面向任务和基于几何特征的机器人多指手抓取模式规划

针对机器人多指手自身的特点,通过分析人手的抓取特性,对其可能具有的抓取模式进行分类。考虑被抓取物体的几何特征和任务要求,采用基函数为高斯核函数的RBF神经网络来表示被抓物体的样本特征和抓取模式之间的复杂非线性映射。将抓取模式分为10类,对于新的被抓物体,利用训练好的神经网络自动生成抓取模式,并利用VC++/OpenGL建立了可视化仿真平台,进行了抓取模式分类仿真实验,结果表明对于新的物体,机器人可以选择适当的抓取模式进行抓取。     

面向任务的多指机械手抓取构形评价与优选

基于机械手抓取物体时手指与物体之间构成有摩擦的固定点接触，提出一种面向任务的机械手抓取构形评价方法，在保证一定抓取稳定性和满足任务要求的前提下，以整个操作过程中手指接触力的最大值最小为目标确定最佳的抓取构形。结合实例说明了抓取构形评价与优选方法的应用。     

基于抓取稳定性的手指接触力规划算法

在手指指端与物体构成有摩擦固定点接触模型的前提下,通过对三指灵巧手抓取内力与抓取稳定性关系的分析,构建指端接触力的规划算法。在接触点摩擦锥内将指端接触力分解为操作力和抓取内力,分析抓取内力对于接触稳定性的作用和内力汇交点与汇交多边形的相互位置,建立抓取内力与抓取稳定性之间的关系,通过合理规划内力汇交点的位置获得较高的抓取稳定性。通过使内力汇交点到汇交多边形各条边的最短距离达到最大,建立基于抓取稳定性的内力汇交点优化模型和指端接触力的规划算法。该算法将抓取布局规划和接触力分配融为一个统一的过程,可以有效提高抓取的规划效率,有利于抓取的在线规划。结合抓取实例说明了算法的有效性。     

三指手抓取力封闭性的简单判别算法

根据力封闭抓取的定义建立了利用严格内力的存在条件进行三指灵巧手力封闭抓取判别的研究方法,构建了三指灵巧手力封闭抓取的等价判别算法。在手指指端与物体构成有摩擦的固定点接触模型的前提下,三指灵巧手力封闭抓取条件等价于抓取矩阵满秩并存在严格内力。首先,基于对物体表面接触位置的几何特性分析,建立了摩擦扇存在条件,每个指端接触点的内法矢与接触平面所成夹角小于接触点摩擦圆锥角;其次,通过研究接触点摩擦扇边界线及其交点的相互位置关系,推导了内力汇交多边形的存在条件,任意两个摩擦扇的边界线交点中,至少有一个位于第3个摩擦扇的严格内部,或者至少有3个边界线交点位于第3个摩擦扇边界线上;最后,在考虑内力平衡关系的基础上,建立了基于矢量计算的三指灵巧手力封闭抓取的快速判别算法。     

力抓取的接触力建模与可行抓取力分析

为了解决操作机和夹持系统性能设计中可行抓取力和承载条件计算问题,研究过约束、超静定条件下的接触力建模和抓取分析方法。基于抓取的接触动力学约束,采用拉格朗日乘子方法建立摩擦抓取的约束力求解模型。讨论超静定抓取中接触约束的不确定性。通过识别可行速度集合与约束力集合的旋量基及其张成的子空间,给出表达摩擦接触约束力可行解的主动内力子空间结构,即物体所受内力子空间与可调整驱动力的交集,通过对接触处约束力旋量系统的双重正置运算,识别出内力旋量系及其张成的子空间,使得能够显式计算动态平衡条件下的可行接触力集合,建立动态载荷作用下可行抓取力与承载力的求解方法。     

机器人多指抓取力封闭分析的非线性规划算法

提出了一个同时适于机器人多指抓取力封闭定性与定量分析的非线性规划算法。从抓取的力平衡方程出发,研究了外力旋量空间与任一接触处操作力空间之间的关系。在接触力空间定义了约束力集、严格约束力集及法向力集,分析了三者之间的关系。基于摩擦约束的凸性,给出了力封闭和部分力封闭的充要条件,推导了相应的判别算法。提出了衡量抓取接触稳定性的定量指标。最后做了实例说明。     

热轧润滑剂用基础油的点接触和线接触高温减摩性能研究

用多功能SRV试验机评价了3种热轧润滑剂用基础油在点接触和线接触形式下的高温减摩性能。试验结果表明：在试验负荷范围内，当试验温度不大于300℃时，3种基础油的点接触和线接触摩擦因数都非常小，其摩擦因数在0．10—0．14之间，而且摩擦因数在试验过程中变化非常平稳；除了合成油和矿物油在试验负荷为20N的情况外，3种基础油在400℃时的点接触摩擦因数和线接触摩擦因数之间存在非常大的差异，点接触的摩擦因数非常高，而线接触摩擦因数则非常低；试验温度为500℃时，3种基础油的点接触摩擦因数和线接触摩擦因数都比较高，但线接触摩擦因数要比点接触摩擦因数低。     

多模式绳牵引刚柔结合夹持器设计与分析

为了解决夹持器难以稳定夹持轮廓复杂的易损异形物体的问题，以提高夹持器的自适应性和保护抓取能力，设计了一种包括绳索驱动系统和柔性铰链结构的绳牵引刚柔结合夹持器。通过设计夹持器的绳索路径实现多种抓取模式之间的相互转换。在载荷平衡方程的基础上对两种抓取模式进行静力学建模，得到输入力与输出夹持力之间的力学模型，并通过数值仿真和物理试验验证力学模型的正确性与夹持器的抓取性能。最后，分别对不同几何形状和不同刚度的抓取目标进行抓取试验，验证了夹持器自适应保护抓取的能力。     

多关节欠驱动机器人手爪包络抓取稳定性分析与仿真

研究了欠驱动手爪包络抓取稳定性.定义了抓取构形,推导了关节数和抓取构形之间的关系,采用抓取构形之间的转移难度作为表征抓取稳定性测度指标,给出了抓取不稳的主要原因以及影响包络抓取稳定性的主要因素.通过仿真对抓取不同形状物体时的稳定性,及影响稳定性的主要因素进行了对比分析和验证.结果表明:理论分析和仿真研究结果是一致的,欠驱动手爪包络抓取时存在多种抓取构形是影响抓取稳定性的直接原因,稳定抓取取决于物体形状和初始的抓取姿态,其中物体凸边的线接触的长度是决定抓取稳定的最重要的因素,物体越是接近圆形稳定性越差,当物体和抓取构形比较吻合时,并且都是以线接触相互作用,抓取稳定性最好.     

三指灵巧手内力汇交多边形存在性判别算法

抓取内力对三指灵巧手实现稳定抓取具有重要的作用.当手指指端与物体构成有摩擦的固定点接触模型时,三指内力必须位于相应接触点摩擦扇内并交于内力汇交多边形内的一点.因此,内力优化算法的研究必须建立在内力汇交多边形存在的基础之上.基于对物体表面接触位置的几何特性分析,建立摩擦扇存在条件,每个指端接触点的内法矢与接触平面所成夹角小于接触点摩擦圆锥角.通过研究接触点摩擦扇边界线及其交点的相互位置关系,推导内力汇交多边形的存在条件,任意两个摩擦扇的边界线交点中,至少有一个位于第三个摩擦扇的严格内部,或者至少有3个边界线交点位于第三个摩擦扇边界线上.最后构建判别三指内力汇交多边形存在性的通用算法,由此可确定能实现对指端操作力起调整作用的自平衡内力.结合抓取实例说明了判别算法的有效性.     

Posture optimization for a humanoid robot using a simple genetic algorithm

This study proposes a method of real-time posture optimization of humanoid robots using a genetic algorithm and neural network. Here, the motion of a humanoid robot pushing an object is considered. When the robot starts pushing the object, the palms of its hands and the soles of its feet are assumed to be fixed on the object and on the ground, respectively, and they sense the reaction force from those surfaces. The reaction force results in changes of torques in the joints. This study determines an optimized posture using a genetic algorithm such that either the torques are evenly distributed over all joints or the torque of the weakest joint is rapidly reduced. Several different optimized postures are then generated by varying the reaction forces at the palms and the soles. The data is used as training patterns for a multilayer perceptron neural network with a back-propagation learning algorithm. Using the trained neural network, the humanoid robot can find the optimal posture for different reaction forces in real time. Several simulations were conducted to confirm the effectiveness of the proposed method. The simulation results showed that the proposed method can be used for real-time posture optimization of humanoid robots.     

Control of whole finger manipulation utilizing frictionless sliding contact—theory and experiment

We propose a whole finger manipulation (WFM) to improve the robustness of manipulation by a multifingered robot hand. The WFM is characterized by enveloping an object by finger links of the multifingered robot hand, and utilizing the frictionless sliding contacts between the surfaces of finger links and the grasped object. Therefore, the WFM can achieve a large manipulation without regrasping the object and can be highly robust in nature because of form-closure due to multiple contact points on the grasped object. This paper discusses the property and control of the WFM. The WFM with frictionless contact is shown to be a manipulable and force closure grasp using kinematics and statics with frictionless sliding contact. We discuss that the force closure for the WFM with frictionless contact can be viewed as form closure. An internal force-based impedance control is proposed for the stable manipulation in consideration of sliding contacts between the object and the finger links. A prototype hand and sensing systems are presented. The validity of the proposed control law is verified through the experiments.     

On the influence of surface roughness on real area of contact in normal,dry, friction free,rough contact by using a neural network

A model previously developed at LTU was used in order to perform numerical simulations of normal, dry, friction free, linear elastic contact of rough surfaces. A variational approach was followed and the FFT-technique was used to speed up the numerical solution process. Five different steel surfaces were measured using a Wyko optical profilometer and several 2D profiles were taken. The real area of contact and the pressure distribution over the contact length were calculated for all the 2D profiles. A new slope parameter was defined. An artificial neural network was applied to determine the relationship between the roughness parameters and the real area of contact. The trained model was able to capture the dependence of the real area of contact on the roughness parameters. The ability of the neural network to generalize on unseen data was tested. The neural network was able to prove the correlation between the roughness parameters and the real area of contact.     

基于多尺度特征融合和抓取质量评估的抓取生成方法

在非结构化场景中，物体的6-Ddf抓取是智能服务机器人领域的一项极具挑战性的任务。在该场景中，机器人需要应对不同大小和形状的物体以及环境噪声等因素的干扰，因此难以生成准确的抓取姿态。针对此问题本文提出一种基于多尺度特征融合和抓取质量评估的6-Dof抓取姿态生成方法。首先，提出了自适应半径查询法，解决真实环境中点云采样不均匀导致的关键点查询异常的问题；其次设计了一种将多尺度特征和抓取质量评估融合的抓取生成网络，可以生成丰富的6-Dof抓取域：最后，定义了一种抓取质量评估方法，包含力闭合分数、接触面平整度、棱边分析和质心分数，并将这些标准应用在标准数据集上生成新的抓取置信分数标签，同时将这些标准融人抓取生成网络中。实验结果表明所述的方法与当前较为先进的方法FGC-GraspNet相比平均精度提升了5.9%，单物体抓取成功率提升了5.8%，多物体场景的抓取成功率提升了1.1%。综上所述，本文所提出的方法具备有效性和可行性，在单物体场景和多物体场景中具有较好的适应性。     

METHOD OF CLASSIFYING GRASPS BY ROBOT HANDS

ＭＥＴＨＯＤＯＦＣＬＡＳＳＩＦＹＩＮＧＧＲＡＳＰＳＢＹＲＯＢＯＴＨＡＮＤＳＺｈａｎｇＹｕｒｕ（ＢｅｉｊｉｎｇＵｎｉｖｅｒｓｉｔｙｏｆＡｅｒｏｎａｕｔｉｃｓａｎｄＡｓｔｒｏｎａｕｔｉｃｓＷｉｌｌｉａｍＡ．ＧｒｕｖｅｒＳｉｍｏｎＦｒａｓｅｒＵｎｉｖｅｒｓ...     

机器人手爪的柔顺指端

模仿人的皮肤与肌肉组织,本文提出了一种具有电流变流体的机器人手指柔顺指端的结构,在机器人的双指手爪中,指端是类似于皮肤的弹性薄膜和在皮肤下填充电流变流体的聚氨酯泡沫层组成,在未施加电场时,指端极为柔软,当电流变流体受电场作用时,指端成类固体状,试验结果表明,柔顺指端对物体的抓握过程是从柔性抓所至到物体被锁固在两指间的硬性抓握,因而在同样抓握力下,提高了机器人手爪举升物体的能力。     

机器人多指操作的抓取力分析和实时优化

在非线性摩擦锥约束转换为对称矩阵正定线性约束的基础上,首先将抓取力摩擦锥约束进一步转换为线性矩阵不等式约束,使抓取力封闭判别问题转化为带线性矩阵不等式约束的可行解问题,完成抓取的力封闭条件判别,然后根据摩擦锥的正定线性约束矩阵和抓取力平衡约束矩阵的分块特性和稀疏特性,实现多指滚动操作实时抓取力优化计算.最后在多指手实验平台上实现并验证了算法.     

Amontons' law at the molecular level

One of the fundamental postulates of friction is that at the microscopic or molecular level, the “real” area of contact is proportional to the load applied over the macroscopic or “apparent” area. This has both theoretical and experimental support and has formed the basis of many theoretical analyses, including an explanation of one of the most basic observations of everyday friction, i.e., that the friction force F is proportional to the load L or weight of the moving object (Amontons' law) where the ratio of F to L defines the coefficient of friction μ=F/L. We have carried out friction experiments between two molecularly smooth non-adhering surfaces under conditions where all the relevant macroscopic and microscopic parameters were directly measured. We find that even at the microscopic level the friction force is proportional to the net applied load and not to the real area of contact. One implication of this finding is that Amontons' law is also obeyed directly at the molecular level and does not emerge indirectly because of some fortuitous correlation between the net applied load and the local contact area or shear strength, as is commonly supposed. A physical model, based on intermolecular forces and thermodynamic considerations, is offered to explain why the friction force is proportional to the net applied load, and why the case of adhering surfaces - where the friction force is found to be proportional to the molecular contact area -is quite different from that of non-adhering surfaces. This revised version was published online in June 2006 with corrections to the Cover Date.