目视觉车顶焊缝涂胶机械臂系统的有效性。

果园机械化疏花技术与装备研究进展

为促进果园机械化疏花技术与装备发展,提高果园全程机械化水平,通过论述疏花的必要性与作业标准,以期为疏花工作提供理论指导。根据国内外手持式、振动式、机载式和智能式疏花机的工作方法与特点,概括分析其关键技术和优缺点。重点阐述机器视觉作为智能疏花机核心技术在花朵识别应用上的研究现状,通过对比YOLO、Faster R-CNN等典型花朵识别算法的平均精度、召回率和F1分数总结其制约因素和存在的主要问题。针对目前主流机载疏花机存在的工作模式单一、精准作业水平低、对果园标准化水平要求高、适用范围窄等主要问题,从规范果园种植方式、研发新型主轴结构与疏花绳材料、构建果园生产管理经验专家库、花朵识别技术的重点研究方向、智能疏花机未来研发重点5个方面进行展望。

三指机器人手爪设计与抓取研究

传统的多指仿人灵巧手虽然已实现较高的灵巧性与仿人性,但存在设计制作繁琐或价格昂贵以及手指末端抓力偏弱的弊端。而三指机械手作为一种简化形式一般又存在灵巧性不高的问题。针对以上弊端设计了一款基于滑块-摇杆机构的三指机器人手爪。将该结构直接类比人手所设计的手指关节能够实现中节指骨与远节指骨的耦合运动,且指尖力会随手指弯曲程度的增加而增大,有利于对细小目标物进行指尖捏取;采取与传统偏航-俯仰型不同的翻滚-俯仰型掌指结构,在增加了沿掌根关节的旋转自由度的同时尽可能地丰富了手爪的抓取姿态,使其能够适应更多的目标物,增加了灵巧性。在MATLAB中建立手指运动学模型并求得手指末端工作空间。经由实验验证得出该手爪具备较大的抓取力与丰富的抓取姿态。

基于并行混合注意力的复杂背景小尺度手部检测方法

针对复杂背景中手部特征不明显及尺度变化较大,难以满足高精度水平检测,易出现误检、漏检的问题,以YOLOv5为基础结构提出一种小尺度手部检测方法。将并行混合机制的注意力模块(parallel mixed attention mechanism,PMAM)嵌入到主干网络中,提高对手部特征的提取能力;设计一种结合路径聚合网络(path aggregation network,PAN)和加权双向特征金字塔网络(bidirectional feature pyramid network,BiFPN)改进的特征融合网络PB-FPN(path bidirectional-feature pyramid network),引入新的路径参与底部特征融合,提高算法对小尺度手部目标的检测能力;通过将骨干网络中的空间金字塔池化(spatial pyramid pooling-fast,SPPF)引入特征融合网络并与模型预测头连接,进一步提高算法的性能。在此基础上,使用FReLU作为网络模型的激活函数,增强网络的空间敏感度,提高网络鲁棒性。为验证所提方法的有效性,构建了更符合研究背景的新的数据集TV-COCO-Hand,并在此数据集上进行了相关实验,结果表明,改进后的模型在构建的数据集上mAP达到91.4%,比基线网络模型提高了3.8个百分点,且检测效果优于目前主流检测网络模型。在公开数据集上进行了数据集对比实验以及真实场景的检测实验,验证了模型的泛化性。

耳温计检定系统机械臂位姿自动修正方法研究

针对耳温计自动检定系统中机械定位随机偏移的问题,提出一种基于机器视觉的机械臂位姿自动修正算法。在恒温槽基准位建立机械臂运动的基准参考坐标系,在此坐标系下设计机械臂作业路径动作;系统作业过程中,通过定位恒温槽上的特征图案生成变换矩阵,产生新参考坐标系;机械臂来到新参考坐标系下拍照,反推变换矩阵偏差值并对其迭代更新;通过对机械臂运动的参考坐标系进行修正,完成对机械臂三维作业的位姿修正。实验结果表明,所提算法对恒温槽上3个平面的定位精度达到0.5 mm以内,为相似场景下的视觉引导提供了一种新思路。

基于折纸的被动触发双稳态机器人手设计

传统机器人手通常需要驱动器连续地提供保持抓握状态的扭矩或力。若驱动器失效,则机器人手无法稳定地抓取物体。针对上述问题,提出了一种基于折纸的被动触发双稳态机器人手。该机器人手由具有单自由度的抓握机构和具有双稳态特性的驱动机构组成。基于抓握机构和驱动机构的运动学模型,结合抓握状态的要求对机器人手的结构进行了设计,可通过设置扭转弹簧的刚度参数来实现对能量壁垒的灵活调整。最后,通过开展跌落捕获实验来验证所设计机器人手的抓握性能。结果表明,当跌落高度为400 mm时,机器人手的抓握运动未被触发;当跌落高度为440 mm时,机器人手成功抓握物体;当跌落高度为480 mm时,机器人手虽被触发了抓握运动但未能抓住物体。实验结果不仅验证了机器人手在无驱动条件下稳定抓取特定大小物体的能力,还揭示了不同外部冲击下能量壁垒的存在。基于折纸的被动触发双稳态机器人手在被动式和自适应机器人中具有潜在应用前景。

选矸机器人手爪机构设计及其分拣功能分析

选矸机器人手爪作为分拣矸石的执行机构,对矸石的分拣方式、分拣精度与分拣效率具有重要影响。传统手爪只能实现抓取或拨离的单一分拣方式,对于某些特定的场景,无法满足应用需求。为此,提出了一种多功能选矸机器人手爪,完成了机构设计和优化,分析了手爪机构的3种分拣模式,并制定了手爪分拣功能变换控制策略。该机器人手爪整体结构紧凑,可实现抓取、拨离与协同抱取3种功能形态之间的快速变换,拓展了选矸机器人的适用范围,可有效提高矸石的分拣精度与分拣效率。