Humanoid grasping is a critical ability for anthropomorphic hand, and plays a significant role in the development of humanoid robots. In this article, we present a deep learning-based control framework for humanoid grasping, incorporating the dynamic contact process among the anthropomorphic hand, the object, and the environment. This method efficiently eliminates the constraints imposed by inaccessible grasping points on both the contact surface of the object and the table surface. To mimic human-like grasping movements, an underactuated anthropomorphic hand is utilized, which is designed based on human hand data. The utilization of hand gestures, rather than controlling each motor separately, has significantly decreased the control dimensionality. Additionally, a deep learning framework is used to select gestures and grasp actions. Our methodology, proven both in simulation and on real robot, exceeds the performance of static analysis-based methods, as measured by the standard grasp metric Q1. It expands the range of objects the system can handle, effectively grasping thin items such as cards on tables, a task beyond the capabilities of previous methodologies.

Robotic hands have long strived to reach the performance of human hands. The physical complexity and extraordinary capabilities of the human hand, in terms of sensing, actuation, and cognitive abilities, make achieving this goal challenging. At the heart of the physical structure of the hand is its\' passive behaviors. Seen most clearly in soft robotic hands, these behaviors influence and affect the mechanical, sensing, and control functionalities. With this perspective, we present a framework through which passivity in robot hands can be understood, by concretely identifying the role of passivity in the design, fabrication, and control of soft hands. In this framework we focus on the interactions between the physical hand and the: environment, internal actuation, sensor morphology, and wrist control. Taking these surrounding systems away, we are left with a passive soft hand whose behaviors emerge from external interactions. Inspired by the human hand, we define the role of these four key interacting pillars and review how state-of-the art robot hands utilize these four elements to aid functionality. We show how these pillars promote hybrid soft-rigid hands with rich behaviors, providing benefits in terms of the increased adaptability to uncertain environments, improved scalability and reduction in the cost of actuation, sensing, and control. This review provides a conceptual framework for approaching hand design and analysis through consideration of the passive behaviors. This highlights not only the advances that can be made by approaching the problem in this way but also the outstanding challenges that stem from this outlook.

The postural synergies have great potentials to replicate human grasp characteristics, simplify grasp control and reduce the number of hardware needed actuators. However, due to the complex mapping relationship and jagged transmission ratio, the implemented mechanisms are always too bulky and loose which greatly limits its application. For current solutions, the replicating accuracy of motion characteristics or control intuition are compromised, and hitherto no work reports the replicating errors in literatures. To overcome these limitations, we present a novel design framework to determine the actuation configuration, implemented scheme and physical parameters. In this way, the mechanism is miniaturized and can be compactly embedded in hand palm, a self-contained synergy robot hand that integrated with mechanism, sensors and suited electrical system is built. The experiments demonstrate that the robot hand can accurately replicate the motion characteristics of two primary synergies, keep the control intuition to simplify grasp control, perform a better anthropomorphic motion capability and grasp different objects with versatile grasp functionality.

Soft robotic hands provide better safety and adaptability than rigid robotic hands. Furthermore, a multijointed structure that imitates the movement of a human hand represents significant progress in realizing its anthropomorphism. In this study, we present a multijointed pneumatic soft anthropomorphic hand that is capable of expressing letters through sign language and grasping different objects using three grasping modes, namely thumb grasping, precision grasping, and power grasping. This novel soft hand is composed of multijointed soft fingers, a thumb, thenar, and 3D-printed palm. Tests were performed to characterize the displacement track and force performance of the fingers, thumb, and thenar, which was made by mold casting silicone rubber. In addition, a dedicated pneumatic control system was designed and built to enable the soft hand to automatically perform the tasks set by specific programs. This new multijointed hand with a flexible thenar represents significant progress in the development of anthropomorphic bionic hands, offering the benefits of fast response, low cost, as well as ease of fabrication, assembly, and replacement.

The proposal of postural synergy theory has provided a new approach to solve the problem of controlling anthropomorphic hands with multiple degrees of freedom. However, generating the grasp configuration for new tasks in this context remains challenging. This study proposes a method to learn grasp configuration according to the shape of the object by using postural synergy theory. By referring to past research, an experimental paradigm is first designed that enables the grasping of 50 typical objects in grasping and operational tasks. The angles of the finger joints of 10 subjects were then recorded when performing these tasks. Following this, four hand primitives were extracted by using principal component analysis, and a low-dimensional synergy subspace was established. The problem of planning the trajectories of the joints was thus transformed into that of determining the synergy input for trajectory planning in low-dimensional space. The average synergy inputs for the trajectories of each task were obtained through the Gaussian mixture regression, and several Gaussian processes were trained to infer the inputs trajectories of a given shape descriptor for similar tasks. Finally, the feasibility of the proposed method was verified by simulations involving the generation of grasp configurations for a prosthetic hand control. The error in the reconstructed posture was compared with those obtained by using postural synergies in past work. The results show that the proposed method can realize movements similar to those of the human hand during grasping actions, and its range of use can be extended from simple grasping tasks to complex operational tasks.

The structure of the human musculo-skeletal systems shows complex passive dynamic properties, critical for adaptive grasping and motions. Through wrist and arm actuation, these passive dynamic properties can be exploited to achieve nuanced and diverse environment interactions. We have developed a passive anthropomorphic robot hand that shows complex passive dynamics. We require arm/wrist control with the ability to exploit these. Due to the soft hand structures and high degrees of freedom during passive-object interactions, bespoke generation of wrist trajectories is challenging. We propose a new approach, which takes existing wrist trajectories and adapts them to changes in the environment, through analysis and classification of the interactions. By analysing the interactions between the passive hand and object, the required wrist motions to achieve them can be mapped back to control of the hand. This allows the creation of trajectories which are parameterized by object size or task. This approach shows up to 86% improvement in grasping success rate with a passive hand for object size changes up to +/-50%.

We propose a novel wearable robotic glove or exo-glove design scalable to the variation of the hand kinematics. While most of the traditional robot hand is driven by rotating the joint directly with a rigid body, our exo-glove deforms a robotic finger\'s skin and, thus, the hand skeleton joints. Multiple tendons woven on the exo-glove\'s surface can make multi-DOF finger joint motions. We allocated tendons to mimic a hand\'s intrinsic and extrinsic muscles. Thus, a robotic hand actuated with the exo-glove can perform natural finger motions, including abduction/adduction and flexion/extension of finger joints. Moreover, additional tendons for the thumb enable power grips and the robotic hand\'s human-like motion. The proposed design approach places all the actuators on the surface without directly actuating any of the hand skeleton\'s joint. Therefore, a random hand skeleton can work as a robotic hand by putting the wearable robotic glove on it. Thus, the proposed model provides a high degree of freedom on choosing hand skeletons. We expect the aforementioned biomimetic features of our proposed method will benefit not only traditional robotic hands design but also the design of prosthetic hands and robot power-assisted hand glove.

Affordable 3D-printed tendon-driven prosthetic hands are a rising trend because of their availability and easy customization. Nevertheless, comparative studies about the functionality of this kind of prostheses are lacking. The tradeoff between the number of actuators and the grasping ability of prosthetic hands is a relevant issue in their design. The analysis of synergies among fingers is a common method used to reduce dimensionality without any significant loss of dexterity. Therefore, the purpose of this study is to assess the functionality and motion synergies of different tendon-driven hands using an able-bodied adaptor. The use of this adaptor to control the hands by means of the fingers of healthy subjects makes it possible to take advantage of the human brain control while obtaining the synergies directly from the artificial hand. Four artificial hands (IMMA, Limbitless, Dextrus v2.0, InMoov) were confronted with the Anthropomorphic Hand Assessment Protocol, quantifying functionality and human-like grasping. Three subjects performed the tests by means of a specially designed able-bodied adaptor that allows each tendon to be controlled by a different human finger. The tendon motions were registered, and correlation and principal component analyses were used to obtain the motion synergies. The grasping ability of the analyzed hands ranged between 48 and 57% with respect to that of the human hand, with the IMMA hand obtaining the highest score. The effect of the subject on the grasping ability score was found to be non-significant. For all the hands, the highest tendon-pair synergies were obtained for pairs of long fingers and were greater for adjacent fingers. The principal component analysis showed that, for all the hands, two principal components explained close to or more than 80% of the variance. Several factors, such as the friction coefficient of the hand contact surfaces, limitations on the underactuation, and impairments for a correct thumb opposition need to be improved in this type of prostheses to increase their grasping stability. The principal components obtained in this study provide useful information for the design of transmission or control systems to underactuate these hands.

Underactuation is widely used when designing anthropomorphic hand, which involves fewer degrees of actuation than degrees of freedom. However, the similarities between coordinated joint movements and movement variances across different grasp tasks have not been suitably examined. This work suggests a systematic approach to identify the actuation strategy with the minimum number for degrees of actuation for anthropomorphic hands. This work evaluates the correlations of coordinated movements in human hands during 23 grasp tasks to suggest actuation strategies for anthropomorphic hands. Our approach proceeds as follows: first, we find the best description for each coordinated joint movement in each grasp task by using multiple linear regression; then, based on the similarities between joint movements, we classify hand joints into groups by using hierarchical cluster analysis; finally, we reduce the dimensionality of each group of joints by employing principal components analysis. The metacarpophalangeal joints and proximal interphalangeal joints have the best and most consistent description of their coordinated movements across all grasp tasks. The thumb metacarpophalangeal and abduction/adduction between the ring and little fingers exhibit relatively high independence of movement. The distal interphalangeal joints show a high degree of independent movement but not for all grasp tasks. Analysis of the results indicates that for the distal interphalangeal joints, their coordinated movements are better explained when all fingers wrap around the object. Our approach fails to provide more information for the other joints. We conclude that 19 degrees of freedom for an anthropomorphic hand can be reduced to 13 degrees of actuation distributed between six groups of joints. The number of degrees of actuation can be further reduced to six by relaxing the dimensionality reduction criteria. Other resolutions are as follows: (a) the joint coupling scheme should be joint-based rather than finger-based and (b) hand designs may need to include finger abduction/adduction movements.

The ability to regulate the mechanical stiffness in a large range could be extremely important for soft robots to interact with the environment more effectively. In this article, we propose a novel chain-like granular jamming mechanism to achieve a large range of stiffness variation instantly, based on a method that is totally different from existing vacuum-based granular jamming systems. Theoretical modeling is introduced to find the best combination of granules to form the chain-like structure (CLS) and experiments are conducted to demonstrate it. The experimental results indicate that the novel jamming structure is able to achieve a stiffness variation range as large as 50.7 folds. To further validate the effectiveness of the CLS, a soft-rigid hybrid actuator based on the jamming structure is proposed and an integrated fabrication method is provided. Furthermore, an anthropomorphic hand based on the hybrid actuators is developed and the experimental results show that the hand is not only versatile enough to manipulate various objects with different weights, material properties, shapes, and surface characteristics at the soft state, like existing soft grippers, but also can lift heavy objects (1.5 kg in a cylindrical grasping gesture and 3.52 kg in a hook gesture) at the rigid state, which could be difficult for other soft grippers. Finally, the hand is integrated into our homemade service robot, significantly improving the practicability and safety of the robot when serving humans.