The GRASP Taxonomy of Human Grasp Types
TL;DR: The resulting taxonomy incorporates all grasps found in the reviewed taxonomies that complied with the grasp definition and is shown that due to the nature of the classification, the 33 grasp types might be reduced to a set of 17 more generalgrasps if only the hand configuration is considered without the object shape/size.
Abstract: In this paper, we analyze and compare existing human grasp taxonomies and synthesize them into a single new taxonomy (dubbed “The GRASP Taxonomy” after the GRASP project funded by the European Commission). We consider only static and stable grasps performed by one hand. The goal is to extract the largest set of different grasps that were referenced in the literature and arrange them in a systematic way. The taxonomy provides a common terminology to define human hand configurations and is important in many domains such as human–computer interaction and tangible user interfaces where an understanding of the human is basis for a proper interface. Overall, 33 different grasp types are found and arranged into the GRASP taxonomy. Within the taxonomy, grasps are arranged according to 1) opposition type, 2) the virtual finger assignments, 3) type in terms of power, precision, or intermediate grasp, and 4) the position of the thumb. The resulting taxonomy incorporates all grasps found in the reviewed taxonomies that complied with the grasp definition. We also show that due to the nature of the classification, the 33 grasp types might be reduced to a set of 17 more general grasps if only the hand configuration is considered without the object shape/size.
Citations
More filters
TL;DR: This work uses reinforcement learning (RL) to learn dexterous in-hand manipulation policies that can perform vision-based object reorientation on a physical Shadow Dexterous Hand, and these policies transfer to the physical robot despite being trained entirely in simulation.
Abstract: We use reinforcement learning (RL) to learn dexterous in-hand manipulation policies that can perform vision-based object reorientation on a physical Shadow Dexterous Hand. The training is performed...
1,428 citations
TL;DR: Tactile patterns obtained from a scalable sensor-embedded glove and deep convolutional neural networks help to explain how the human hand can identify and grasp individual objects and estimate their weights.
Abstract: Humans can feel, weigh and grasp diverse objects, and simultaneously infer their material properties while applying the right amount of force—a challenging set of tasks for a modern robot1. Mechanoreceptor networks that provide sensory feedback and enable the dexterity of the human grasp2 remain difficult to replicate in robots. Whereas computer-vision-based robot grasping strategies3–5 have progressed substantially with the abundance of visual data and emerging machine-learning tools, there are as yet no equivalent sensing platforms and large-scale datasets with which to probe the use of the tactile information that humans rely on when grasping objects. Studying the mechanics of how humans grasp objects will complement vision-based robotic object handling. Importantly, the inability to record and analyse tactile signals currently limits our understanding of the role of tactile information in the human grasp itself—for example, how tactile maps are used to identify objects and infer their properties is unknown6. Here we use a scalable tactile glove and deep convolutional neural networks to show that sensors uniformly distributed over the hand can be used to identify individual objects, estimate their weight and explore the typical tactile patterns that emerge while grasping objects. The sensor array (548 sensors) is assembled on a knitted glove, and consists of a piezoresistive film connected by a network of conductive thread electrodes that are passively probed. Using a low-cost (about US$10) scalable tactile glove sensor array, we record a large-scale tactile dataset with 135,000 frames, each covering the full hand, while interacting with 26 different objects. This set of interactions with different objects reveals the key correspondences between different regions of a human hand while it is manipulating objects. Insights from the tactile signatures of the human grasp—through the lens of an artificial analogue of the natural mechanoreceptor network—can thus aid the future design of prosthetics7, robot grasping tools and human–robot interactions1,8–10. Tactile patterns obtained from a scalable sensor-embedded glove and deep convolutional neural networks help to explain how the human hand can identify and grasp individual objects and estimate their weights.
623 citations
TL;DR: A model of hands and bodies interacting together and fit it to full-body 4D sequences that move naturally with detailed hand motions and a realism not seen before in full body performance capture is formulated.
Abstract: Humans move their hands and bodies together to communicate and solve tasks. Capturing and replicating such coordinated activity is critical for virtual characters that behave realistically. Surprisingly, most methods treat the 3D modeling and tracking of bodies and hands separately. Here we formulate a model of hands and bodies interacting together and fit it to full-body 4D sequences. When scanning or capturing the full body in 3D, hands are small and often partially occluded, making their shape and pose hard to recover. To cope with low-resolution, occlusion, and noise, we develop a new model called MANO (hand Model with Articulated and Non-rigid defOrmations). MANO is learned from around 1000 high-resolution 3D scans of hands of 31 subjects in a wide variety of hand poses. The model is realistic, low-dimensional, captures non-rigid shape changes with pose, is compatible with standard graphics packages, and can fit any human hand. MANO provides a compact mapping from hand poses to pose blend shape corrections and a linear manifold of pose synergies. We attach MANO to a standard parameterized 3D body shape model (SMPL), resulting in a fully articulated body and hand model (SMPL+H). We illustrate SMPL+H by fitting complex, natural, activities of subjects captured with a 4D scanner. The fitting is fully automatic and results in full body models that move naturally with detailed hand motions and a realism not seen before in full body performance capture. The models and data are freely available for research purposes at http://mano.is.tue.mpg.de.
536 citations
TL;DR: In this paper, a low-cost open-source consumer 3D printer and a commercially available printing material were identified for printing soft pneumatic actuators with complex inner geometry and high degree of freedom.
Abstract: This work presents a novel technique for direct 3D printing of soft pneumatic actuators using 3D printers based on fused deposition modeling (FDM) technology. Existing fabrication techniques for soft pneumatic actuators with complex inner geometry are normally time-consuming and involve multistep processes. A low-cost open-source consumer 3D printer and a commercially available printing material were identified for printing soft pneumatic actuators with complex inner geometry and high degree of freedom. We investigated the material properties of the printing material, simulated the mechanical behavior of the printed actuators, characterized the performances of the actuators in terms of their bending capability, output forces, as well as durability, and demonstrated the potential soft robotic applications of the 3D printed actuators. Using the 3D printed actuators, we developed a soft gripper that was able to grasp and lift heavy objects with high pay–-to-weight ratio, which demonstrated that the ...
413 citations
15 Jun 2019
TL;DR: This work presents an end-to-end learnable model that exploits a novel contact loss that favors phys- ically plausible hand-object constellations, and improves grasp quality metrics over baselines, using RGB images as input.
Abstract: Estimating hand-object manipulations is essential for in- terpreting and imitating human actions. Previous work has made significant progress towards reconstruction of hand poses and object shapes in isolation. Yet, reconstructing hands and objects during manipulation is a more challeng- ing task due to significant occlusions of both the hand and object. While presenting challenges, manipulations may also simplify the problem since the physics of contact re- stricts the space of valid hand-object configurations. For example, during manipulation, the hand and object should be in contact but not interpenetrate. In this work, we regu- larize the joint reconstruction of hands and objects with ma- nipulation constraints. We present an end-to-end learnable model that exploits a novel contact loss that favors phys- ically plausible hand-object constellations. Our approach improves grasp quality metrics over baselines, using RGB images as input. To train and evaluate the model, we also propose a new large-scale synthetic dataset, ObMan, with hand-object manipulations. We demonstrate the transfer- ability of ObMan-trained models to real data.
343 citations
Cites background from "The GRASP Taxonomy of Human Grasp T..."
...We find that in the grasps produced by GraspIt, power grasps, as defined by [3] in which larger surfaces of the hand and the object are in contact, are rarely produced....
[...]
References
More filters
TL;DR: It is shown that movements of the hand consist of two basic patterns of movements which are termed precision grip and power grip, which appear to cover the whole range of prehensile activity of the human hand.
Abstract: 1. The prehensile movements of the hand as a whole are analysed from both an anatomical anda functional viewpoint. 2. It is shown that movements of the hand consist of two basic patterns of movements which are termed precision grip and power grip. 3. In precision grip the object is pinched between the flexor aspects of the fingers and that of the opposing thumb. 4. In power grip the object is held as in a clamp between the flexed fingers and the palm, counter pressure being applied by the thumb lying more or less in the plane of the palm. 5. These two patterns appear to cover the whole range of prehensile activity of the human hand.
1,446 citations
01 Jun 1989
TL;DR: Comparisons of the grasp taxonomy, the expert system, and grasp-quality measures derived from the analytic models reveal that the analytic measures are useful for describing grasps in manufacturing tasks despite the limitations in the models.
Abstract: Current analytical models of grasping and manipulation with robotic hands contain simplifications and assumptions that limit their application to manufacturing environments. To evaluate these models, a study was undertaken of the grasps used by machinists in a small batch manufacturing operation. Based on the study, a taxonomy of grasps was constructed. An expert system was also developed to clarify the issues involved in human grasp choice. Comparisons of the grasp taxonomy, the expert system, and grasp-quality measures derived from the analytic models reveal that the analytic measures are useful for describing grasps in manufacturing tasks despite the limitations in the models. In addition, the grasp taxonomy provides insights for the design of versatile robotic hands for manufacturing. >
1,414 citations
"The GRASP Taxonomy of Human Grasp T..." refers background or methods in this paper
...The taxonomy of Cutkosky [53], which is widely used in the field of robotics, lists 15 different grasps....
[...]
...While Cutkosky [53] distinguishes grasps also by object size, many other authors do not....
[...]
...[51] M. Cutkosky and P. Wright, “Modeling manufacturing grips and correlations with the design of robotic hands,” in Proc....
[...]
...As Cutkosky mainly differentiates grasps by the object properties, this reduction is only natural....
[...]
...[53] M. R. Cutkosky, “On grasp choice, grasp models, and the design of hands for manufacturing tasks,” IEEE Trans....
[...]
Book•
01 Jan 1974
TL;DR: Functional anatomy of the tibiofibular joints explains the architecture of the lower limb and orientation of the articular surfaces of the subtalar joint.
Abstract: Chapter 1: The Hip The Hip Joint (Coxo-femoral Joint) The hip: the joint at the root of the lower limb Movements of flexion at the hip joint Movements of extension at the hip joint Movements of abduction at the hip joint Movements of adduction at the hip joint Movements of axial rotation at the hip joint Movements of circumduction at the hip joint Orientation of the femoral head and of the acetabulum Relationships of articular surfaces Architeccture of the femur and of the pelvis The acetabular labrum and the ligament of the head of femur The capsular ligament of the hip joint The ligaments of the hip joint Role of the ligaments in flexion-extension Role of the ligaments in lateral-medial rotation Role of the ligaments in adduction-abduction Functional anatomy of the ligament of head of femur Coaptation of the articular surfaces of the hip joint Muscular and bony factors maintaining the stability of the hip joint The flexor muscles of the hip joint The extensor muscles of the hip joint The abductor muscles of the hip joint Hip abduction Transverse stability of the pelvis The adductor muscles of the hip joint The lateral rotator muscles of the hip joint The rotator muscles of the hip joint Inversion of muscular actions Successive recruitment of the abductor muscles Chapter 2: The Knee The axes of the knee joint Medial and lateral deviations of the knee Movements of flexion-extension Axial rotation of the knee General architecture of the lower limb and orientation of the articular surfaces The articular surfaces of flexion-extension The tibial articular surfaces in relation to axial rotation Profiles of the femoral condyles and of the tibial articular surfaces Determinants of the condylotrochlear profile Movements of the femoral condyles on the tibial plateau during flexion-extension Movements of the femoral condyles on the tibial plateau during axial rotation The articular capsule The ligamentum mucosum, the synovial plicae and the joint capacity The inter-articular menisci Meniscal displacements during flexion-extension Meniscal displacements during axial rotation - meniscal lesions Patellar displacements relative to the femur Femoropatellar relationships Patellar movements relative to the tibia The collateral ligaments of the knee Transverse stability of the knee Anteroposterior stability of the knee The peri-articular defence system of the knee The cruciate ligaments of the knee Relations between the capsule and the cruciate ligaments Direction of the cruciate ligaments Mechanical role of the cruciate ligaments Rotational stability of the extended knee Dynamic tests of the knee during medial rotation Dynamic tests for rupture of the anterior cruciate ligament Dynamic tests of the knee during lateral rotation The extensor muscles of the knee Physiological actions of the rectus femoris The flexor muscles of the knee The rotator muscles of the knee Automatic rotation of the knee Dynamic equilibrium of the knee Chapter 3: The Ankle The articular complex of the foot Flexion-extension The articular surfaces of the ankle joint The ligaments of the ankle joint Anteroposterior stability of the ankle and factors limiting flexion-extension Transverse stability of the ankle joint The tibiofibular joints Functional anatomy of the tibiofibular joints Why does the leg have two bones? Chapter 4: The Foot Axial rotation and side-to-side movements of the foot The articular surfaces of the subtalar joint Congruence and incongruence of the articular surfaces of the subtalar joint The talus: the unusual bone The ligaments of the subtalar joint The transverse tarsal joint and its ligaments Movements at the subtalar joint Movements at the subtalar and transverse tarsal joints Movements at the transverse tarsal joint Overall functioning of the posterior tarsal joints The heterokinetic universal joint of the hindfoot The ligamentous chains during inversion and eversion The cuneonavicular, intercuneiform and tarsometatarsal joints Movements at the anterior tarsal and tarsometatarsal joints Extension of the toes The compartments of the leg The interosseous and the lumbrical muscles The muscles of the sole of the foot The fibrous tunnels of the instep and of the sole of the foot The flexor muscles of the ankle The triceps surae The other extensor muscles of the ankle The abductor-pronator muscles: the fibularis muscles The adductor-supinator muscles:the tibialis muscles Chapter 5: The Plantar Vault Overview of the plantar vault The medial arch The lateral arch The anterior arch and the transverse arch of the foot The distribution of loads and static distortions of the plantar vault Architectural equilibrium of the foot Dynamic distortions of the plantar vault during walking Dynamic distortions of the plantar vault secondary to inclination of the leg on the inverted foot Dynamic distortions of the plantar vault secondary to inclination of the leg on the everted foot Adaptation of the plantar vault to the terrain The various types of pes cavus The various types of pes planus Imbalances of the interior arch Types of feet Chapter 6: Walking The move to bipedalism The miracle of bipedalism The initial step Swing phase of the gait cycle Loading response phase The footprints Pelvic oscillations The tilts of the pelvis Torsion of the trunk Swinging of the upper limbs Muscular chains during running Appendices Walking is freedom The nerves of the lower limb The sensory compartments of the lower limb (text) The sensory compartments of the lower limb: Figures 1 and 2 Bibliography Models of Articular Biomechanics
991 citations
Book•
[...]
20 Apr 2006
TL;DR: This book discusses the evolution and anatomy of the hand, sensory neurophysiology, and applications across the lifespan, as well as some of the applications currently in use.
Abstract: 1. Historical Overview and general introduction 2. Evolutionary development and anatomy of the hand 3. Sensory neurophysiology 4. Tactile sensing 5. Active haptic sensing 6. Prehension 7. Non-prehensile skilled movements 8. End-effector constraints 9. Hand function across the lifespan 10. Applications 11. Summary, conclusions and future directions
544 citations
"The GRASP Taxonomy of Human Grasp T..." refers background in this paper
...in more than 20 degrees of freedom [15], [16]....
[...]
TL;DR: Extrinsic and intrinsic tendons of the thumb were found to sustain forces of up to 10.0 and thirty kilograms during pinch, producing five kilograms of force at the thumb tip and forces of as much as fifty kilograms during grasp.
Abstract: Using a three-dimensional analysis, the internal forces in the joints and soft tissues of the thumb during pinch and grasp were calculated. To do this, mechanical equivalents were ascribed to the anatomical system, the joint orientation and tendon locations were determined from biplanar roentgenograms of five marked normal cadaver specimens, and the magnitudes of forces in the tendons, intrinsic muscles, joint contact surfaces, and constraining ligaments were calculated based on assumed loads applied to the tip of the thumb in various types of pinch and grasp. These results are the direct extension of a two-dimensional analysis that proved inadequate for the determination of static tendon and joint forces. Extrinsic and intrinsic tendons of the thumb were found to sustain forces of up to 10.0 and thirty kilograms during pinch, producing five kilograms of force at the thumb tip and forces of as much as fifty kilograms during grasp. The joint compression (contact) forces averaged three kilograms of force at the interphalangeal joint, 5.4 kilograms at the metacarpophalangeal joint, and 12.0 kilograms at the carpometacarpal joint during simple pinch (one kilogram of applied force). Compression forces of as much as 120 kilograms may occur at the carpometacarpal joint during strong grasp.
541 citations