Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Anthropomorphism Index of Mobility for Artificial Hands

Anthropomorphism Index of Mobility for Artificial Hands Hindawi Applied Bionics and Biomechanics Volume 2019, Article ID 7169034, 11 pages https://doi.org/10.1155/2019/7169034 Research Article Immaculada Llop-Harillo , Antonio Pérez-González , and Verónica Gracia-Ibáñez Grupo de Biomecánica y Ergonomía, Departamento de Ingeniería Mecánica y Construcción, Universitat Jaume I (UJI), 12071, Spain Correspondence should be addressed to Immaculada Llop-Harillo; illop@uji.es Received 21 March 2019; Accepted 17 June 2019; Published 28 July 2019 Guest Editor: Francesca Cordella Copyright © 2019 Immaculada Llop-Harillo et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The increasing development of anthropomorphic artificial hands makes necessary quick metrics that analyze their anthropomorphism. In this study, a human grasp experiment on the most important grasp types was undertaken in order to obtain an Anthropomorphism Index of Mobility (AIM) for artificial hands. The AIM evaluates the topology of the whole hand, joints and degrees of freedom (DoFs), and the possibility to control these DoFs independently. It uses a set of weighting factors, obtained from analysis of human grasping, depending on the relevance of the different groups of DoFs of the hand. The computation of the index is straightforward, making it a useful tool for analyzing new artificial hands in early stages of the design process and for grading human-likeness of existing artificial hands. Thirteen artificial hands, both prosthetic and robotic, were evaluated and compared using the AIM, highlighting the reasons behind their differences. The AIM was also compared with other indexes in the literature with more cumbersome computation, ranking equally different artificial hands. As the index was primarily proposed for prosthetic hands, normally used as nondominant hands in unilateral amputees, the grasp types selected for the human grasp experiment were the most relevant for the human nondominant hand to reinforce bimanual grasping in activities of daily living. However, it was shown that the effect of using the grasping information from the dominant hand is small, indicating that the index is also valid for evaluating the artificial hand as dominant and so being valid for bilateral amputees or robotic hands. freedom (DoFs), range of motion, and weight and number 1. Introduction of actuators, but an index to compare those properties with In recent years, it has been an increasing development of new the human hand was not defined. Some other previous stud- affordable and anthropomorphic prosthetic hands [1, 2] as a ies tried to quantify the anthropomorphism of artificial consequence of the improvements in 3D-printing technolo- hands with a numerical index. Feix et al. [3] proposed a met- ric for comparing the anthropomorphic motion capability of gies. The human hand is a complex and marvelous tool whose dexterity has not been achieved by any artificial hand. robotic and prosthetic hands, the anthropomorphism index Evaluating the functional similarity of artificial hands with (AI), being its computation cumbersome and based only on the human hand is essential for improving current anthropo- the position and orientation of the distal phalanges in differ- morphic hand designs. Assessing the capability of the pros- ent GTs. Liarokapis et al. [4] defined an anthropomorphism theses to perform the main grasp types (GTs) of human index to assess the robot’s ability to mimic the human hand grasping could give an insight into the level of functionality based on the comparison of the finger phalanx workspaces restored in patients. Metrics or indexes that quantify numer- and also the workspaces of the fingers’ base frames. Liu ically the level of anthropomorphism are the way to grade et al. [5] proposed twelve quantified prosthetic hand anthro- human-likeness and to provide specifications for maximizing pomorphism evaluation indexes including physical and the anthropomorphic functionality while designing new actuation properties, among which is included a DoF config- artificial hands. uration evaluation. This index was based on a matrix of DoF Belter et al. [1] reviewed and compared the mechanical configuration where the element of the matrix is set to 1 if properties of different prosthetic hands, as their degrees of there exist an artificial DoF in the corresponding position, 2 Applied Bionics and Biomechanics PC1 PC2 PC3 PC4 PC5 tests developed in this study on the main GTs. Furthermore, a preliminary study [12] carried out on four human healthy subjects encouraged us to go deep in the study by increasing the number of subjects, improving the definition of the index, and widening the analysis of its validity to the different types of artificial hands. The AIM is intended to be a quick compu- tation index based on the biomechanics of the human hand and thus providing a way to compare their functional anthro- pomorphism. Moreover, the relevance of each DoF for func- tionality, obtained by tests on the human hand in this study, is intended to be useful for other applications in artificial Figure 1: Principal components of the degrees of freedom of the human hand performing activities of daily living obtained in [10] hand design. (PC1: digit arching, PC2: closure, PC3: palmar arching, PC4: lateral pinch, and PC5: opposition). 2. Materials and Methods otherwise is set to 0. However, this approach does not take into account the relevance of each DoF for grasping during 2.1. Human Grasp Experiment. With the purpose of taking activities of daily living (ADL) nor the underactuation in into account in the AIM the relevance of each DoF according the joints. Underactuation in artificial hands [6] allows to to its importance for functional grasping, an experiment to use less actuators than DoFs while keeping versatility to measure the kinematics of the human hand in functional adapt GTs to different object shapes. grasps was carried out. Twenty subjects, ten males and ten Prostheses design could be different depending on its use females, all of whom were right-handed and free of hand for a dominant or nondominant hand; however, in the case pathologies or injuries, performed the most relevant GTs of a patient who still has a healthy hand, the most appropriate for a nondominant hand to reinforce bimanual grasping in strategy would be to consider the remaining hand as domi- ADL (PP, EG, TP, and TVG [9]). Although the grasps were nant [7, 8]. Thereby, the design of the prosthesis should be selected for a nondominant hand (most common use of a focused for a nondominant hand reinforcing bimanual hand prosthesis for unilateral amputees), subjects were asked grasping. The importance of the different GTs for personal to perform grasps with their dominant hand to get the most autonomy of the patients in ADL has been studied previously natural performance of human grasping. The study was by the authors [9], being pulp pinch (PP) (26%), extension approved by the Ethics Committee of the University, and grip (EG) (20.8%), tripod pinch (TP) (10.4%), and transverse all the subjects gave their written informed consent. The ages volar grip (TVG) (8.7%), the most relevant GTs for a non- of the subjects ranged intentionally between 20 and 51, being dominant hand to reinforce bimanual grasping, representing the average 35 ± 8, in order to prevent kinematic alterations together with the nonprehensile one, almost 90% of relevance due to joint degeneration from ageing. Subjects were selected for autonomy. so that the distribution of hand sizes was representative of the In the previous studies by the authors [10], the posture of population [13]. The hand width ranged from 70 to 96 mm the right hand from 24 healthy subjects performing 24 repre- with an average of 83 mm, and the hand length ranged from sentative ADL was recorded with an instrumented glove. 170 to 210 mm with an average of 185 mm. Twelve objects of different sizes were selected from the ADL were selected from the WHO’s International Classifica- tion of Functioning, Disability and Health [11]. By applying Yale-CMU-Berkeley Object and Model Set [14], three for principal component analysis (PCA), five factors explaining each of the four GTs (PP, EG, TP, and TVG), in order to 73.7% of the variance were obtained. As shown in Figure 1, cover most common requirements in ADL for each one the five main principal components (PCs) of the DoFs of (Figure 2). The subjects were sitting with the hands in the the human hand in ADL were “PC1: digit arching” (flexion table in a comfortable way: the arms close to the body and of the interphalangeal joints), “PC2: closure” (combination parallel to the sagittal plane, the elbows flexed 90 , the wrist of abduction of the fingers, except for the thumb, with flexion on the edge of the table, and the hands laying on the table of the metacarpophalangeal joints), “PC3: palmar arching,” palms down in a natural posture. This was the starting and “PC4: lateral pinch” (represents the lateral opposition of the ending posture for each grasping action. Subjects were thumb to the index), and “PC5: opposition” (represents the instructed on the different GTs to perform with each object, pad-to-pad opposition of the thumb to the little finger). and objects to be grasped were situated one by one by the The aim of this study is to propose an index to measure researcher at a distance of 30 cm in front of the subjects. Sub- the anthropomorphism of prosthetic hands, based on the jects were free to practice the grasps to be sure that it is in the comparison of the topology of the whole hand (joints and correct posture before starting the recordings. The steps to DoFs) and on the possibility to control these DoFs indepen- perform the grasps during the experiment consisted of the dently. The computation of the index, referred to as Anthro- following: grasping the object from the table with the correct pomorphism Index of Mobility (AIM), should weight each hand posture/GT, lift it up during two seconds, and finally, DoF depending on its importance for grasping in ADL. To release the object again on the table and return the hand to define this importance, we used the information from previ- the starting position. The sequence of the twelve objects to ous experimental tests performed in the group and specific grasp during the experiment is shown in Figure 2 in the Applied Bionics and Biomechanics 3 (a) T large marker (TP) (b) T plate (EG) (c) T chips can (TVG) 01 02 03 (d) T small marker (PP) (e) T tuna can (TP) (f) T cracker box (EG) 04 05 06 (g) T coffee can (TVG) (h) T plastic pear (PP) (i) T golf ball (TP) 07 08 09 (j) T pudding box (EG) (k) T power drill (TVG) (l) T washer 10 mm (PP) 10 11 12 Figure 2: Grasping tasks of the experiment (a-l). T are the tasks ordered (g: indicates the order) followed by the object of the Yale-CMU- Berkeley Object and Model Set [14] to grasp and in brackets the grasp type to be performed in each task (TP: tripod pinch, EG: extension grip, TVG: transverse volar grip, and PP: pulp pinch). specified order. The experiment was repeated three times geal flexion (MCP1 to MCP5, 1 to 5 meaning thumb to little per subject. digits), interphalangeal flexion of the thumb (IP1), proximal The kinematics of the hand while performing the grasp- interphalangeal flexion of the fingers (PIP2 to PIP5), flex- ing postures was recorded (100 Hz) using an instrumented ion and abduction of the carpometacarpal joint of the right hand glove with 18 sensors (CyberGlove Systems LLC; thumb (CMC1), relative abduction between finger MCPs San Jose, CA). DoF kinematics corresponding to 16 joint (index-middle, middle-ring, and ring-little), and palmar angles (marked with an asterisk in Table 1) was obtained arching. Prior to the tests with objects, the CyberGlove was using a previously validated protocol [15]: metacarpophalan- calibrated for each subject following the calibration 4 Applied Bionics and Biomechanics Table 1: Joints and degrees of freedom (DoFs) of the human hand where the summation extends for i = 1,2,3,4, corresponding corresponding to the four different groups of DoFs defined. to each one of the four groups of DoFs (Table 1: F/E, AB/AD, P.ARC, and T.OPP), the factor k accounts for the type of Groups of DoFs Joints and DoFs of the human hand actuation of the DoFs included in this group, and the factor MCP2_Flexion w is a weighting coefficient depending on the relevance of this group of DoF for grasping in ADL. Both the term k PIP2_Flexion and the weighting factor w were defined to have a range DIP2_Flexion between 0 and 1, and the sum of weighting coefficients w ∗ i MCP3_Flexion for the four groups is unity, so that the AIM reach a maxi- PIP3_Flexion mum value of 1 for the human hand and a very low value DIP3_Flexion for an artificial hand with very poor anthropomorphism. Finger flexion-extension The factor k for each group i was defined to get a high (F/E) MCP4_Flexion value if the method of actuation for the DoFs in that group PIP4_Flexion allows to control them independently, as in the human hand, DIP4_Flexion and a lower value if the motions of these DoFs are highly MCP5_Flexion coupled during motion. To this end, each DoF in the evalu- ated hand was classified according to the types included in PIP5_Flexion Table 2. DIP5_Flexion The independent mobility of a DoF can be ranked quali- MCP2_Abduction tatively from better to worse, depending on its class, as MCP3_Abduction A>B>C>D>E. Note that B class was considered better than Finger abduction-adduction (AB/AD) C because it allows mechanical adaptation of the finger to MCP4_Abduction ∗ the shape of the object to be grasped and do not suffer from MCP5_Abduction mechanical singular configurations [6]. Pugh’s method used CMC5_Flexion Palmar arching in concept design evaluation [20] was employed to convert (P.ARC) CMC4_Flexion the ranked list of methods of actuation of the DoFs into a list of numerical coefficients c (last column in Table 2). However, CMC1_Flexion the independent mobility of a DoF is associated not only with CMC1_Abduction the type of actuation in this particular DoF but also with that Thumb opposition MCP1_Flexion of the DoFs more proximal in the same serial chain of a digit, (T.OPP) MCP1_Abduction i.e., for a finger, the mobility for flexion in the PIP joint is dependent on the mobility for flexion in the MCP joint. IP1_Flexion Consequently, for that case, the coefficient c for the DoF j ij 1: thumb, 2: index finger, 3: middle finger, 4: ring finger, 5: little finger; CMC: of the group i was obtained as the multiplication of the coef- carpometacarpal joint, MCP: metacarpophalangeal joint, PIP: proximal ficient c of this DoF and those located proximally in the same interphalangeal joint, DIP: distal interphalangeal joint, IP: interphalangeal joint; 16 joint angles measured during the experiment with the CyberGlove. serial kinematic chain. In addition, for assigning the coeffi- cient c to several DoFs underactuated by the same motor or procedure [15]. Starting and final positions while the hand is actuator, class A was considered for only one of them and not moving were trimmed from the recordings. Then, they class B or C for others. If a motor actuates several DoFs were filtered with a 2nd-order 2-way low-pass Butterworth included in different groups i, the coefficient 1 corresponding filter with cut-off frequency of 5 Hz [16, 17]. The tests were to class A was divided among the number of groups and this video recorded. fraction was assigned to only one of the DoFs in this group, being others classified as either B or C. Finally, the factor k 2.2. Index Definition. The Anthropomorphism Index of for each group i was defined with equation (2), by summing Mobility (AIM) for an artificial hand was defined based on the terms c in the group i and dividing by the number of ij two main factors: (1) the DoFs present in the hand along with DoFs of the human hand in this group (n ), which is, its method of actuation and (2) the relevance of these DoFs according to Table 1, 12 for i =1, 4 for i =2, 2 for i =3, for grasping in ADL. and 5 for i =4. The DoFs of the human hand (HH) [18, 19] were classi- fied into four different functional groups for defining the ∑ c j ij AIM (Table 1): finger flexion-extension (12 in HH), finger k = 2 abduction-adduction (4 in HH), palmar arching (2 in HH), and thumb opposition (5 in HH). The Anthropomorphism Index of Mobility (AIM) was The weighting factor w in equation (1), accounting for defined with the relative relevance of the DoFs of the group i for grasping in ADL, was defined with AIM =〠 k · w , 1 i i w =〠 r · s 3 i ik k k Applied Bionics and Biomechanics 5 Table 2: Classification of the DoF depending on the type of actuation and numeric coefficient associated. Class Type of actuation of the DoF c A DoF actuated by one independent motor or actuator 1 B DoF underactuated with other DoFs without a rigid coupling, allowing adaptive grasps (tendons, elastic elements) 0.75 C DoF underactuated with other DoFs with a rigid coupling, not allowing adaptive grasp (linkages) 0.5 D No actuation on the DoF, but passive motion allowed 0.25 E DoF absent in the artificial hand 0 In equation (3), r weights the relative contribution of ing [9]: 39.5% for PP, 31.6% for EG, 15.8% for TP, and ik the group of DoFs i (i = 1,2,3,4) in human hand functionality 13.2% for TVG. represented through PC (k = 1,2,3,4,5), corresponding to each of the five kinematic functional synergies (see ∑ f t=1 tk v = , gbp Figure 1) found in a previous study [10]. These PCs account k for 73.7% of the variance when performing a wide set of rep- resentative ADL. The loading matrix of the PCs obtained in ∑ ∑ v b p gbp v = , that study, which can be found in Supplementary Materials g b · p (available here), was used to calculate r as shown in equa- ik tions (4) and (5). For a PC , r was computed as the sum k ik v n = , of absolute values of the loadings l for the DoFs j ijk k included in the group i (according to Table 1) divided by the sum of the absolute value of all the loadings of that PC . h =〠 v , 9 g g ∑ l j ijk ∑ n · z g g g r = , k 10 ik a s = k k 2.3. Artificial Hands. With the objective of exemplifying the use of the AIM and verifying its utility, it was computed for a =〠〠 l 5 k ijk i j several artificial hands with different topologies and actua- tion systems. The AIM was obtained for different affordable On the other hand, s in equation (3) contains the 3D-printed prosthetic hands, including the IMMA hand information about the importance of the PC in the most designed by the authors [21], some advanced commercial relevant GTs. To compute this term, first, the human hand prosthetic hands, and other artificial hands. Some hands kinematics was obtained from the human grasp experiment of these two later groups have been evaluated with other explained above, but to consider the relation with the func- indexes of anthropomorphism in the literature, such as tionality of the human hand during ADL, kinematics was the anthropomorphism index (AI) [3] and the Total Score transformed to be expressed as scores f referred to the tk of Anthropomorphism (A ) [4]. The main characteristics five functional PCs (Figure 1) instead of being expressed R of the hands analyzed are described below. in the original sixteen variables (joint angles). This infor- mation can be found in Supplementary Materials. A greater 2.3.1. Affordable 3D-Printed Prosthetic Hands absolute value of the score of a PC in one particular instant t indicates that the position of the hand is better (i) IMMA hand [21]: 3D-printed five-digit prosthetic represented by this PC . Next, for each of the twelve grasp- hand, with 6 DoFs actuated by tendons: flexion in ing tasks g (Figure 2), the absolute value of the scores f tk each finger and flexion and abduction in the thumb. for each PC was averaged during the task (equation (6)), It has three phalanges per finger and its joints are and then (equation (7)) these means v were averaged gbp elastic elements. This hand is just a prototype and across subjects b and repetitions p. The resulting means cannot be used as a prosthesis directly, it needs a v were normalized (equation (8)) with respect to their g socket with motors and a control system, but after sum across PCs h (equation (9)), providing the relative a study of the authors [22], it is being considered here to be actuated by two motors following the contribution of the five PCs to each grasping task n . two actuation synergies obtained from experiments Finally, these relative contributions were weight-averaged with human actuation by the relative relevance of the GT for autonomy of each grasping task z and divided by 3 because three objects (ii) Cyborg Beast [23]: five-digit low-cost 3D-printed were considered for each GT (equation (10)). The weight prosthetic hand for children with upper-limb differ- z for averaging was obtained from the relative use of the ences. It is body-powered using the wrist of the four main GT for a nondominant hand in bimanual grasp- amputee as the unique actuator to drive all the finger 6 Applied Bionics and Biomechanics lates the ring and little finger using a metal bar tendons. It has two phalanges per finger and 5 DoFs: flexion in each finger and flexion of the thumb. Finger within the glove, which couples these fingers to the flexion is driven by tendons along the palmar surface movements of the middle finger of each finger. Elastic cords placed inside the dorsal (iv) Michelangelo hand: five-digit technologically aspect of the fingers provide passive finger extension. advanced prosthetic hand of Ottobock [28]. Actively Its joints are Chicago screws and the materials used to driven elements are the thumb, index finger, and print the different parts of the hand are PLA and ABS middle finger while the ring finger and little finger (iii) Flexy-Hand [24]: 3D-printed five-digit prosthetic passively follow the other fingers. The six joints are hand, with 5 DoFs actuated by tendons: flexion in controlled by two actuators (one for the flexion/- each finger and flexion in the thumb. It has three extension of the five digits and the second one allows phalanges per finger and two phalanges in the the thumb to be electronically positioned in an addi- thumb. The retraction is made through flexible tional axis of movement being abducted/adducted). 3D-printed joints. It is body-powered using the wrist The fingers are slightly abducted when the MCP of the amputee as the unique actuator to drive all the joints are extended, and when flexed, the fingertips finger tendons adduct and touch each other, providing a finger abduction/adduction mode (iv) KIT prosthetic hand [25]: five-digit 3D-printed hand prosthesis with underactuated mechanism, 2.3.3. Other Artificial Hands sensors, and embedded control system, developed by the Institute for Anthropomatics and Robotics (i) FRH-4 hand: this is a robot hand built for the (Karlsruhe Institute of Technology). Two motors mobile-assisting robot ARMAR [29]. It has eight (one for the four long fingers and other for the independent fluidic actuators: one in the metacarpus thumb) actuate 10 DoFs (flexion of two joints in each that allows the palm to flex in the middle, the index finger) by means of tendons. The four long fingers and middle fingers have two each, the ring and little are simultaneously driven via a force-distributing fingers have one for both, and the thumb has two transmission based on the TUAT/Karlsruhe mecha- actuators nism providing shape adaptivity (all fingers keep (ii) Barrett [30]: three-digit programmable grasper of closing until contact regardless of blocked movement Robotnik. It has four brushless motors and three in other joints). The passive reopening of the fingers multijointed fingers (two phalanges connected by is obtained through custom made springs belt transmission), two of them have an extra DoF (v) ADA [26]: Ada Hand V1.1 by Open Bionics is with 180 of lateral mobility a five-digit myoelectric prosthetic hand entirely (iii) DLR/HIT II [31]: used on Space Justin (humanoid 3D-printed with flexible material. It is tendon driven upper body) for telemanipulation is a multisensory and has two phalanges in each finger and one linear robot hand from Harbin Institute of Technology actuator for each digit driving their flexion and DLR Institute for Robotics and Mechatronic. It has 15 DoFs, five identical modular fingers with two 2.3.2. Commercial Prosthetic Hands flat BLDC motors placed in the base. Each finger has three active DoFs (2 DoFs offlexion and one of abduc- (i) i-Limb: myoelectronically controlled, externally tion) and four joints (the motions of distal and medial powered, tendon linking, multiarticulating pros- phalanges are coupled by a linkage). The thumb is thetic hand of Touch Bionics [27] with eleven joints fixed in an appropriate orientation of the palm (two joints in each long finger and three in the thumb). It has five individually powered digits and (iv) Shadow [32]: the Shadow Dexterous Hand is a powered thumb rotation, with manual override humanoid robot hand created by the Shadow Robot Company. The four fingers have 2 one-axis joints (ii) Bebionic [28]: multiarticulating myoelectric pros- (DIP and PIP) and one universal MCP joint; the lit- thetic hand developed by RSL Steeper with eleven tle finger has an extra one-axis joint on the metacar- joints (two joints in each long finger and three in the pal to provide arching. The thumb has a one-axis thumb). It has five actuators, one for each finger, joint (IP) and two universal joints (MCP and and the thumb has two positions manually placed by CMC). It contains 20 motors in the forearm (3 DoFs the user with an inbuilt sensor detecting the position. per finger, 5 DoFs in the thumb, 1 DoF in the palm, Folding links allow the fingers to flex. Among the 14 and 2 DoFs in the wrist) different grip patterns and hand positions that it can achieve, one of them is the finger adduction grip 3. Results and Discussion (iii) SensorHand Speed: the Ottobock SensorHand Speed [28] is a myoelectronically controlled hand with Figure 3 shows a representative plot of the mean value and three actuated fingers, which are driven by the same standard deviation across all the subjects (20) and repetitions motor. It is covered with a cosmetic glove that emu- (3) of the 16 joint angles measured with the CyberGlove Applied Bionics and Biomechanics 7 CMC1 CMC1 MCP1 IP1 F A F F 100 100 100 100 0 00 00 00 −100 −100 −100 −100 −100 −100 −100 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 M MCP2 CP2 M MC CP3 P3 M MC CP4 P4 MC MCP5 P5 F F F F F F F F 100 100 100 100 100 100 100 0 00 00 00 −100 −100 −100 −100 −100 −100 −100 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 PIP2 PIP2 PIP3 PIP3 PIP4 PIP4 PIP5 PIP5 F F F F F F F F 100 100 100 100 100 100 100 0 00 00 00 −100 −100 −100 −100 −100 −100 −100 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 M MCP CP22 M MCP4 CP4 M MCP5 CP5 Pa Palm lmaarrAr Arcch h A A A A A A 100 100 100 100 100 100 100 0 00 00 00 −100 −100 −100 −100 0 0.5 1 0 0.5 1 0 0.5 1 0 0.5 1 Figure 3: Mean and standard deviation of the 16 joint angles (in degrees) measured with the CyberGlove during the extension grip of the plate (T ). The angles are normalized across time (1: thumb, 2: index finger, 3: middle finger, 4: ring finger, and 5: little finger; CMC: carpometacarpal joint, MCP: metacarpophalangeal joint, PIP: proximal interphalangeal joint, DIP: distal interphalangeal joint, IP: interphalangeal joint; F: flexion/extension; and A: abduction/adduction). while performing a grasp of one object of the human grasp “Digit arching” (PC1) is the most significant synergy, as experiment (T of Figure 2). indicated by the higher value of s . 02 k Table 3 shows the relative contribution of the five PCs to Table 4 shows the matrix r (equation (4)) containing the ik each grasping task of the human grasp experiment n influence of the different groups of DoFs of the human hand (equation (8)) together with the final value of the parameter on the five kinematic functional synergies (Figure 1) and the s for each PC (equation (10)), obtained from these relative resulting parameter w after applying equation (3) with this k i contributions weight-averaged by their importance in ADL matrix r and the vector s (Table 3). ik k z [9]. The parameter w weights the relevance of the different g i The contributions of the different PCs to each grasping groups of DoFs in the human grasps in ADL. These results task ranged between 9.6% and 39.1%, indicating that all the indicate that finger flexion-extension is by far the most rele- five PCs have a nonnegligible importance in the twelve grasp- vant group of DoFs accounting for more than half of the ing tasks analyzed. It can be seen that PC1, corresponding to functionality, followed by thumb opposition and finger “digit arching,” is predominant in grasping tasks involving abduction-adduction. The palmar arching has a relevance PP (g = 4,8,12 ). Moreover, “palmar arching” (PC3) and of only 5%. This result by itself is valuable for making deci- “lateral pinch” (PC4) synergies are less represented in aver- sions during the design of new artificial hand prototypes to age in the four main GTs considered in the experiment, lead- maximize their functionality. ing to lower values of the final parameter s for these PCs, Table 5 shows the value of the parameter k (equation (2)) k i although the difference with “opposition” (PC5) is small. for each group of DoFs for the artificial hands analyzed. The 8 Applied Bionics and Biomechanics Table 3: Mean (SD) of the relative contribution n of the five PCs in each grasping task T (g: indicates the order of the tasks in Figure 2) and g g final value of the parameter s for each PC. PC1 (%) PC2 (%) PC3 (%) PC4 (%) PC5 (%) 19.2 (6.4) 24.4 (7.6) 18.4 (7.7) 13.0 (7.7) 24.9 (9.7) T 22.6 (6.9) 14.2 (6.3) 25.1 (6.1) 22.4 (8.1) 15.7 (8.4) 25.1 (4.7) 23.1 (4.7) 15.6 (4.3) 24.9 (8.9) 11.3 (5.5) T 33.5 (6.7) 19.1 (3.8) 12.6 (3.5) 12.1 (4.2) 22.7 (10.6) 22.4 (8.3) 21.4 (8.9) 18.6 (8.5) 17.4 (10.6) 20.1 (13.9) 28.2 (6.0) 21.0 (5.0) 17.9 (5.4) 17.0 (9.5) 16.0 (10.2) 28.3 (5.0) 21.1 (3.8) 12.9 (4.3) 27.0 (9.2) 10.8 (5.8) 39.1 (6.2) 23.6 (4.0) 12.9 (4.6) 11.8 (7.7) 12.5 (7.3) 19.5 (8.2) 24.5 (7.1) 14.3 (7.9) 20.3 (10.1) 21.3 (9.0) 30.9 (6.2) 18.6 (5.2) 20.1 (6.2) 9.6 (5.4) 20.7 (11.7) T 19.9 (3.4) 19.8 (7.7) 19.2 (7.9) 18.3 (10.8) 22.8 (13.3) 34.0 (7.4) 20.4 (5.1) 12.8 (4.3) 12.1 (6.5) 20.8 (9.6) s 29.0 20.5 16.5 15.6 18.4 Table 4: Matrix r and resulting w (equation (3)). ik i Functional synergies Groups of DoFs w (%) PC1 (%) PC2 (%) PC3 (%) PC4 (%) PC5 (%) Finger flexion-extension 79.6 50.7 42.5 51.8 34.9 55 Finger abduction-adduction 8.8 37.8 6.9 20.7 5.4 16 Palmar arching 4.4 3.4 15.3 0.2 1.9 5 Thumb opposition 7.2 8.1 35.3 27.3 57.9 24 Table 5: Parameter k (equation (2)) for each group of DoFs for the Table 6: Results of the Anthropomorphism Index of Mobility different artificial hands. (AIM) for different artificial hands and comparison with other indexes of the literature. Artificial hand F/E AB/AD P.ARC T.OPP Artificial hand AIM (%) AI (%) [3] A (%) [4] IMMA 0.48 0.25 0 0.50 IMMA 42 Cyborg Beast 0.29 0 0 0.18 Cyborg Beast 20 Flexy-Hand 0.39 0.25 0 0.23 Flexy-Hand 31 KIT 0.47 0 0 0.35 KIT 34 ADA 0.58 0.25 0 0.40 ADA 46 i-Limb 0.58 0 0 0.55 i-Limb 45 Bebionic 0.50 0 0 0.50 Bebionic 40 SensorHand 0.13 0 0 0.10 SensorHand 10 0.25 Michelangelo 0.13 0 0 0.30 Michelangelo 14 2.80 FRH-4 0.46 0 0.50 0.40 FRH-4 37 5.20 Barrett 0.25 0.38 0 0.30 Barret 27 10.38 DLR/HIT II 0.83 1 0 0.70 DLR/HIT II 78 26.61 Shadow 0.83 1 0.50 1 Shadow 88 39.93 F/E: finger flexion-extension, AB/AD: finger abduction-adduction, P.ARC: palmar arching, T.OPP: thumb opposition. details about the computation for each hand (c ) are supplied by the parameter w (Table 4). Notwithstanding, some hands ij i as Supplementary Materials. It can be seen that F/E and as the SensorHand and Michelangelo showed low scores in T.OPP are the groups of DoFs mainly included and actively F/E because of their rigid fingers without interphalangeal driven in the artificial hands, manifested by higher values joints. The unique hand with the 5 DoFs in T.OPP actively of k . It is worth to note that this fact is coherent with the driven is the Shadow hand. AB/AD is included actively in greater relevance of these groups of DoFs in ADL, as indicated DLR/HIT II and Shadow hands and passively through the Applied Bionics and Biomechanics 9 Table 7: Range of motion of the hand joints (in degrees) obtained in the human grasp experiment. ° ° ° ° ° Thumb ( ) Index ( ) Middle ( ) Ring ( ) Little ( ) F A F F F A FFFF A FF A F CMC CMC MCP IP MCP MCP PIP MCP PIP MCP MCP PIP MCP MCP PIP Min -27 0 -24 -32 -22 -9 0 -16 0 -13 -7 -1 -13 -7 -2 Max 32 28 13 42 51 24 62 65 66 68 16 76 69 12 68 P5 -6 0 -10 -5 -4 -3 1 -1 1 -1 -1 1 -3 -1 0 P95 15 19 2 17 30 9 36 40 42 30 8 49 26 7 40 CMC: carpometacarpal joint, MCP: metacarpophalangeal joint, PIP: proximal interphalangeal joint, DIP: distal interphalangeal joint, IP: interphalangeal joint, F: flexion (+)/extension (-), A: abduction (+)/adduction (-), P: percentile. use of deformable joints in some 3D-printed hands. Finally, the AIM as a method for evaluating the anthropomorphism P.ARC is only present in FRH-4 and Shadow hands. of an artificial hand. Finally, Table 6 shows the AIM for the different artificial (i) It is really quick to obtain: simply, the parameter k hands, obtained using equation (1) and considering the parameters shown in Tables 4 and 5. Two factors affect the has to be calculated, according to the DoFs and actu- ation methods of the artificial hand and equation (1) final AIM obtained by a hand (equation (1)): its mobility has to be applied (w is provided above) and type of actuation, represented by the number of DoF, the number of actuators, the number of digits and phalanges (ii) It analyzes not only the topology but also the func- per digit, and the type of underactuation, affecting to the final tionality of the artificial hand because it takes into parameters k ; and how this mobility and actuation system is account the results obtained in grasping tests and distributed among the different groups of DOFs, with regard ADL with the human hand to the human hand, affecting through the weighting factor w (Table 4). The most advanced robotic hands (DLR/HIT II and Shadow) with a significant amount of motors and DoFs, Notwithstanding, some important aspects in the design and located in the important groups of DoFs, with higher of an artificial hand are not within the scope of the AIM: weight w (F/E, T.OPP, and AB/AD), obtained the highest the orientation of the joint axes, the range of motion of the AIM scores, above 75%. The commercial prosthetic hands different hand joints, the dimension of the phalanges, the i-Limb and Bebionic as well as some 3D-printed hands friction coefficient of the parts of the hand in contact with (ADA, IMMA) obtained AIM scores between 40% and the objects, the grasping force exerted by the actuators, the 50%. These hands include a reasonable number of motors efficiency of the driving linkages, the control system, etc. and DoFs in the important groups (F/E and T.OPP). The rest Some previous studies [33–35] have shown the relevance of of the hands obtained scores below 40% with the lowest score these aspects. In this sense, the AIM, involving mainly the being for the SensorHand. The reason behind this lower AIM topological structure, the number of actuators, and the type is an improvable number of DoFs, motors, or type of under- of underactuation, can be considered as an index especially actuation in the groups of F/E, T.OPP, or both. useful in the concept design stage. The other design consider- The results shown in Table 6 indicate that the artificial ations cited above should be taken into account in later hands analyzed in the literature with other anthropomor- design stages: preliminary or detail design. Additional phism indexes, such as AI [3] or A [4], are ranked equally indexes that take into account these aspects could be interest- by the AIM and the other metrics, although the scores are ing, and future works can go in this way. The index proposed different. The method used to compute the indexes justify by Liu et al. [5] considers some of these aspects, but it does these different scores. The AI is obtained from the achievable not include their relevance for functionality according to workspace of positions and orientations of the fingers’ distal human grasping tests. With respect to the phalanx dimen- segments and compares this with information obtained sions and the joints’ range of motion, the authors developed experimentally from human hand grasping. The A is based some studies [13, 16, 36] helping to obtain anthropomorphic on the computation of the finger phalanx workspace com- designs. However, the evaluation of some of the design bined with that of the finger base frames, and the compari- aspects cited above is difficult to be performed with indexes, son with the human hand is made through a simplified requiring experimentation, after detailed design of the artifi- model of their joints and geometry. It is worth to note that cial hand and manufacturing a prototype. The authors have obtaining AI and A involves using complex algorithms proposed methods for this experimental evaluation consider- and detailed information of the hand design, not easily avail- ing the main GTs in ADL and a special device for actuating able, while obtaining the AIM just requires information the hand prototype [21]. about the number of DoFs and the possibility to control The ranges of motion of the hand joints obtained in the them independently. Despite these differences in the human grasp experiment undertaken in this study are shown method used to obtain each index, the fact that they rank in Table 7. A wide range of motion for the different joints equally, the hands as the AIM can be seen as a kind of val- was covered with the objects selected in comparison to the idation of our index. Two main points can justify the use of functional range of motion of the human hand joints in 10 Applied Bionics and Biomechanics ADL [16]. These ranges could be considered as a minimum findings of this study are included within the supplementary for prostheses with functional grasping for the main GTs, information files. although general manipulation would recommend using larger ranges if possible. Conflicts of Interest This study was primarily focused on prosthetic hands, and therefore, the scoring system takes into account the The authors declare that there is no conflict of interest capability of the hand to perform the most important GTs regarding the publication of this paper. for a nondominant hand to reinforce bimanual grasping (through parameter z ). For the case of a dominant hand Acknowledgments reinforcing bimanual grasping, the parameter z for the four GTs considered in this study changes to [9] PP (58.0%), EG This work was supported by the Spanish Ministry of (16.6%), TP (9.5%), and TVG (16.0%). The effect of this Economy and Competitiveness and ESF (grant number change on the resulting w is negligible and implies a dispar- BES-2015-076005); the Spanish Ministry of Economy and ity of the AIM obtained for the artificial hands analyzed Competitiveness, AEI, and ERDF (grant numbers DPI2014- (Table 6) of a maximum of 1%. Therefore, the AIM is con- 60635-R, DPI2017-89910-R); and Universitat Jaume I sidered useful to evaluate the anthropomorphism of both (grant numbers UJI-B2017-70, UJI-B2017-51). dominant and nondominant hands. With this result and the result obtained from the comparison of the AIM with Supplementary Materials other indexes of the literature [3, 4], we can conclude that the index proposed can be valid for artificial both robotic The excel file “JointAngles” contains the joint angles for each and prosthetic hands, regardless of whether they are domi- instant for the twelve tasks repeated three times by twenty nant or nondominant hands. subjects during the human grasp experiment. The excel file “Scores” contains the human hand kinematics in the human grasp experiment transformed to scores f referred to five 4. Conclusion tk functional synergies for the twelve tasks repeated three times In this study, we have presented an anthropomorphism by twenty subjects. The excel file “LoadingsMatrix” contains index (AIM) that can be used to evaluate and compare the the loadings l of the five functional synergies during ADL. ijk mobility of artificial hands in relation to the human hand The excel file “ArtificialHands” contains the value c accord- ij functionality, especially in concept design. The AIM evalu- ing to the method of actuation for each DoF in the artificial ates the topology of the whole hand (joints and DoFs) and hands analyzed and the corresponding k of each hand. the possibility to control these DoFs independently according (Supplementary Materials) to their functionality. We have shown that the index can be valid for both prosthetic and robotic hands, dominant and References nondominant hands. To define the index, the functionality of the different groups of DoFs of the hand (F/E, AB/AD, [1] J. T. Belter, J. L. Segil, A. M. Dollar, and R. F. Weir, “Mechan- P.ARC, and T.OPP) was analyzed according to a human ical design and performance specifications of anthropomor- grasp experiment on twenty subjects with the four main phic prosthetic hands: a review,” Journal of Rehabilitation GTs for personal autonomy in ADL. It was concluded Research and Development, vol. 50, no. 5, pp. 599–618, 2013. that the relevance of the different groups of DoFs (w ) [2] J. ten Kate, G. Smit, and P. Breedveld, “3D-printed upper limb was 55% for F/E, 16% for AB/AD, 5% for P.ARC, and prostheses: a review,” Disability and Rehabilitation. Assistive Technology, vol. 12, no. 3, pp. 300–314, 2017. 24% for T.OPP. Thirteen artificial hands, including affordable 3D-printed prosthetic hands, advanced commercial pros- [3] T. Feix, J. Romero, C. H. Ek, H. B. Schmiedmayer, and D. Kragic, “A metric for comparing the anthropomorphic thetic hands, and robotic hands, were evaluated and motion capability of artificial hands,” IEEE Transactions on compared with the AIM, and the reason for their differences Robotics, vol. 29, no. 1, pp. 82–93, 2013. was discussed. The results obtained in this study should be [4] M. V. Liarokapis, P. K. Artemiadis, and K. J. Kyriakopoulos, taken into account in the concept design stage of new proto- “Quantifying anthropomorphism of robot hands,” in 2013 types in order to obtain new designs that maximize their func- IEEE International Conference on Robotics and Automation, tionality. Further research will focus on new metrics for later pp. 2041–2046, Karlsruhe, Germany, May 2013. design stages considering other design aspects (range of [5] Y. Liu, D. Yang, L. Jiang, and H. Liu, “A synthetic framework motion of the joints, relative length of the phalanges, orienta- for evaluating the anthropomorphic characteristics of pros- tion of the joints axes, etc.) and on experimental benchmarks thetic hands,” in 2015 IEEE International Conference on to measure the grasping capability of artificial hands. Advanced Intelligent Mechatronics (AIM), pp. 877–884, Busan, South Korea, July 2015. Data Availability [6] L. Birglen, T. Laliberté, and C. Gosselin, Underactuated Robotic Hands, Springer Tracts in Advanced Robotics book The human hand kinematics expressed as joint angles and series (STAR, volume 40), Springer, 2008. scores referred to the five PCs, the loading matrix corre- [7] M. W. Cornwall, “Prosthetics and patient management: a sponding to these PCs, and the c values for the joints of ij comprehensive clinical approach,” Physical Therapy, vol. 87, the artificial hands analyzed; the data used to support the no. 4, p. 482, 2007. Applied Bionics and Biomechanics 11 [24] Gyrobot, “Flexy-Hand 2,” 2014, March 2019, https://www. [8] K. E. Yancosek and D. R. Mullineaux, “Stability of handwriting performance following injury-induced hand-dominance thingiverse.com/thing:380665. transfer in adults: a pilot study,” Journal of Rehabilitation [25] P. Weiner, J. Starke, F. Hundhausen, J. Beil, and T. Asfour, Research and Development, vol. 48, no. 1, pp. 59–68, 2011. “The KIT prosthetic hand: design and control,” in 2018 [9] V. Gracia-Ibáñez, J. L. Sancho-Bru, and M. Vergara, “Rele- IEEE/RSJ International Conference on Intelligent Robots and vance of grasp types to assess functionality for personal Systems (IROS), Madrid, Spain, October 2018. autonomy,” Journal of Hand Therapy, vol. 31, no. 1, [26] Open Bionics, “Open Bionics hands,” 2018, March 2019, pp. 102–110, 2018. https://openbionicslabs.com/downloads. [10] V. Gracia Ibañez, Contribution to hand functional assessment [27] Touch Bionics, “Touch Bionics products,” 2019, March 2019, based on its kinematics, Universitat Jaume I, Castelló de la http://touchbionics.com/products. Plana, 2016. [28] Ottobock, “Ottobock myoelectric prosthetics,” 2017, March [11] World Health Organization, Towards a common language for 2019, https://www.ottobockus.com/prosthetics/upper-limb- functioning, disability and health ICF the International Classi- prosthetics/solution-overview/myoelectric-prosthetics/. fication of Functioning, Disability and Health, World Heal. [29] I. Gaiser, S. Schulz, A. Kargov et al., “A new anthropomorphic Organ, Geneva, 2002. robotic hand,” in Humanoids 2008 - 8th IEEE-RAS Interna- [12] I. Llop-Harillo, A. Pérez-González, and V. Gracia-Ibáñez, tional Conference on Humanoid Robots, pp. 418–422, Daejeon, Anthropomorphism index of mobility for hand prostheses, South Korea, December 2008. XXVI Congress of the International Society of Biomechanics, [30] Barrett Technology, “BarrettHand,” March 2019, https://www. barrett.com/about-barretthand. [13] M. Vergara, M. J. Agost, and V. Gracia-Ibáñez, “Dorsal and [31] H. Liu, K. Wu, P. Meusel et al., “Multisensory five-finger palmar aspect dimensions of hand anthropometry for design- dexterous hand: the DLR/HIT hand II,” in 2008 IEEE/RSJ ing hand tools and protections,” Human Factors and International Conference on Intelligent Robots and Systems, Ergonomics in Manufacturing & Service Industries, vol. 28, Nice, France, September 2008. no. 1, pp. 17–28, 2018. [32] Shadow Robot Company, “Shadow dexterous hand,” 2019, [14] B. Calli, A. Walsman, A. Singh, S. Srinivasa, P. Abbeel, and March 2019, https://www.shadowrobot.com/products/ A. M. Dollar, “Benchmarking in manipulation research: dexterous-hand/. using the Yale-CMU-Berkeley object and model set,” IEEE Robotics and Automation Magazine, vol. 22, no. 3, pp. 36– [33] M. Controzzi, M. D'Alonzo, C. Peccia, C. M. Oddo, M. C. Carrozza, and C. Cipriani, “Bioinspired fingertip for anthro- 52, 2015. pomorphic robotic hands,” Applied Bionics and Biomechan- [15] V. Gracia-Ibáñez, M. Vergara, J. H. Buffi, W. M. Murray, and ics, vol. 11, no. 1-2, 38 pages, 2014. J. L. Sancho-Bru, “Across-subject calibration of an instru- mented glove to measure hand movement for clinical pur- [34] M. Ceccarelli, N. E. N. Rodríguez, G. Carbone, and C. Lopez- poses,” Computer Methods in Biomechanics and Biomedical Cajùn, “An optimal design of driving mechanism in a 1 degree of freedom (d.o.f.) anthropomorphic finger,” Applied Bionics Engineering, vol. 20, no. 6, pp. 587–597, 2017. and Biomechanics, vol. 2, no. 2, pp. 103–110, 2005. [16] V. Gracia-Ibáñez, M. Vergara, J. L. Sancho-Bru, M. C. Mora, and C. Piqueras, “Functional range of motion of the hand [35] F. J. Andrés, A. Pérez-González, C. Rubert, J. Fuentes, and joints in activities of the International Classification of Func- B. Sospedra, “Comparison of grasping performance of tendon tioning, Disability and Health,” Journal of Hand Therapy, and linkage transmission systems in an electric-powered low- vol. 30, no. 3, pp. 337–347, 2017. cost hand prosthesis,” Journal of Mechanisms and Robotics, vol. 11, no. 1, article 011018, 2018. [17] V. Patel, M. Burns, R. Chandramouli, and R. Vinjamuri, “Biometrics based on hand synergies and their neural repre- [36] I. Llop-Harillo, V. Gracia-Ibáñez, and A. Pérez-González, sentations,” IEEE Access, vol. 5, pp. 13422–13429, 2017. “Analysis of anthropometric dimensions and joints range of motion of the human hand for application to the design of [18] I. A. Kapandji, The Physiology of the Joints: Upper Limb, hand prostheses,” in 8th World Congress on Biomechanics, Churchill Livingstone, 1982. [19] J. L. Sancho-Bru, Modèl biomecànic de la mà orientat al disseny d’eines manuals, Universitat Jaume I, 2000. [20] D. G. Ullman, The Mechanical Design Process, McGraw-Hill, 4th edition, 2010. [21] I. Llop-Harillo and A. Pérez-González, “System for the exper- imental evaluation of anthropomorphic hands. Application to a new 3D-printed prosthetic hand prototype,” International Biomechanics, vol. 4, no. 2, pp. 50–59, 2017. [22] I. Llop-Harillo, A. Pérez-González, and J. Cantero-Ramis, “Analysis of motion synergies in grasping tasks for a 6-DOF tendon-driven prosthetic hand,” in 16th International Sympo- sium on Computer Methods in Biomechanics and Biomedical Engineering and 4th Conference on Imaging and Visualization, New York, 2019. [23] J. Zuniga, D. Katsavelis, J. Peck et al., “Cyborg beast: a low-cost 3d-printed prosthetic hand for children with upper-limb differences,” BMC Research Notes, vol. 8, no. 1, p. 10, 2015. International Journal of Advances in Rotating Machinery Multimedia Journal of The Scientific Journal of Engineering World Journal Sensors Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Hindawi Hindawi Hindawi Volume 2018 Volume 2018 Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com www.hindawi.com www.hindawi.com Volume 2018 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Bionics and Biomechanics Hindawi Publishing Corporation

Anthropomorphism Index of Mobility for Artificial Hands

Loading next page...
 
/lp/hindawi-publishing-corporation/anthropomorphism-index-of-mobility-for-artificial-hands-AgaF3cYuoS
Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2019 Immaculada Llop-Harillo et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
1176-2322
eISSN
1754-2103
DOI
10.1155/2019/7169034
Publisher site
See Article on Publisher Site

Abstract

Hindawi Applied Bionics and Biomechanics Volume 2019, Article ID 7169034, 11 pages https://doi.org/10.1155/2019/7169034 Research Article Immaculada Llop-Harillo , Antonio Pérez-González , and Verónica Gracia-Ibáñez Grupo de Biomecánica y Ergonomía, Departamento de Ingeniería Mecánica y Construcción, Universitat Jaume I (UJI), 12071, Spain Correspondence should be addressed to Immaculada Llop-Harillo; illop@uji.es Received 21 March 2019; Accepted 17 June 2019; Published 28 July 2019 Guest Editor: Francesca Cordella Copyright © 2019 Immaculada Llop-Harillo et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The increasing development of anthropomorphic artificial hands makes necessary quick metrics that analyze their anthropomorphism. In this study, a human grasp experiment on the most important grasp types was undertaken in order to obtain an Anthropomorphism Index of Mobility (AIM) for artificial hands. The AIM evaluates the topology of the whole hand, joints and degrees of freedom (DoFs), and the possibility to control these DoFs independently. It uses a set of weighting factors, obtained from analysis of human grasping, depending on the relevance of the different groups of DoFs of the hand. The computation of the index is straightforward, making it a useful tool for analyzing new artificial hands in early stages of the design process and for grading human-likeness of existing artificial hands. Thirteen artificial hands, both prosthetic and robotic, were evaluated and compared using the AIM, highlighting the reasons behind their differences. The AIM was also compared with other indexes in the literature with more cumbersome computation, ranking equally different artificial hands. As the index was primarily proposed for prosthetic hands, normally used as nondominant hands in unilateral amputees, the grasp types selected for the human grasp experiment were the most relevant for the human nondominant hand to reinforce bimanual grasping in activities of daily living. However, it was shown that the effect of using the grasping information from the dominant hand is small, indicating that the index is also valid for evaluating the artificial hand as dominant and so being valid for bilateral amputees or robotic hands. freedom (DoFs), range of motion, and weight and number 1. Introduction of actuators, but an index to compare those properties with In recent years, it has been an increasing development of new the human hand was not defined. Some other previous stud- affordable and anthropomorphic prosthetic hands [1, 2] as a ies tried to quantify the anthropomorphism of artificial consequence of the improvements in 3D-printing technolo- hands with a numerical index. Feix et al. [3] proposed a met- ric for comparing the anthropomorphic motion capability of gies. The human hand is a complex and marvelous tool whose dexterity has not been achieved by any artificial hand. robotic and prosthetic hands, the anthropomorphism index Evaluating the functional similarity of artificial hands with (AI), being its computation cumbersome and based only on the human hand is essential for improving current anthropo- the position and orientation of the distal phalanges in differ- morphic hand designs. Assessing the capability of the pros- ent GTs. Liarokapis et al. [4] defined an anthropomorphism theses to perform the main grasp types (GTs) of human index to assess the robot’s ability to mimic the human hand grasping could give an insight into the level of functionality based on the comparison of the finger phalanx workspaces restored in patients. Metrics or indexes that quantify numer- and also the workspaces of the fingers’ base frames. Liu ically the level of anthropomorphism are the way to grade et al. [5] proposed twelve quantified prosthetic hand anthro- human-likeness and to provide specifications for maximizing pomorphism evaluation indexes including physical and the anthropomorphic functionality while designing new actuation properties, among which is included a DoF config- artificial hands. uration evaluation. This index was based on a matrix of DoF Belter et al. [1] reviewed and compared the mechanical configuration where the element of the matrix is set to 1 if properties of different prosthetic hands, as their degrees of there exist an artificial DoF in the corresponding position, 2 Applied Bionics and Biomechanics PC1 PC2 PC3 PC4 PC5 tests developed in this study on the main GTs. Furthermore, a preliminary study [12] carried out on four human healthy subjects encouraged us to go deep in the study by increasing the number of subjects, improving the definition of the index, and widening the analysis of its validity to the different types of artificial hands. The AIM is intended to be a quick compu- tation index based on the biomechanics of the human hand and thus providing a way to compare their functional anthro- pomorphism. Moreover, the relevance of each DoF for func- tionality, obtained by tests on the human hand in this study, is intended to be useful for other applications in artificial Figure 1: Principal components of the degrees of freedom of the human hand performing activities of daily living obtained in [10] hand design. (PC1: digit arching, PC2: closure, PC3: palmar arching, PC4: lateral pinch, and PC5: opposition). 2. Materials and Methods otherwise is set to 0. However, this approach does not take into account the relevance of each DoF for grasping during 2.1. Human Grasp Experiment. With the purpose of taking activities of daily living (ADL) nor the underactuation in into account in the AIM the relevance of each DoF according the joints. Underactuation in artificial hands [6] allows to to its importance for functional grasping, an experiment to use less actuators than DoFs while keeping versatility to measure the kinematics of the human hand in functional adapt GTs to different object shapes. grasps was carried out. Twenty subjects, ten males and ten Prostheses design could be different depending on its use females, all of whom were right-handed and free of hand for a dominant or nondominant hand; however, in the case pathologies or injuries, performed the most relevant GTs of a patient who still has a healthy hand, the most appropriate for a nondominant hand to reinforce bimanual grasping in strategy would be to consider the remaining hand as domi- ADL (PP, EG, TP, and TVG [9]). Although the grasps were nant [7, 8]. Thereby, the design of the prosthesis should be selected for a nondominant hand (most common use of a focused for a nondominant hand reinforcing bimanual hand prosthesis for unilateral amputees), subjects were asked grasping. The importance of the different GTs for personal to perform grasps with their dominant hand to get the most autonomy of the patients in ADL has been studied previously natural performance of human grasping. The study was by the authors [9], being pulp pinch (PP) (26%), extension approved by the Ethics Committee of the University, and grip (EG) (20.8%), tripod pinch (TP) (10.4%), and transverse all the subjects gave their written informed consent. The ages volar grip (TVG) (8.7%), the most relevant GTs for a non- of the subjects ranged intentionally between 20 and 51, being dominant hand to reinforce bimanual grasping, representing the average 35 ± 8, in order to prevent kinematic alterations together with the nonprehensile one, almost 90% of relevance due to joint degeneration from ageing. Subjects were selected for autonomy. so that the distribution of hand sizes was representative of the In the previous studies by the authors [10], the posture of population [13]. The hand width ranged from 70 to 96 mm the right hand from 24 healthy subjects performing 24 repre- with an average of 83 mm, and the hand length ranged from sentative ADL was recorded with an instrumented glove. 170 to 210 mm with an average of 185 mm. Twelve objects of different sizes were selected from the ADL were selected from the WHO’s International Classifica- tion of Functioning, Disability and Health [11]. By applying Yale-CMU-Berkeley Object and Model Set [14], three for principal component analysis (PCA), five factors explaining each of the four GTs (PP, EG, TP, and TVG), in order to 73.7% of the variance were obtained. As shown in Figure 1, cover most common requirements in ADL for each one the five main principal components (PCs) of the DoFs of (Figure 2). The subjects were sitting with the hands in the the human hand in ADL were “PC1: digit arching” (flexion table in a comfortable way: the arms close to the body and of the interphalangeal joints), “PC2: closure” (combination parallel to the sagittal plane, the elbows flexed 90 , the wrist of abduction of the fingers, except for the thumb, with flexion on the edge of the table, and the hands laying on the table of the metacarpophalangeal joints), “PC3: palmar arching,” palms down in a natural posture. This was the starting and “PC4: lateral pinch” (represents the lateral opposition of the ending posture for each grasping action. Subjects were thumb to the index), and “PC5: opposition” (represents the instructed on the different GTs to perform with each object, pad-to-pad opposition of the thumb to the little finger). and objects to be grasped were situated one by one by the The aim of this study is to propose an index to measure researcher at a distance of 30 cm in front of the subjects. Sub- the anthropomorphism of prosthetic hands, based on the jects were free to practice the grasps to be sure that it is in the comparison of the topology of the whole hand (joints and correct posture before starting the recordings. The steps to DoFs) and on the possibility to control these DoFs indepen- perform the grasps during the experiment consisted of the dently. The computation of the index, referred to as Anthro- following: grasping the object from the table with the correct pomorphism Index of Mobility (AIM), should weight each hand posture/GT, lift it up during two seconds, and finally, DoF depending on its importance for grasping in ADL. To release the object again on the table and return the hand to define this importance, we used the information from previ- the starting position. The sequence of the twelve objects to ous experimental tests performed in the group and specific grasp during the experiment is shown in Figure 2 in the Applied Bionics and Biomechanics 3 (a) T large marker (TP) (b) T plate (EG) (c) T chips can (TVG) 01 02 03 (d) T small marker (PP) (e) T tuna can (TP) (f) T cracker box (EG) 04 05 06 (g) T coffee can (TVG) (h) T plastic pear (PP) (i) T golf ball (TP) 07 08 09 (j) T pudding box (EG) (k) T power drill (TVG) (l) T washer 10 mm (PP) 10 11 12 Figure 2: Grasping tasks of the experiment (a-l). T are the tasks ordered (g: indicates the order) followed by the object of the Yale-CMU- Berkeley Object and Model Set [14] to grasp and in brackets the grasp type to be performed in each task (TP: tripod pinch, EG: extension grip, TVG: transverse volar grip, and PP: pulp pinch). specified order. The experiment was repeated three times geal flexion (MCP1 to MCP5, 1 to 5 meaning thumb to little per subject. digits), interphalangeal flexion of the thumb (IP1), proximal The kinematics of the hand while performing the grasp- interphalangeal flexion of the fingers (PIP2 to PIP5), flex- ing postures was recorded (100 Hz) using an instrumented ion and abduction of the carpometacarpal joint of the right hand glove with 18 sensors (CyberGlove Systems LLC; thumb (CMC1), relative abduction between finger MCPs San Jose, CA). DoF kinematics corresponding to 16 joint (index-middle, middle-ring, and ring-little), and palmar angles (marked with an asterisk in Table 1) was obtained arching. Prior to the tests with objects, the CyberGlove was using a previously validated protocol [15]: metacarpophalan- calibrated for each subject following the calibration 4 Applied Bionics and Biomechanics Table 1: Joints and degrees of freedom (DoFs) of the human hand where the summation extends for i = 1,2,3,4, corresponding corresponding to the four different groups of DoFs defined. to each one of the four groups of DoFs (Table 1: F/E, AB/AD, P.ARC, and T.OPP), the factor k accounts for the type of Groups of DoFs Joints and DoFs of the human hand actuation of the DoFs included in this group, and the factor MCP2_Flexion w is a weighting coefficient depending on the relevance of this group of DoF for grasping in ADL. Both the term k PIP2_Flexion and the weighting factor w were defined to have a range DIP2_Flexion between 0 and 1, and the sum of weighting coefficients w ∗ i MCP3_Flexion for the four groups is unity, so that the AIM reach a maxi- PIP3_Flexion mum value of 1 for the human hand and a very low value DIP3_Flexion for an artificial hand with very poor anthropomorphism. Finger flexion-extension The factor k for each group i was defined to get a high (F/E) MCP4_Flexion value if the method of actuation for the DoFs in that group PIP4_Flexion allows to control them independently, as in the human hand, DIP4_Flexion and a lower value if the motions of these DoFs are highly MCP5_Flexion coupled during motion. To this end, each DoF in the evalu- ated hand was classified according to the types included in PIP5_Flexion Table 2. DIP5_Flexion The independent mobility of a DoF can be ranked quali- MCP2_Abduction tatively from better to worse, depending on its class, as MCP3_Abduction A>B>C>D>E. Note that B class was considered better than Finger abduction-adduction (AB/AD) C because it allows mechanical adaptation of the finger to MCP4_Abduction ∗ the shape of the object to be grasped and do not suffer from MCP5_Abduction mechanical singular configurations [6]. Pugh’s method used CMC5_Flexion Palmar arching in concept design evaluation [20] was employed to convert (P.ARC) CMC4_Flexion the ranked list of methods of actuation of the DoFs into a list of numerical coefficients c (last column in Table 2). However, CMC1_Flexion the independent mobility of a DoF is associated not only with CMC1_Abduction the type of actuation in this particular DoF but also with that Thumb opposition MCP1_Flexion of the DoFs more proximal in the same serial chain of a digit, (T.OPP) MCP1_Abduction i.e., for a finger, the mobility for flexion in the PIP joint is dependent on the mobility for flexion in the MCP joint. IP1_Flexion Consequently, for that case, the coefficient c for the DoF j ij 1: thumb, 2: index finger, 3: middle finger, 4: ring finger, 5: little finger; CMC: of the group i was obtained as the multiplication of the coef- carpometacarpal joint, MCP: metacarpophalangeal joint, PIP: proximal ficient c of this DoF and those located proximally in the same interphalangeal joint, DIP: distal interphalangeal joint, IP: interphalangeal joint; 16 joint angles measured during the experiment with the CyberGlove. serial kinematic chain. In addition, for assigning the coeffi- cient c to several DoFs underactuated by the same motor or procedure [15]. Starting and final positions while the hand is actuator, class A was considered for only one of them and not moving were trimmed from the recordings. Then, they class B or C for others. If a motor actuates several DoFs were filtered with a 2nd-order 2-way low-pass Butterworth included in different groups i, the coefficient 1 corresponding filter with cut-off frequency of 5 Hz [16, 17]. The tests were to class A was divided among the number of groups and this video recorded. fraction was assigned to only one of the DoFs in this group, being others classified as either B or C. Finally, the factor k 2.2. Index Definition. The Anthropomorphism Index of for each group i was defined with equation (2), by summing Mobility (AIM) for an artificial hand was defined based on the terms c in the group i and dividing by the number of ij two main factors: (1) the DoFs present in the hand along with DoFs of the human hand in this group (n ), which is, its method of actuation and (2) the relevance of these DoFs according to Table 1, 12 for i =1, 4 for i =2, 2 for i =3, for grasping in ADL. and 5 for i =4. The DoFs of the human hand (HH) [18, 19] were classi- fied into four different functional groups for defining the ∑ c j ij AIM (Table 1): finger flexion-extension (12 in HH), finger k = 2 abduction-adduction (4 in HH), palmar arching (2 in HH), and thumb opposition (5 in HH). The Anthropomorphism Index of Mobility (AIM) was The weighting factor w in equation (1), accounting for defined with the relative relevance of the DoFs of the group i for grasping in ADL, was defined with AIM =〠 k · w , 1 i i w =〠 r · s 3 i ik k k Applied Bionics and Biomechanics 5 Table 2: Classification of the DoF depending on the type of actuation and numeric coefficient associated. Class Type of actuation of the DoF c A DoF actuated by one independent motor or actuator 1 B DoF underactuated with other DoFs without a rigid coupling, allowing adaptive grasps (tendons, elastic elements) 0.75 C DoF underactuated with other DoFs with a rigid coupling, not allowing adaptive grasp (linkages) 0.5 D No actuation on the DoF, but passive motion allowed 0.25 E DoF absent in the artificial hand 0 In equation (3), r weights the relative contribution of ing [9]: 39.5% for PP, 31.6% for EG, 15.8% for TP, and ik the group of DoFs i (i = 1,2,3,4) in human hand functionality 13.2% for TVG. represented through PC (k = 1,2,3,4,5), corresponding to each of the five kinematic functional synergies (see ∑ f t=1 tk v = , gbp Figure 1) found in a previous study [10]. These PCs account k for 73.7% of the variance when performing a wide set of rep- resentative ADL. The loading matrix of the PCs obtained in ∑ ∑ v b p gbp v = , that study, which can be found in Supplementary Materials g b · p (available here), was used to calculate r as shown in equa- ik tions (4) and (5). For a PC , r was computed as the sum k ik v n = , of absolute values of the loadings l for the DoFs j ijk k included in the group i (according to Table 1) divided by the sum of the absolute value of all the loadings of that PC . h =〠 v , 9 g g ∑ l j ijk ∑ n · z g g g r = , k 10 ik a s = k k 2.3. Artificial Hands. With the objective of exemplifying the use of the AIM and verifying its utility, it was computed for a =〠〠 l 5 k ijk i j several artificial hands with different topologies and actua- tion systems. The AIM was obtained for different affordable On the other hand, s in equation (3) contains the 3D-printed prosthetic hands, including the IMMA hand information about the importance of the PC in the most designed by the authors [21], some advanced commercial relevant GTs. To compute this term, first, the human hand prosthetic hands, and other artificial hands. Some hands kinematics was obtained from the human grasp experiment of these two later groups have been evaluated with other explained above, but to consider the relation with the func- indexes of anthropomorphism in the literature, such as tionality of the human hand during ADL, kinematics was the anthropomorphism index (AI) [3] and the Total Score transformed to be expressed as scores f referred to the tk of Anthropomorphism (A ) [4]. The main characteristics five functional PCs (Figure 1) instead of being expressed R of the hands analyzed are described below. in the original sixteen variables (joint angles). This infor- mation can be found in Supplementary Materials. A greater 2.3.1. Affordable 3D-Printed Prosthetic Hands absolute value of the score of a PC in one particular instant t indicates that the position of the hand is better (i) IMMA hand [21]: 3D-printed five-digit prosthetic represented by this PC . Next, for each of the twelve grasp- hand, with 6 DoFs actuated by tendons: flexion in ing tasks g (Figure 2), the absolute value of the scores f tk each finger and flexion and abduction in the thumb. for each PC was averaged during the task (equation (6)), It has three phalanges per finger and its joints are and then (equation (7)) these means v were averaged gbp elastic elements. This hand is just a prototype and across subjects b and repetitions p. The resulting means cannot be used as a prosthesis directly, it needs a v were normalized (equation (8)) with respect to their g socket with motors and a control system, but after sum across PCs h (equation (9)), providing the relative a study of the authors [22], it is being considered here to be actuated by two motors following the contribution of the five PCs to each grasping task n . two actuation synergies obtained from experiments Finally, these relative contributions were weight-averaged with human actuation by the relative relevance of the GT for autonomy of each grasping task z and divided by 3 because three objects (ii) Cyborg Beast [23]: five-digit low-cost 3D-printed were considered for each GT (equation (10)). The weight prosthetic hand for children with upper-limb differ- z for averaging was obtained from the relative use of the ences. It is body-powered using the wrist of the four main GT for a nondominant hand in bimanual grasp- amputee as the unique actuator to drive all the finger 6 Applied Bionics and Biomechanics lates the ring and little finger using a metal bar tendons. It has two phalanges per finger and 5 DoFs: flexion in each finger and flexion of the thumb. Finger within the glove, which couples these fingers to the flexion is driven by tendons along the palmar surface movements of the middle finger of each finger. Elastic cords placed inside the dorsal (iv) Michelangelo hand: five-digit technologically aspect of the fingers provide passive finger extension. advanced prosthetic hand of Ottobock [28]. Actively Its joints are Chicago screws and the materials used to driven elements are the thumb, index finger, and print the different parts of the hand are PLA and ABS middle finger while the ring finger and little finger (iii) Flexy-Hand [24]: 3D-printed five-digit prosthetic passively follow the other fingers. The six joints are hand, with 5 DoFs actuated by tendons: flexion in controlled by two actuators (one for the flexion/- each finger and flexion in the thumb. It has three extension of the five digits and the second one allows phalanges per finger and two phalanges in the the thumb to be electronically positioned in an addi- thumb. The retraction is made through flexible tional axis of movement being abducted/adducted). 3D-printed joints. It is body-powered using the wrist The fingers are slightly abducted when the MCP of the amputee as the unique actuator to drive all the joints are extended, and when flexed, the fingertips finger tendons adduct and touch each other, providing a finger abduction/adduction mode (iv) KIT prosthetic hand [25]: five-digit 3D-printed hand prosthesis with underactuated mechanism, 2.3.3. Other Artificial Hands sensors, and embedded control system, developed by the Institute for Anthropomatics and Robotics (i) FRH-4 hand: this is a robot hand built for the (Karlsruhe Institute of Technology). Two motors mobile-assisting robot ARMAR [29]. It has eight (one for the four long fingers and other for the independent fluidic actuators: one in the metacarpus thumb) actuate 10 DoFs (flexion of two joints in each that allows the palm to flex in the middle, the index finger) by means of tendons. The four long fingers and middle fingers have two each, the ring and little are simultaneously driven via a force-distributing fingers have one for both, and the thumb has two transmission based on the TUAT/Karlsruhe mecha- actuators nism providing shape adaptivity (all fingers keep (ii) Barrett [30]: three-digit programmable grasper of closing until contact regardless of blocked movement Robotnik. It has four brushless motors and three in other joints). The passive reopening of the fingers multijointed fingers (two phalanges connected by is obtained through custom made springs belt transmission), two of them have an extra DoF (v) ADA [26]: Ada Hand V1.1 by Open Bionics is with 180 of lateral mobility a five-digit myoelectric prosthetic hand entirely (iii) DLR/HIT II [31]: used on Space Justin (humanoid 3D-printed with flexible material. It is tendon driven upper body) for telemanipulation is a multisensory and has two phalanges in each finger and one linear robot hand from Harbin Institute of Technology actuator for each digit driving their flexion and DLR Institute for Robotics and Mechatronic. It has 15 DoFs, five identical modular fingers with two 2.3.2. Commercial Prosthetic Hands flat BLDC motors placed in the base. Each finger has three active DoFs (2 DoFs offlexion and one of abduc- (i) i-Limb: myoelectronically controlled, externally tion) and four joints (the motions of distal and medial powered, tendon linking, multiarticulating pros- phalanges are coupled by a linkage). The thumb is thetic hand of Touch Bionics [27] with eleven joints fixed in an appropriate orientation of the palm (two joints in each long finger and three in the thumb). It has five individually powered digits and (iv) Shadow [32]: the Shadow Dexterous Hand is a powered thumb rotation, with manual override humanoid robot hand created by the Shadow Robot Company. The four fingers have 2 one-axis joints (ii) Bebionic [28]: multiarticulating myoelectric pros- (DIP and PIP) and one universal MCP joint; the lit- thetic hand developed by RSL Steeper with eleven tle finger has an extra one-axis joint on the metacar- joints (two joints in each long finger and three in the pal to provide arching. The thumb has a one-axis thumb). It has five actuators, one for each finger, joint (IP) and two universal joints (MCP and and the thumb has two positions manually placed by CMC). It contains 20 motors in the forearm (3 DoFs the user with an inbuilt sensor detecting the position. per finger, 5 DoFs in the thumb, 1 DoF in the palm, Folding links allow the fingers to flex. Among the 14 and 2 DoFs in the wrist) different grip patterns and hand positions that it can achieve, one of them is the finger adduction grip 3. Results and Discussion (iii) SensorHand Speed: the Ottobock SensorHand Speed [28] is a myoelectronically controlled hand with Figure 3 shows a representative plot of the mean value and three actuated fingers, which are driven by the same standard deviation across all the subjects (20) and repetitions motor. It is covered with a cosmetic glove that emu- (3) of the 16 joint angles measured with the CyberGlove Applied Bionics and Biomechanics 7 CMC1 CMC1 MCP1 IP1 F A F F 100 100 100 100 0 00 00 00 −100 −100 −100 −100 −100 −100 −100 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 M MCP2 CP2 M MC CP3 P3 M MC CP4 P4 MC MCP5 P5 F F F F F F F F 100 100 100 100 100 100 100 0 00 00 00 −100 −100 −100 −100 −100 −100 −100 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 PIP2 PIP2 PIP3 PIP3 PIP4 PIP4 PIP5 PIP5 F F F F F F F F 100 100 100 100 100 100 100 0 00 00 00 −100 −100 −100 −100 −100 −100 −100 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 00 0.5 0.5 11 M MCP CP22 M MCP4 CP4 M MCP5 CP5 Pa Palm lmaarrAr Arcch h A A A A A A 100 100 100 100 100 100 100 0 00 00 00 −100 −100 −100 −100 0 0.5 1 0 0.5 1 0 0.5 1 0 0.5 1 Figure 3: Mean and standard deviation of the 16 joint angles (in degrees) measured with the CyberGlove during the extension grip of the plate (T ). The angles are normalized across time (1: thumb, 2: index finger, 3: middle finger, 4: ring finger, and 5: little finger; CMC: carpometacarpal joint, MCP: metacarpophalangeal joint, PIP: proximal interphalangeal joint, DIP: distal interphalangeal joint, IP: interphalangeal joint; F: flexion/extension; and A: abduction/adduction). while performing a grasp of one object of the human grasp “Digit arching” (PC1) is the most significant synergy, as experiment (T of Figure 2). indicated by the higher value of s . 02 k Table 3 shows the relative contribution of the five PCs to Table 4 shows the matrix r (equation (4)) containing the ik each grasping task of the human grasp experiment n influence of the different groups of DoFs of the human hand (equation (8)) together with the final value of the parameter on the five kinematic functional synergies (Figure 1) and the s for each PC (equation (10)), obtained from these relative resulting parameter w after applying equation (3) with this k i contributions weight-averaged by their importance in ADL matrix r and the vector s (Table 3). ik k z [9]. The parameter w weights the relevance of the different g i The contributions of the different PCs to each grasping groups of DoFs in the human grasps in ADL. These results task ranged between 9.6% and 39.1%, indicating that all the indicate that finger flexion-extension is by far the most rele- five PCs have a nonnegligible importance in the twelve grasp- vant group of DoFs accounting for more than half of the ing tasks analyzed. It can be seen that PC1, corresponding to functionality, followed by thumb opposition and finger “digit arching,” is predominant in grasping tasks involving abduction-adduction. The palmar arching has a relevance PP (g = 4,8,12 ). Moreover, “palmar arching” (PC3) and of only 5%. This result by itself is valuable for making deci- “lateral pinch” (PC4) synergies are less represented in aver- sions during the design of new artificial hand prototypes to age in the four main GTs considered in the experiment, lead- maximize their functionality. ing to lower values of the final parameter s for these PCs, Table 5 shows the value of the parameter k (equation (2)) k i although the difference with “opposition” (PC5) is small. for each group of DoFs for the artificial hands analyzed. The 8 Applied Bionics and Biomechanics Table 3: Mean (SD) of the relative contribution n of the five PCs in each grasping task T (g: indicates the order of the tasks in Figure 2) and g g final value of the parameter s for each PC. PC1 (%) PC2 (%) PC3 (%) PC4 (%) PC5 (%) 19.2 (6.4) 24.4 (7.6) 18.4 (7.7) 13.0 (7.7) 24.9 (9.7) T 22.6 (6.9) 14.2 (6.3) 25.1 (6.1) 22.4 (8.1) 15.7 (8.4) 25.1 (4.7) 23.1 (4.7) 15.6 (4.3) 24.9 (8.9) 11.3 (5.5) T 33.5 (6.7) 19.1 (3.8) 12.6 (3.5) 12.1 (4.2) 22.7 (10.6) 22.4 (8.3) 21.4 (8.9) 18.6 (8.5) 17.4 (10.6) 20.1 (13.9) 28.2 (6.0) 21.0 (5.0) 17.9 (5.4) 17.0 (9.5) 16.0 (10.2) 28.3 (5.0) 21.1 (3.8) 12.9 (4.3) 27.0 (9.2) 10.8 (5.8) 39.1 (6.2) 23.6 (4.0) 12.9 (4.6) 11.8 (7.7) 12.5 (7.3) 19.5 (8.2) 24.5 (7.1) 14.3 (7.9) 20.3 (10.1) 21.3 (9.0) 30.9 (6.2) 18.6 (5.2) 20.1 (6.2) 9.6 (5.4) 20.7 (11.7) T 19.9 (3.4) 19.8 (7.7) 19.2 (7.9) 18.3 (10.8) 22.8 (13.3) 34.0 (7.4) 20.4 (5.1) 12.8 (4.3) 12.1 (6.5) 20.8 (9.6) s 29.0 20.5 16.5 15.6 18.4 Table 4: Matrix r and resulting w (equation (3)). ik i Functional synergies Groups of DoFs w (%) PC1 (%) PC2 (%) PC3 (%) PC4 (%) PC5 (%) Finger flexion-extension 79.6 50.7 42.5 51.8 34.9 55 Finger abduction-adduction 8.8 37.8 6.9 20.7 5.4 16 Palmar arching 4.4 3.4 15.3 0.2 1.9 5 Thumb opposition 7.2 8.1 35.3 27.3 57.9 24 Table 5: Parameter k (equation (2)) for each group of DoFs for the Table 6: Results of the Anthropomorphism Index of Mobility different artificial hands. (AIM) for different artificial hands and comparison with other indexes of the literature. Artificial hand F/E AB/AD P.ARC T.OPP Artificial hand AIM (%) AI (%) [3] A (%) [4] IMMA 0.48 0.25 0 0.50 IMMA 42 Cyborg Beast 0.29 0 0 0.18 Cyborg Beast 20 Flexy-Hand 0.39 0.25 0 0.23 Flexy-Hand 31 KIT 0.47 0 0 0.35 KIT 34 ADA 0.58 0.25 0 0.40 ADA 46 i-Limb 0.58 0 0 0.55 i-Limb 45 Bebionic 0.50 0 0 0.50 Bebionic 40 SensorHand 0.13 0 0 0.10 SensorHand 10 0.25 Michelangelo 0.13 0 0 0.30 Michelangelo 14 2.80 FRH-4 0.46 0 0.50 0.40 FRH-4 37 5.20 Barrett 0.25 0.38 0 0.30 Barret 27 10.38 DLR/HIT II 0.83 1 0 0.70 DLR/HIT II 78 26.61 Shadow 0.83 1 0.50 1 Shadow 88 39.93 F/E: finger flexion-extension, AB/AD: finger abduction-adduction, P.ARC: palmar arching, T.OPP: thumb opposition. details about the computation for each hand (c ) are supplied by the parameter w (Table 4). Notwithstanding, some hands ij i as Supplementary Materials. It can be seen that F/E and as the SensorHand and Michelangelo showed low scores in T.OPP are the groups of DoFs mainly included and actively F/E because of their rigid fingers without interphalangeal driven in the artificial hands, manifested by higher values joints. The unique hand with the 5 DoFs in T.OPP actively of k . It is worth to note that this fact is coherent with the driven is the Shadow hand. AB/AD is included actively in greater relevance of these groups of DoFs in ADL, as indicated DLR/HIT II and Shadow hands and passively through the Applied Bionics and Biomechanics 9 Table 7: Range of motion of the hand joints (in degrees) obtained in the human grasp experiment. ° ° ° ° ° Thumb ( ) Index ( ) Middle ( ) Ring ( ) Little ( ) F A F F F A FFFF A FF A F CMC CMC MCP IP MCP MCP PIP MCP PIP MCP MCP PIP MCP MCP PIP Min -27 0 -24 -32 -22 -9 0 -16 0 -13 -7 -1 -13 -7 -2 Max 32 28 13 42 51 24 62 65 66 68 16 76 69 12 68 P5 -6 0 -10 -5 -4 -3 1 -1 1 -1 -1 1 -3 -1 0 P95 15 19 2 17 30 9 36 40 42 30 8 49 26 7 40 CMC: carpometacarpal joint, MCP: metacarpophalangeal joint, PIP: proximal interphalangeal joint, DIP: distal interphalangeal joint, IP: interphalangeal joint, F: flexion (+)/extension (-), A: abduction (+)/adduction (-), P: percentile. use of deformable joints in some 3D-printed hands. Finally, the AIM as a method for evaluating the anthropomorphism P.ARC is only present in FRH-4 and Shadow hands. of an artificial hand. Finally, Table 6 shows the AIM for the different artificial (i) It is really quick to obtain: simply, the parameter k hands, obtained using equation (1) and considering the parameters shown in Tables 4 and 5. Two factors affect the has to be calculated, according to the DoFs and actu- ation methods of the artificial hand and equation (1) final AIM obtained by a hand (equation (1)): its mobility has to be applied (w is provided above) and type of actuation, represented by the number of DoF, the number of actuators, the number of digits and phalanges (ii) It analyzes not only the topology but also the func- per digit, and the type of underactuation, affecting to the final tionality of the artificial hand because it takes into parameters k ; and how this mobility and actuation system is account the results obtained in grasping tests and distributed among the different groups of DOFs, with regard ADL with the human hand to the human hand, affecting through the weighting factor w (Table 4). The most advanced robotic hands (DLR/HIT II and Shadow) with a significant amount of motors and DoFs, Notwithstanding, some important aspects in the design and located in the important groups of DoFs, with higher of an artificial hand are not within the scope of the AIM: weight w (F/E, T.OPP, and AB/AD), obtained the highest the orientation of the joint axes, the range of motion of the AIM scores, above 75%. The commercial prosthetic hands different hand joints, the dimension of the phalanges, the i-Limb and Bebionic as well as some 3D-printed hands friction coefficient of the parts of the hand in contact with (ADA, IMMA) obtained AIM scores between 40% and the objects, the grasping force exerted by the actuators, the 50%. These hands include a reasonable number of motors efficiency of the driving linkages, the control system, etc. and DoFs in the important groups (F/E and T.OPP). The rest Some previous studies [33–35] have shown the relevance of of the hands obtained scores below 40% with the lowest score these aspects. In this sense, the AIM, involving mainly the being for the SensorHand. The reason behind this lower AIM topological structure, the number of actuators, and the type is an improvable number of DoFs, motors, or type of under- of underactuation, can be considered as an index especially actuation in the groups of F/E, T.OPP, or both. useful in the concept design stage. The other design consider- The results shown in Table 6 indicate that the artificial ations cited above should be taken into account in later hands analyzed in the literature with other anthropomor- design stages: preliminary or detail design. Additional phism indexes, such as AI [3] or A [4], are ranked equally indexes that take into account these aspects could be interest- by the AIM and the other metrics, although the scores are ing, and future works can go in this way. The index proposed different. The method used to compute the indexes justify by Liu et al. [5] considers some of these aspects, but it does these different scores. The AI is obtained from the achievable not include their relevance for functionality according to workspace of positions and orientations of the fingers’ distal human grasping tests. With respect to the phalanx dimen- segments and compares this with information obtained sions and the joints’ range of motion, the authors developed experimentally from human hand grasping. The A is based some studies [13, 16, 36] helping to obtain anthropomorphic on the computation of the finger phalanx workspace com- designs. However, the evaluation of some of the design bined with that of the finger base frames, and the compari- aspects cited above is difficult to be performed with indexes, son with the human hand is made through a simplified requiring experimentation, after detailed design of the artifi- model of their joints and geometry. It is worth to note that cial hand and manufacturing a prototype. The authors have obtaining AI and A involves using complex algorithms proposed methods for this experimental evaluation consider- and detailed information of the hand design, not easily avail- ing the main GTs in ADL and a special device for actuating able, while obtaining the AIM just requires information the hand prototype [21]. about the number of DoFs and the possibility to control The ranges of motion of the hand joints obtained in the them independently. Despite these differences in the human grasp experiment undertaken in this study are shown method used to obtain each index, the fact that they rank in Table 7. A wide range of motion for the different joints equally, the hands as the AIM can be seen as a kind of val- was covered with the objects selected in comparison to the idation of our index. Two main points can justify the use of functional range of motion of the human hand joints in 10 Applied Bionics and Biomechanics ADL [16]. These ranges could be considered as a minimum findings of this study are included within the supplementary for prostheses with functional grasping for the main GTs, information files. although general manipulation would recommend using larger ranges if possible. Conflicts of Interest This study was primarily focused on prosthetic hands, and therefore, the scoring system takes into account the The authors declare that there is no conflict of interest capability of the hand to perform the most important GTs regarding the publication of this paper. for a nondominant hand to reinforce bimanual grasping (through parameter z ). For the case of a dominant hand Acknowledgments reinforcing bimanual grasping, the parameter z for the four GTs considered in this study changes to [9] PP (58.0%), EG This work was supported by the Spanish Ministry of (16.6%), TP (9.5%), and TVG (16.0%). The effect of this Economy and Competitiveness and ESF (grant number change on the resulting w is negligible and implies a dispar- BES-2015-076005); the Spanish Ministry of Economy and ity of the AIM obtained for the artificial hands analyzed Competitiveness, AEI, and ERDF (grant numbers DPI2014- (Table 6) of a maximum of 1%. Therefore, the AIM is con- 60635-R, DPI2017-89910-R); and Universitat Jaume I sidered useful to evaluate the anthropomorphism of both (grant numbers UJI-B2017-70, UJI-B2017-51). dominant and nondominant hands. With this result and the result obtained from the comparison of the AIM with Supplementary Materials other indexes of the literature [3, 4], we can conclude that the index proposed can be valid for artificial both robotic The excel file “JointAngles” contains the joint angles for each and prosthetic hands, regardless of whether they are domi- instant for the twelve tasks repeated three times by twenty nant or nondominant hands. subjects during the human grasp experiment. The excel file “Scores” contains the human hand kinematics in the human grasp experiment transformed to scores f referred to five 4. Conclusion tk functional synergies for the twelve tasks repeated three times In this study, we have presented an anthropomorphism by twenty subjects. The excel file “LoadingsMatrix” contains index (AIM) that can be used to evaluate and compare the the loadings l of the five functional synergies during ADL. ijk mobility of artificial hands in relation to the human hand The excel file “ArtificialHands” contains the value c accord- ij functionality, especially in concept design. The AIM evalu- ing to the method of actuation for each DoF in the artificial ates the topology of the whole hand (joints and DoFs) and hands analyzed and the corresponding k of each hand. the possibility to control these DoFs independently according (Supplementary Materials) to their functionality. We have shown that the index can be valid for both prosthetic and robotic hands, dominant and References nondominant hands. To define the index, the functionality of the different groups of DoFs of the hand (F/E, AB/AD, [1] J. T. Belter, J. L. Segil, A. M. Dollar, and R. F. Weir, “Mechan- P.ARC, and T.OPP) was analyzed according to a human ical design and performance specifications of anthropomor- grasp experiment on twenty subjects with the four main phic prosthetic hands: a review,” Journal of Rehabilitation GTs for personal autonomy in ADL. It was concluded Research and Development, vol. 50, no. 5, pp. 599–618, 2013. that the relevance of the different groups of DoFs (w ) [2] J. ten Kate, G. Smit, and P. Breedveld, “3D-printed upper limb was 55% for F/E, 16% for AB/AD, 5% for P.ARC, and prostheses: a review,” Disability and Rehabilitation. Assistive Technology, vol. 12, no. 3, pp. 300–314, 2017. 24% for T.OPP. Thirteen artificial hands, including affordable 3D-printed prosthetic hands, advanced commercial pros- [3] T. Feix, J. Romero, C. H. Ek, H. B. Schmiedmayer, and D. Kragic, “A metric for comparing the anthropomorphic thetic hands, and robotic hands, were evaluated and motion capability of artificial hands,” IEEE Transactions on compared with the AIM, and the reason for their differences Robotics, vol. 29, no. 1, pp. 82–93, 2013. was discussed. The results obtained in this study should be [4] M. V. Liarokapis, P. K. Artemiadis, and K. J. Kyriakopoulos, taken into account in the concept design stage of new proto- “Quantifying anthropomorphism of robot hands,” in 2013 types in order to obtain new designs that maximize their func- IEEE International Conference on Robotics and Automation, tionality. Further research will focus on new metrics for later pp. 2041–2046, Karlsruhe, Germany, May 2013. design stages considering other design aspects (range of [5] Y. Liu, D. Yang, L. Jiang, and H. Liu, “A synthetic framework motion of the joints, relative length of the phalanges, orienta- for evaluating the anthropomorphic characteristics of pros- tion of the joints axes, etc.) and on experimental benchmarks thetic hands,” in 2015 IEEE International Conference on to measure the grasping capability of artificial hands. Advanced Intelligent Mechatronics (AIM), pp. 877–884, Busan, South Korea, July 2015. Data Availability [6] L. Birglen, T. Laliberté, and C. Gosselin, Underactuated Robotic Hands, Springer Tracts in Advanced Robotics book The human hand kinematics expressed as joint angles and series (STAR, volume 40), Springer, 2008. scores referred to the five PCs, the loading matrix corre- [7] M. W. Cornwall, “Prosthetics and patient management: a sponding to these PCs, and the c values for the joints of ij comprehensive clinical approach,” Physical Therapy, vol. 87, the artificial hands analyzed; the data used to support the no. 4, p. 482, 2007. Applied Bionics and Biomechanics 11 [24] Gyrobot, “Flexy-Hand 2,” 2014, March 2019, https://www. [8] K. E. Yancosek and D. R. Mullineaux, “Stability of handwriting performance following injury-induced hand-dominance thingiverse.com/thing:380665. transfer in adults: a pilot study,” Journal of Rehabilitation [25] P. Weiner, J. Starke, F. Hundhausen, J. Beil, and T. Asfour, Research and Development, vol. 48, no. 1, pp. 59–68, 2011. “The KIT prosthetic hand: design and control,” in 2018 [9] V. Gracia-Ibáñez, J. L. Sancho-Bru, and M. Vergara, “Rele- IEEE/RSJ International Conference on Intelligent Robots and vance of grasp types to assess functionality for personal Systems (IROS), Madrid, Spain, October 2018. autonomy,” Journal of Hand Therapy, vol. 31, no. 1, [26] Open Bionics, “Open Bionics hands,” 2018, March 2019, pp. 102–110, 2018. https://openbionicslabs.com/downloads. [10] V. Gracia Ibañez, Contribution to hand functional assessment [27] Touch Bionics, “Touch Bionics products,” 2019, March 2019, based on its kinematics, Universitat Jaume I, Castelló de la http://touchbionics.com/products. Plana, 2016. [28] Ottobock, “Ottobock myoelectric prosthetics,” 2017, March [11] World Health Organization, Towards a common language for 2019, https://www.ottobockus.com/prosthetics/upper-limb- functioning, disability and health ICF the International Classi- prosthetics/solution-overview/myoelectric-prosthetics/. fication of Functioning, Disability and Health, World Heal. [29] I. Gaiser, S. Schulz, A. Kargov et al., “A new anthropomorphic Organ, Geneva, 2002. robotic hand,” in Humanoids 2008 - 8th IEEE-RAS Interna- [12] I. Llop-Harillo, A. Pérez-González, and V. Gracia-Ibáñez, tional Conference on Humanoid Robots, pp. 418–422, Daejeon, Anthropomorphism index of mobility for hand prostheses, South Korea, December 2008. XXVI Congress of the International Society of Biomechanics, [30] Barrett Technology, “BarrettHand,” March 2019, https://www. barrett.com/about-barretthand. [13] M. Vergara, M. J. Agost, and V. Gracia-Ibáñez, “Dorsal and [31] H. Liu, K. Wu, P. Meusel et al., “Multisensory five-finger palmar aspect dimensions of hand anthropometry for design- dexterous hand: the DLR/HIT hand II,” in 2008 IEEE/RSJ ing hand tools and protections,” Human Factors and International Conference on Intelligent Robots and Systems, Ergonomics in Manufacturing & Service Industries, vol. 28, Nice, France, September 2008. no. 1, pp. 17–28, 2018. [32] Shadow Robot Company, “Shadow dexterous hand,” 2019, [14] B. Calli, A. Walsman, A. Singh, S. Srinivasa, P. Abbeel, and March 2019, https://www.shadowrobot.com/products/ A. M. Dollar, “Benchmarking in manipulation research: dexterous-hand/. using the Yale-CMU-Berkeley object and model set,” IEEE Robotics and Automation Magazine, vol. 22, no. 3, pp. 36– [33] M. Controzzi, M. D'Alonzo, C. Peccia, C. M. Oddo, M. C. Carrozza, and C. Cipriani, “Bioinspired fingertip for anthro- 52, 2015. pomorphic robotic hands,” Applied Bionics and Biomechan- [15] V. Gracia-Ibáñez, M. Vergara, J. H. Buffi, W. M. Murray, and ics, vol. 11, no. 1-2, 38 pages, 2014. J. L. Sancho-Bru, “Across-subject calibration of an instru- mented glove to measure hand movement for clinical pur- [34] M. Ceccarelli, N. E. N. Rodríguez, G. Carbone, and C. Lopez- poses,” Computer Methods in Biomechanics and Biomedical Cajùn, “An optimal design of driving mechanism in a 1 degree of freedom (d.o.f.) anthropomorphic finger,” Applied Bionics Engineering, vol. 20, no. 6, pp. 587–597, 2017. and Biomechanics, vol. 2, no. 2, pp. 103–110, 2005. [16] V. Gracia-Ibáñez, M. Vergara, J. L. Sancho-Bru, M. C. Mora, and C. Piqueras, “Functional range of motion of the hand [35] F. J. Andrés, A. Pérez-González, C. Rubert, J. Fuentes, and joints in activities of the International Classification of Func- B. Sospedra, “Comparison of grasping performance of tendon tioning, Disability and Health,” Journal of Hand Therapy, and linkage transmission systems in an electric-powered low- vol. 30, no. 3, pp. 337–347, 2017. cost hand prosthesis,” Journal of Mechanisms and Robotics, vol. 11, no. 1, article 011018, 2018. [17] V. Patel, M. Burns, R. Chandramouli, and R. Vinjamuri, “Biometrics based on hand synergies and their neural repre- [36] I. Llop-Harillo, V. Gracia-Ibáñez, and A. Pérez-González, sentations,” IEEE Access, vol. 5, pp. 13422–13429, 2017. “Analysis of anthropometric dimensions and joints range of motion of the human hand for application to the design of [18] I. A. Kapandji, The Physiology of the Joints: Upper Limb, hand prostheses,” in 8th World Congress on Biomechanics, Churchill Livingstone, 1982. [19] J. L. Sancho-Bru, Modèl biomecànic de la mà orientat al disseny d’eines manuals, Universitat Jaume I, 2000. [20] D. G. Ullman, The Mechanical Design Process, McGraw-Hill, 4th edition, 2010. [21] I. Llop-Harillo and A. Pérez-González, “System for the exper- imental evaluation of anthropomorphic hands. Application to a new 3D-printed prosthetic hand prototype,” International Biomechanics, vol. 4, no. 2, pp. 50–59, 2017. [22] I. Llop-Harillo, A. Pérez-González, and J. Cantero-Ramis, “Analysis of motion synergies in grasping tasks for a 6-DOF tendon-driven prosthetic hand,” in 16th International Sympo- sium on Computer Methods in Biomechanics and Biomedical Engineering and 4th Conference on Imaging and Visualization, New York, 2019. [23] J. Zuniga, D. Katsavelis, J. Peck et al., “Cyborg beast: a low-cost 3d-printed prosthetic hand for children with upper-limb differences,” BMC Research Notes, vol. 8, no. 1, p. 10, 2015. International Journal of Advances in Rotating Machinery Multimedia Journal of The Scientific Journal of Engineering World Journal Sensors Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Control Science and Engineering Advances in Civil Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com Journal of Journal of Electrical and Computer Robotics Engineering Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 VLSI Design Advances in OptoElectronics International Journal of Modelling & Aerospace International Journal of Simulation Navigation and in Engineering Engineering Observation Hindawi Hindawi Hindawi Hindawi Volume 2018 Volume 2018 Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com www.hindawi.com www.hindawi.com Volume 2018 International Journal of Active and Passive International Journal of Antennas and Advances in Chemical Engineering Propagation Electronic Components Shock and Vibration Acoustics and Vibration Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal

Applied Bionics and BiomechanicsHindawi Publishing Corporation

Published: Jul 28, 2019

References