Three-fingered Robot Hand Research Articles

Underactuated Three-Finger Robot Hand with Human-Like Flexion

Numerous studies have been conducted on three-finger robot hands, which are widely used in industries. These studies led to the development of motorized prosthetic hands for amputees. Although many developers have focused on the functionality of motorized prosthetic hands, prosthetic users place more importance on the human-like motion of the device owing to social implications. Therefore, this study aims to achieve human-like flexion of a three-finger robot hand. Each finger contains three phalanges joined by pivots and torsional return springs. A tendon wire runs through guide pins inside the phalanges up to the linear actuator located in the forearm. When the tendon wire is pulled, the finger is flexed. The return springs are optimally selected based on ADAMS simulations and kinesiology of the human hand. The selection is then verified by experiments

Vision-Based Intelligent Perceiving and Planning System of a 7-DoF Collaborative Robot.

In this paper, an intelligent perceiving and planning system based on deep learning is proposed for a collaborative robot consisting of a 7-DoF (7-degree-of-freedom) manipulator, a three-finger robot hand, and a vision system, known as IPPS (intelligent perceiving and planning system). The lack of intelligence has been limiting the application of collaborative robots for a long time. A system to realize “eye-brain-hand” process is crucial for the true intelligence of robots. In this research, a more stable and accurate perceiving process was proposed. A well-designed camera system as the vision system and a new hand tracking method were proposed for operation perceiving and recording set establishment to improve the applicability. A visual process was designed to improve the accuracy of environment perceiving. Besides, a faster and more precise planning process was proposed. Deep learning based on a new CNN (convolution neural network) was designed to realize intelligent grasping planning for robot hand. A new trajectory planning method of the manipulator was proposed to improve efficiency. The performance of the IPPS was tested with simulations and experiments in a real environment. The results show that IPPS could effectively realize intelligent perceiving and planning for the robot, which could realize higher intelligence and great applicability for collaborative robots.

Kinematic design of a serial-parallel hybrid finger mechanism actuated by twisted-and-coiled polymer

Flexible actuators composed of twisted and coiled polymers (TCPs) can provide excellent performance such as high power-to-weight ratios for robotic hands. This paper presents a systematic methodology for the kinematic design of the TCP-actuated finger mechanism. The finger mechanism has a redundant serial-parallel hybrid topology, which is designed for grasping in space. The local and global performance evaluation indices concerning the dexterity and the extreme value of the velocity are first proposed for the optimal design of mechanism dimensions. Moreover, the flexibility matrix, mapping the stiffnesses of flexible elements including active TCP actuators and passive springs into the end-effector's stiffness, is established, and thus the extreme value of the deformation is obtained as the optimization objective function. Therefore, optimal passive stiffness coefficients are achieved by the genetic algorithm. Finally, the prototype of three-fingered robotic hand with the proposed finger mechanism and the control strategy based on TCP power-strain model were developed. Comparisons of experimental and theoretical data validated the kinematic design of the finger mechanism.

Effects of Voltage Dips on Robotic Grasping

This paper addresses the effects of electric power quality on robotic operations. A general overview is reported to highlight the main characteristics of electric power quality and it’s effects on a powered system by considering an end-user’s viewpoint. Then, the authors outline the influence of voltage dip effects by focusing on robotic grasping applications. A specific case study is reported, namely that of LARM Hand IV, a three-fingered robotic hand which has been designed and built at LARM in Cassino, Italy. A dedicated test rig has been developed and set up to generate predefined voltage dips. Experimental tests are carried out to evaluate the effects of different types of voltage dip on the grasping of objects.

A Sense of Touch for the Shadow Modular Grasper

In this letter, we have designed and built a set of tactile fingertips for integration with a commercial three-fingered robot hand, the Shadow Modular Grasper. The fingertips are an evolution of an established optical biomimetic tactile sensor, the TacTip. In developing the tactile fingertips, we have progressed the technology in areas such as miniaturization, development of custom-shaped finger-pads, and integration of multiple sensors. From these fingertips, we extract a set of high-level features with intuitive relationships to tactile quantities such as contact location and pressure. We present a simple linear-regression method for predicting roll and pitch of the finger-pad relative to a surface normal and show that the method generalizes to unknown depths and shapes. Finally, we apply this prediction to a grasp-control method with the Modular Grasper and show that it can adjust the grasp on three real-world objects from the YCB object set in order to attain a greater area of contact at each fingertip.

Three-Fingered Robot Hand with Gripping Force Generating Mechanism Using Small Gas Springs - Mechanical Design and Basic Experiments –

This paper presents the mechanical design of a new three-fingered robot hand for a robot designed to handle tableware. The finger mechanism has three joints and consists of a pair of fourbar linkage mechanisms, one small gas spring, and one feed screw mechanism. As the feed screw moves, the finger mechanism performs flexion and extension operations with its joints interlocked. The gas spring generates gripping force, which is adjusted at the position of the moving part moved by the feed screw. Therefore, the three-fingered robot hand can open and close synchronously, powered by a single motor in the base of the hand. The hand grips with mechanical flexibility. In addition, it can maintain its grip with no power supply. Tests show that the hand can successfully perform the movements required to grasp various kinds of tableware.

Caging Polygonal Objects Using Formationally Similar Three-Finger Hands

Caging offers a robust strategy for grasping objects with robot hands. This letter describes an efficient caging-to-grasping algorithm for polygonal objects using minimalistic three-finger robot hands. This letter describes how to cage and then grasp polygonal objects, using single actuator triangular three-finger formations, whose shape is determined by any desired immobilizing grasp of the polygonal object. While the hand's configuration space is four-dimensional, the algorithm uses the hand's two-dimensional contact space, which represents all two- and three-finger contacts along the grasped object boundary. This letter describes how the problem of computing the critical cage formation that allows the object to escape the hand is reduced to a search along a caging graph constructed in the hand's contact space. Starting from a desired immobilizing grasp, the graph is searched for the critical cage formation, which is used to determine the caging regions surrounding the immobilizing grasp. Any three-finger placement within these regions guarantees robust object grasping. The technique is demonstrated with a detailed computational example and a video clip, which shows caging experiments with a single actuator three-finger robot hand.

Grasp configuration planning for a low-cost and easy-operation underactuated three-fingered robot hand

This paper proposes a method for modeling and planning the grasping configuration of a robotic hand with underactuated finger mechanisms. The proposed modeling algorithm is based on analysis and mimicking of human grasping experience. Results of the analysis is preprocessed and stored in a database. The grasp configuration planning algorithm can be used within a real time online grasp control as based on artificial neural networks. Namely, shapes and sizes of task objects are described by taxonomy data, which are used to generate grasp configurations. Then, a robot hand grasp control system is designed as based on the proposed grasp planning with close-loop position and force feedback. Simulations and experiments are carried out to show the basic features of the proposed formulation for identifying the grasp configurations while dealing with target objects of different shapes and sizes. It is hoped that the well-trained underactuated robot hand can solve most of grasping tasks in our life. The research approach is aimed to research low-cost easy-operation solution for feasible and practical implementation.

Implementation of Two-Axis Position-Based Impedance Control with Inverse Kinematics Solution for A 2-DOF Robotic Finger

Position-based impedance control is a force control approach which consists of a single control law that accommodates the external force to achieve the desired dynamics of the body. A previously developed three-fingered robot hand was very rigid in its motion due to the application of position control alone. The position control scheme was inadequate for the tasks that involves the interaction of a robot end-effector with its environment which could damage fragile objects or be prone to slippage when provided with incorrect object's position. This paper introduces the application of two-axis position-based impedance control to one of the 2 degree-of-freedom (DOF) robotic finger of the robot hand. The goal of the control is to produce a mass-spring-dashpot system for the robot hand which considers the external force exerted by the object or environment onto the finger to modify the targeted position of the robot's tip-end. The position-based impedance control which was successfully performed however could not directly drive the DC-micromotors at the finger joints since it was expressed in the Cartesian position (X,Y,Z) form. Therefore, inverse kinematics was derived using geometrical approach to convert the Cartesian position (X,Y,Z) to angle position of motor which is controlled by PID. The proposed control and the developed kinematics were programmed using Matlab Simulink and tested in real-time experiments. The validation result has proven that the proposed position-based impedance control could modify the initial fingertip position according to the amount and direction of the applied external force, thus produced softness to the robotic finger.

A 3D-printed Robot Hand with Three Linkage-driven Underactuated Fingers

In this study, we improved an underactuated finger mechanism by using Solidworks to simulate the grasp operation of a finger in some different situations. In addition, a robot palm is designed for the three-finger robot hand with the designed underactuated fingers. A Solidworks simulation was used to verify the rationality of the design. Some parts of the hand were modified to fit for 3D printing, and a prototype of the hand was produced by 3D printing, which could reduce the cost of the production process, as well as provide design flexibility and other advantages. Finally, some grasping experiments were made with the prototype. The results showed that the robot could grasp objects with different sizes, and further verified the rationality of the design and feasibility of fabricating the robot hand using 3D printing.

Feeling the Shape: Active Exploration Behaviors for Object Recognition With a Robotic Hand

Autonomous exploration in robotics is a crucial feature to achieve robust and safe systems capable to interact with and recognize their surrounding environment. In this paper, we present a method for object recognition using a three-fingered robotic hand actively exploring interesting object locations to reduce uncertainty. We present a novel probabilistic perception approach with a Bayesian formulation to iteratively accumulate evidence from robot touch. Exploration of better locations for perception is performed by familiarity and novelty exploration behaviors, which intelligently control the robot hand to move toward locations with low and high levels of interestingness, respectively. These are active behaviors that, similar to the exploratory procedures observed in humans, allow robots to autonomously explore locations they believe that contain interesting information for recognition. Active behaviors are validated with object recognition experiments in both offline and real-time modes. Furthermore, the effects of inhibiting the active behaviors are analyzed with a passive exploration strategy. The results from the experiments demonstrate the accuracy of our proposed methods, but also their benefits for active robot control to intelligently explore and interact with the environment.

Workspace of Multifingered Hands Using Monte Carlo Method

Abstract Multifingered hands have the capability of dexterous manipulation of grasped objects and thus significantly increase the capabilities of a robot equipped with multifingered hands. Inspired by a multijointed human finger and the hand, we propose a six degree-of-freedom (DOF) model of a three-fingered robotic hand as a parallel manipulator. Two kinds of contact, namely point contact with friction and rolling without slipping between the fingertips and the grasped object, are considered. The point contact with friction is modeled as a three DOF spherical joint, and for rolling without slipping, we use the resultant nonholonomic constraints between the grasped object and the fingers. With realistic limits on the joints in the fingers and dimensions of finger segments, we obtain the well-conditioned manipulation workspace of the parallel manipulator using a Monte Carlo-based method. Additionally, we present two new general results—it is shown that maximum position and orientation workspace is obtained when the cross-sectional area of the grasped object is approximately equal to the area of the palm of the hand and when rolling without slipping is ensured the size of the well-conditioned workspace is significantly larger (∼1.2–1.5 times). We also present representative experiments of manipulation by a human hand and show that the experimental results are in reasonable agreement with those obtained from simulations.

Variable motion mapping to enhance stiffness discrimination and identification in robot hand teleoperation

In robot hand teleoperation system, force feedback is critical for operator to perceive the physical property of grasped objects. However, it is challenging to implement accurate force feedback if the force accuracy of a haptic device is limited. This paper proposes a variable motion mapping method to enhance stiffness discrimination of the remote object. In this method, the motion mapping coefficient is regulated according to the object stiffness so that the operator can perceive the stiffness difference of the remote objects. To validate the proposed approach, we conducted two experiments on a three-finger robot hand (BarrettHand BH8-280) teleoperation system. In the first experiment, we measured the minimum relative change in feedback stiffness that could be detected by the operator. The result shows that the relative change should be 70% for a correct detection rate of about 91%. Based on the result, the motion mapping coefficient was selected for the system. In the second experiment, the variable motion mapping method was compared to a constant motion mapping method. The experimental task in the comparison is identifying four objects through stiffness perception alone. The experimental results show that the variable motion mapping method enhances the discrimination of the objects. The operator can identify individual object within a group of four without visual feedback with the variable motion mapping method.

An Adaptive Three-Fingered Prismatic Gripper With Passive Rotational Joints

In this letter, we present the design of an underactuated three-fingered robotic hand and evaluate its performance. The hand utilizes radially symmetric, prismatically actuated fingers controlled by a single actuator. Each finger consists of a single joint finger connected to the prismatic joint via a passive rotational joint perpendicular to the palm. The rotational joints allow the fingers to passively switch between spherical and cylindrical grasps while the finger joint allows the fingers to wrap about the grasped object. We compare the performance of this design to that of a concentric gripper with cylindrical fingers and two other underactuated hand designs using the YCB grasping benchmark. This evaluation shows that the three finger prismatic hand performs well especially when equipped with the single joint fingers in comparison to other designs.

FINGERTIP STRUCTURAL ANALYSIS–A SIMULATED DESIGN EVALUATION

This paper discusses the result on structural analysis of a recently designed fingertip for a three-fingered robot hand. This analysis is necessary to understand the structural changes occur under certain range of external force acting on the fingertip. The key parameters to be observed are deformation and equivalent (von-misses) stress. These simulated tests had given good results since there was no permanent deformation occurred and no yielding happened even in the area that had been considered critical. The fingertip will be produced as a prototype and equipped with the miniature load cells for force control of the three-fingered robot hand in the future.

Visual Servoing for Object Manipulation with a Multifingered Robot Hand

This paper deals with the problem of controlling the 3D Cartesian position of an object grasped by a multifingered robot hand using the visual servoing approach. We consider that the geometry of the object is known and the contact constraints are always satisfied during the grasping motion. An RGB camera and a depth sensor are simultaneously used to provide visual positioning and depth information of the grasped object. The control design is based on the kinematic control approach and two visual servo controllers are developed in the image space, ensuring the zero-convergence of the output tracking errors and high disturbance-rejection capability. Practical tests for visual tracking tasks of a grasped object are performed with a three-fingered robot hand and a natural interface device. Experimental results are shown to illustrate the performance and feasibility of the proposed methodology.

Error-tolerant manipulation by caging

This paper delivers a preliminary attempt to find the optimized caging positions for a three-finger robotic hand designed for home-use logistical environments. The main idea behind optimizing caging positions falls in that optimal caging can afford largest margins to stop target objects from escaping into infinity. By employing the advantages of largest margins, optimal caging grasp can be robust enough to endure dramatic perception noises or errors and low sensing resolutions. This paper optimizes object grasping towards caging. Specifically, our algorithm utilizes Genetic Algorithm (GA) to accelerate the searching procedure and evaluate a fitness of the GA population by examining a combination of max–min, which corresponds to intersections of neighbour fingers’ CC space margins, and least inter-finger distance for optimization. Simulation results show that the manipulation strategy proposed in this paper could in the worse case coordinate with sensors whose resolution are less than one pixel per centimeter.

Design and control of a three-fingered tendon-driven robotic hand with active and passive tendons

This paper presents a design of a three-fingered robotic hand driven by active and passive tendons and proposes control methods for this hand. The tendon-driven robotic hand consists of the thumb, the index and the middle fingers. The robotic thumb can move all the joints independently. In contrast, the index and the middle robotic fingers are under-actuated using the combination of active and passive tendons, and move the terminal two joints synchronously, which is one of the important features of the human digits. We present passivity-based impedance and force controllers for tendon-driven robotic fingers and discuss how to combine them for fast and secure grasps. We experimentally validate that the robotic hand moves fast and manipulates an object and demonstrate that the robotic hand grasps objects in diverse ways.

Software Modules for Command and Control of the Anthropomorphic Gripper for Robots

This paper presents two software modules of the general structure and implementation features of the original software for command and control of the anthropomorphic gripper for robots. The software has three modules: RoboVISION, RoboSIM and RoboCOMMANDER. RoboVISION is a software module for tracing, visualization, shape recognition, measurement, and efficiency evaluation for 3D model generation of objects to grip by an anthropomorphic gripper. It was presented in another paper. This paper discribes RoboSIM nad Robo COMMANDER. RoboSIM product is an advanced software module that aims primarily to reduce achievement time of functional simulation of anthropomorphic grippers. RoboCOMMANDER is a software module for controlling a three-fingered anthropomorphic robotic hand in both virtual and real environments. The approach is using a multimodal interface based on voice commands, remote manipulation, and smart, remote controlling mobile devices.

Soft Object Deformation Monitoring and Learning for Model-Based Robotic Hand Manipulation

This paper discusses the design and implementation of a framework that automatically extracts and monitors the shape deformations of soft objects from a video sequence and maps them with force measurements with the goal of providing the necessary information to the controller of a robotic hand to ensure safe model-based deformable object manipulation. Measurements corresponding to the interaction force at the level of the fingertips and to the position of the fingertips of a three-finger robotic hand are associated with the contours of a deformed object tracked in a series of images using neural-network approaches. The resulting model captures the behavior of the object and is able to predict its behavior for previously unseen interactions without any assumption on the object's material. The availability of such models can contribute to the improvement of a robotic hand controller, therefore allowing more accurate and stable grasp while providing more elaborate manipulation capabilities for deformable objects. Experiments performed for different objects, made of various materials, reveal that the method accurately captures and predicts the object's shape deformation while the object is submitted to external forces applied by the robot fingers. The proposed method is also fast and insensitive to severe contour deformations, as well as to smooth changes in lighting, contrast, and background.