A Theoretical Approach of an Intelligent Robot Gripper to Grasp Polygon Shaped Objects

This paper presents an approach for grasp planning and grasp forces optimization of polygon shaped objects. The proposed approach is an intelligent rule-based method that figures out the minimal number of fingers and minimal values of contact forces. These fingers are required to securely grasp a rigid body in the presence of friction and under the action of some external force. This is accomplished by finding optimal contact points on the object boundary along with minimal number of fingers required for achieving the aforementioned goal. Our system handles every object case independently. It generates a rule base for each object based on adequate values of external forces. The system uses the genetic algorithm as its search mechanism, and a rule evaluation mechanism called bucket brigade for the reinforcement learning of the rules. The process mainly consists of two stages; learning then retrieval. Retrievals act on line utilizing previous knowledge and experience embedded in a rule base. If retrievals fail in some cases, learning is presumed until that case is resolved. The algorithm is very general and can be adapted for interface with any object shape. The resulting rule base varies in size according to the degree of difficulty and dimensionality of the grasping problem.     

Optimization of grasping forces in handling of brittle objects

This paper deals with the optimization of grasping brittle objects with a multi-fingered robot hand under general constraints such as finger deformability and object positioning tolerances. First, a general formulation describing hyperstatic grasping is presented. Then an optimization criterion based on the minimization of squeezing forces and torques is introduced. And finally results of numerical simulation for grasping with a special three-fingered gripper are presented.     

Automatic grasp planning for multifingered robot hands

In this paper the development of a planning environment is described which was especially tailored for grasping and manipulating with multifinger robot hands. The research has been concerned with the programming and simulation system of the Karlsruhe dextrous hand, which has been in development for two years. The work presents the result of a geometric-mechanic approach to the object-handling problem with dextrous multifinger hands by selecting grasp points and searching grasp forces to perform desired assembly tasks. The knowledge representation for the sequence planning and command execution is based on object and task restrictions combined with routines for successive optimization and a constraint propagation algorithm.     

Adaptive neuro-fuzzy prediction of grasping object weight for passively compliant gripper

The development of universal grippers able to pick up unfamiliar objects of widely varying shapes and surfaces is a very challenging task. Passively compliant underactuated mechanisms are one way to obtain the gripper which could accommodate to any irregular and sensitive grasping objects. The purpose of the underactuation is to use the power of one actuator to drive the open and close motion of the gripper. The fully compliant mechanism has multiple degrees of freedom and can be considered as an underactuated mechanism. This paper presents a new design of the adaptive underactuated compliant gripper with distributed compliance. The optimal topology of the gripper structure was obtained by iterative finite element method (FEM) optimization procedure. The main points of this paper are in explanation of a new sensing capability of the gripper for grasping and lifting up the gripping objects. Since the sensor stress depends on weight of the grasping object it is appropriate to establish a prediction model for estimation of the grasping object weight in relation to sensor stress. A soft computing based prediction model was developed. In this study an adaptive neuro-fuzzy inference system (ANFIS) was used as soft computing methodology to conduct prediction of the grasping objects weight. The training and checking data for the ANFIS network were obtained by FEM simulations.     

Force and position control of grasp in multiple robotic mechanisms

One of the approaches to increase the dexterity of a robot manipulating system is a design philosophy that consists of multiple robotic mechanisms. Applications of such a collection of manipulators can be in the design of a dextrous end-effector, a reconfigurable fixture to locate and grip various sized objects, or cooperative robotic arms which through their coordinated motions are able to accomplish a given task. Although the applications of such a design philosophy are endless, many problems still remain to be addressed. One of these problems is the control of the contact forces (grasping forces) between the mechanisms and the position of the grasped object. This article addresses this problem. First, a model of the mechanisms in contact with the grasped object is postulated; second, the problem of controlling the grasping forces and the position of the grasped object is formulated in the linear multi-input/multi-output system, and, finally, a centralized optimal controller is proposed for controlling the desired variables. The results of this article are demonstrated using two examples. One of the main advantages of the proposed controller is that it also shapes the transient response of the grasping force, which is an important consideration in cases when grasping fragile objects. © 1996 John Wiley & Sons, Inc.     

一类Power抓持接触力分解的一般表达式

与灵巧抓持相比,Power抓持可承受较大的外部载荷 ,能够更稳定地抓持物体．但由于抓持机构与物体间的约束较多,且接触点可出现在抓持机 构中活动度有限的构件上,因此不能采用已有的适用于灵巧抓持的接触力分解方法对其接触 力进行分解．本文在对一类Power抓持机构的结构特征进行分析的基础上,适当地建立了接 触坐标系和物体坐标系．通过对接触力空间进行分解,给出了抓持接触力分解的一般表达式 ．并根据该表达式,建立了此类Power抓持的外部载荷与关节力矩及接触力之间的显函数关 系．     

面向任务的三指手机器人抓取规划研究

本文给出了一种面向任务的三指手爪抓取规划的思路及研究方法，首先根据人手抓取姿态的分类，总结出典型的机器人抓取姿态，并以三指手爪来完成抓取，然后综合考虑任务要求、对象物体的几何物理特性及环境信息，经任务分析，推理出抓取姿态，并通过寻找特征平衡，确定出抓取平面，再在抓取现上进一步规划出３个抓取点，最终完成抓取规划过程。     

Grip posture and forces during holding cylindrical objects with circular grips

Individual finger position and external grip forces were investigated while subjects held cylindrical objects from above using circular precision grips. Healthy females (n = 11) and males (n = 15) lifted cylindrical objects of various weights (05, 10 and 20kg), and varied diameters (50, 7-5 and 100cm) using the 5-finger grip mode. The effects of 4-, 3- and 2-finger grip modes in the circular grip were also investigated.

Individual finger position was nearly constant for all weights and for diameters of 5-0 and 7-5 cm. The mean angular positions for the index, middle, ring and little fingers relative to the thumb were 98°, 145°, 181°, and 236°, respectively. At the 10-cm diameter, the index and middle finger positions increased, while the ring and little finger positions decreased. There were no differences in individual finger position with regard to gender, hand dimension, or hand strength.

Total grip force increased with weight, and at diameters greater or lesser than 7-5 cm. Total grip force also increased as the number of fingers used for grasping decreased. Although the contribution of the individual fingers to the total grip force changed with weight and diameter, the thumb contribution always exceeded 38% followed by the ring and little fingers, which contributed approximately 18-23% for all weights and diameters. The contribution of the index finger was always smallest (>11%). There was no gender difference for any of the grip force variables. The effects of hand dimension and hand strength on the individual finger grip forces were subtle.     

On the grasping stability and optimality under external perturbations

There are two types of grasping analysis in robotics research: find the grasping force distributions among the grasping fingers when given the contact points and find a good set of the contact points when given the shape of the object. Each kind of problem is associated with optimality and stability analysis. In this article, we investigate the grasping stability and optimality issues under the influence of external perturbations. A rotation‐displacement geometry model is used in computing the changes of grasping forces under external perturbations. Using these results, we present the concept of perturbation closure, which plays the central role in our analysis. A method for finding the local minimal perturbation resisting forces required for non‐slip contacts is developed based on this concept. A grasp so determined is guaranteed to be stable if the external perturbations do not exceed the threshold. Based on this property, we develop a quantitative measurement that can be used to evaluate the performance of different grasping configurations. One can use this measurement to determine the best grasping configuration from a set of perturbation resisting grasps. This actually gives a method which enables the optimal grasping configuration to be found. Both two‐dimensional and three‐dimensional cases are discussed in detail for determining the perturbation closure, the local minimal perturbation resisting force, and the perturbation resisting grasp. Examples are given at the end of the article to illustrate our idea. ©1999 John Wiley & Sons, Inc.     

Artificial neural network dexterous robotics hand optimal control methodology: grasping and manipulation forces optimization

Optimal fingertip forces can always be computed through the well-known optimization algorithms. However, computation time has always remained a real-time constraint. This article presents an efficient scheme to compute optimal grasping and manipulation forces for dexterous robotics hands. This is expressed as a quadratic optimization problem, and an artificial neural network (ANN) is used to learn such quadratic optimization formulations. Computation has been based on a nonlinear model of fingertip contacts and slips. In achieving object grasping while in motion, the hand Jacobian is considered an important matrix to be computed, but it is also highly intensive for real-time computed applications. Consequently, we investigated an efficient approach using artificial neural networks to learn optimal grasping forces. An ANN is used here to learn the optimal contact forces relating hand joint-space torques to the resulting object force. The results have indicated that the ANN has reduced computation times to reasonable values owing to its ability to map nonlinear force relations. Furthermore, the results have revealed that ANNs are capable of learning highly nonlinear relations relating to distributed fingertip forces and joint torques. The technique developed has also proved to be suitable for off-line learning of computed fingertip forces, even with large training samples.     

基于多尺度特征融合的抓取位姿估计

抓取目标多样性、位姿随机性严重制约了机器人抓取的任务适应性,为提高机器人抓取成功率,提出一种融合多尺度特征的机器人抓取位姿估计方法.该方法以RGD信息为输入,采用ResNet-50主干网络,融合FPN(feature pyramid networks)获得多尺度特征作为抓取生成网络的输入,以生成抓取候选框;并将抓取方向...     

Force feedback requirements for efficient laparoscopic grasp control

During laparoscopic grasping, tissue damage may occur due to use of excessive grasp forces and tissue slippage, whereas in barehanded grasping, humans control their grasp to prevent slippage and use of excessive force (safe grasp). This study investigates the differences in grasp control during barehanded and laparoscopic lifts. Ten novices performed lifts in order to compare pinch forces under four conditions: barehanded; using tweezers; a low-efficient grasper; and a high-efficient grasper. Results showed that participants increased their pinch force significantly later during a barehanded lift (at a pull-force level of 2.63 N) than when lifting laparoscopically (from pull-force levels of 0.77 to 1.08 N). In barehanded lifts all participants could accomplish a safe grasp, whereas in laparoscopic lifts excessive force (up to 7.9 N) and slippage (up to 38% of the trials) occurred frequently. For novices, it can be concluded that force feedback (additional to the hand-tool interface), as in skin-tissue contact, is a prerequisite to maintain a safe grasp. Much is known about grasp control during barehanded object manipulation, especially the control of pinch forces to changing loading, whereas little is known about force perception and grasp control during tool usage. This knowledge is a prerequisite for the ergonomic design of tools that are used to manipulate objects.     

基于深度神经网络的多指灵巧手抓取手势优化

基于深度神经网络模型,提出了一种适用于多指灵巧手的抓取手势优化方法。首先,在仿真环境下构建了一个抓取数据集,并在此基础上训练了一个卷积神经网络,依据目标物体单目视觉信息和多指灵巧手抓取位形来预测抓取质量函数,由此可以将多指灵巧手的抓取规划问题转化为使抓取质量最大化的优化问题,进一步,基于深度神经网络中的反向传播和梯度上升算法实现多指灵巧手抓取手势的迭代与优化。在仿真环境中,比较该网络和仿真平台对同一抓取位形的抓取质量评估结果,再利用所提出的优化方法对随机搜索到的初始手势进行优化,比较优化前后手势的力封闭指标。最后,在实际机器人平台上验证本文方法的优化效果,结果表明,本文方法对未知物体的抓取成功率在80%以上,对于失败的抓取,优化后成功的比例达到90%。     

Grasp stability analysis based on acceleration convex polytopes for multi-fingered robot hands

In this paper, we present the analysis of grasp stability for multi-fingered robot hands that is based on translational and rotational acceleration convex polytopes. The aim of the grasp stability analysis is to find the resistance forces and moments of robot hands that can withstand the external disturbance forces and moments applied on objects. We calculate the resistance forces and moments respectively which are considered the properties of objects and robots. Therefore, the resistance forces and moments depend on the joint driving torque limits, the posture and the mass of robot fingers, the configuration and the mass of objects, the grasp position, the friction coefficients between the object surface and the end-effectors of robot fingers. We produce the critical resistance force and moment which are absolutely stable about external disturbances in all directions, the global resistance force and moment which are whole grasp capability of robot hands, and the weighted resistance forces and moments which can be properly used by controlling two indices according to the importance of robot hands. The effectiveness of this method is verified with simulation examples. Recommended by Editorial Board member Hyoukryeol Choi under the direction of Editor Jae-Bok Song. Myeong Eon Jang received the B.S. and M.S. degrees in Mechanical Engineering from Chonnam National University, Gwangju, Korea in 1987 and 1990, the Ph.D. degree in the Department of Mechatronics Engineering at Chungnam National University, Daejeon, Korea in 2009, respectively. Since 1993, he has been a Researcher in the Agency for Defense Development (ADD), Daejeon, Korea. His research interests include robotics and intelligent control. Jihong Lee received the B.S. degree in Electronics Engineering from Seoul National University, Korea in 1983, and the M.S. and Ph.D. degrees from the Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 1985 and 1991, respectively, all in Electrical and Electronics Engineering. Since 2004, he has been a Professor in the Mechatronics Engineering Department of Chungnam National University, Daejeon, Korea. His research interests include robotics, intelligent control, multi-robot localization and path planning.     

The SPRING Hand: Development of a Self-Adaptive Prosthesis for Restoring Natural Grasping

Commercially available prosthetic hands are simple grippers with one or two degrees of freedom; these pinch type devices have two rigid fingers in opposition to a rigid thumb. This paper focuses on an innovative approach for the design of a myoelectric prosthetic hand. The new prosthesis features underactuated mechanisms in order to achieve a natural grasping behavior and a good distribution of pinching forces. In this paper it is shown that underactuation allows reproducing most of the grasping behaviors of the human hand, without augmenting the mechanical and control complexity.     

Computation for Maximum Stable Grasping in Dynamic Force Distribution

One of the fundamental problems in grasping and manipulation of an object by a multi-fingered robotic hand is the computation of contact forces to equilibrate the dynamic external wrench on the grasped object. This paper proposes a two-phase algorithm to calculate the contact forces in order to achieve maximum grasp stability, assuming there is no any slippage in the process of grasping. In the off-line phase, a nonsingular simplex set is obtained by the zone triangulation of the contact primitive wrench set in the wrench space, and the neighbors of the resultant simplexes are recorded by a neighbor-searching procedure. In the on-line phase, a specific simplex in which the required resultant wrench is located can find out rapidly since all neighbors of each simplex have been recorded before. The optimal contact forces can be obtained by the combination of the primitive forces corresponding to the vertices of this simplex. A numerical example shows the proposed algorithm takes a thousandth of the computation time exhausted by the sequential quadratic programming (SQP) or the straightforward bisection method with only a slight lost of optimality, and obtains better solution compared to the decomposition and positive combination (DPC) algorithm at the similar computation speed.     

Multi-fingered robot hand optimal task force distribution: Neural inverse kinematics approach

Grasping and manipulation force distribution optimization of multi-fingered robotic hands can be formulated as a problem for minimizing an objective function subject to form-closure constraints, kinematics, and balance constraints of external force. In this paper we present a novel neural network for dexterous hand-grasping inverse kinematics mapping used in force optimization. The proposed optimization is shown to be globally convergent to the optimal grasping force. The approach followed here is to let an artificial neural network (ANN) learn the nonlinear inverse kinematics functional relating the hand joint positions and displacements to object displacement. This is done by considering the inverse hand Jacobian, in addition to the interaction between hand fingers and the object. The proposed neural-network approach has the advantages that the complexity for implementation is reduced, and the solution accuracy is increased, by avoiding the linearization of quadratic friction constraints. Simulation results show that the proposed neural network can achieve optimal grasping force.     

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．     

Pick-and-verify: verification-based highly reliable picking system for various target objects in clutter

The reliability of picking task for various objects in clutter, as measured on the Amazon Picking Challenge, is far from the expectations of automation companies. Even if the best-performed team, who run object detection before picking the object, had picked a wrong object in the competition. In this paper, we propose a practical method to compose a highly reliable picking system with verification-based approach to reduce the rate of wrong picking and raise the reliability of picking ordered objects. In our approach, which we call pick-and-verify, the robot recognizes object twice: in clutter scene to detect the target and in hand after picking an object with less time loss and rise of reliability of picking the target. For grasping the detected object we do not assume its pose and it is actually the target object, instead, we adopt vision-based grasp planning for vacuum gripper with sensed 3-D point cloud. With the presented approach, the reliability of picking target objects raised 50%, and the score in the APC2015 competition has been improved to be close to the best-performed team by picking 9 out of 12 objects in 10 min with the same hardware in our previous system.     

杂乱场景下小物体抓取检测研究

目的 杂乱场景下的物体抓取姿态检测是智能机器人的一项基本技能。尽管六自由度抓取学习取得了进展,但先前的方法在采样和学习中忽略了物体尺寸差异,导致在小物体上抓取表现较差。方法 提出了一种物体掩码辅助采样方法,在所有物体上采样相同的点以平衡抓取分布,解决了采样点分布不均匀问题。此外,学习时采用多尺度学习策略,在物体部分点云上使用多尺度圆柱分组以提升局部几何表示能力,解决了由物体尺度差异导致的学习抓取操作参数困难问题。通过设计一个端到端的抓取网络,嵌入了提出的采样和学习方法,能够有效提升物体抓取检测性能。结果 在大型基准数据集GraspNet-1Billion上进行评估,本文方法取得对比方法中的最优性能,其中在小物体上的抓取指标平均提升了7%,大量的真实机器人实验也表明该方法具有抓取未知物体的良好泛化性能。结论 本文聚焦于小物体上的抓取,提出了一种掩码辅助采样方法嵌入到提出的端到端学习网络中,并引入了多尺度分组学习策略提高物体的局部几何表示,能够有效提升在小尺寸物体上的抓取质量,并在所有物体上的抓取评估结果都超过了对比方法。