Inter national J our nal of Electrical and Computer Engineering (IJECE) V ol. 10, No. 6, December 2020, pp. 5772 5778 ISSN: 2088-8708, DOI: 10.11591/ijece.v10i6.pp5772-5778 r 5772 A haptic feedback system based on leap motion contr oller f or pr osthetic hand application Hussam K. Abdul-Ameer 1 , Luma Issa Abdul-Kr eem 2 , Huda Adnan 3 , Zahra Sami 4 1,3,4 Biomedical Engineering Department Al-Khw arizmi Colle ge of Engineering, Uni v ersity of Baghdad, Iraq 2 Control and Systems Engineering Department Uni v ersity of T echnology , Iraq Article Inf o Article history: Recei v ed Jul 13, 2019 Re vised Mar 1, 2020 Accepted Mar 8, 2020 K eyw ords: Haptic feedback InMoo v hand Leap motion controller Sensors placement ABSTRA CT Leap motion controller (LMC) is a gest ure sensor consists of three infrared light emitters and tw o infrared stereo cameras as tracking sensors. LMC translates hand mo v ements into graphical data that are used in a v ar iety of applications such as vir - tual/augmented reality and object mo v ements control. In this w ork, we intend to con- trol the mo v ements of a prosthetic hand via (LMC) in which fingers are fle x ed or e xtended in response to hand mo v ements. This will be carried out by passing in the data from the Leap Motion to a processing unit that processes the ra w data by an open- source package (Processing i3) in order to control v e serv o motors using a micro- controller board. In addition, hapt ic setup is proposed using force sensors (FSR) and vibro-motors in which the speed of t hese motors is proportional to the amount of the grasp force e x erted by the prosthetic hand. In v estig ation for optimal placement of the FSRs on a prosthetic hand to obtain con v enient haptic feedback has been carried out. The results sho w the ef fect of object shape and weight on the obtained response of the FSR and ho w the y influence the locations of the sensors. Copyright c 2020 Insitute of Advanced Engineeering and Science . All rights r eserved. Corresponding A uthor: Hussam K. Abdul-Ameer , Al-Khw arizmi Colle ge of Engineering, Uni v ersity of Baghdad, Baghdad, Iraq. Email: hussam@k ecb u.uobaghdad.edu.iq 1. INTR ODUCTION Capturing of human body motion is an increasing research area due to potential applications in robotics and informatics areas. Motion capture of the human hand is comple x task because each finger , and e v en each phalanx, has distinct and independent mo v ements that yield in return higher de grees of freedom [1-4]. Some techniques ha v e been de v eloped to capture hand motions, see [3-6]. Ho we v er , accu- rac y is a major setback to track hand motion. A leap motion controller is a motion sensor that w as de v eloped for hand motion tracking with high detection accurac y that reached 0.01 mm [7]. It consists of three parts, tw o cameras, LEDs, and a microcontroller in which the cameras capture successi v e images of the hand and then the y are passed to the controller to process the images and e xtract spatial information of the hand and fingers. Leap Motion has been used in se v eral projects to recognize hand gestures [8-13]. Although approaching the de xterity of a human hand to control robotic arm is dif ficult, LM w as used to do this task in [14-16], where hand gestures ha v e been translated to joint angles to perform a specific task. A potential application for LM is in robotic sur gery , in which a sur geon could control sur gical i nstruments from a console located in the operat- ing room [17-19]. T ouchless interaction has been in v estig ated in man y w orks. Such studies can impro v e user performance in dif ferent applications such as wheelchair control and maneuv ering and bro wsing of medical im- ages using leap motion [20-23]. Additional w orks ha v e been carried out to in v estig ate the haptic feedback with J ournal homepage: http://ijece .iaescor e .com Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 5773 implantation of leap motion, see [24-26]. Ho we v er , locations of the sensors that are responsible for generating appropriate control signals for haptic actuators need further in v estig ation. The aim of this w ork is to in v estig ate the rob ustness of a modified haptic-prosthetic system for telemedicine applications, where, we suggest using the leap motion controller to control t he mo v ements of a prosthetic hand and create haptic sensation for the subjects. The haptic system consists of three FSRs and tw o vibro-motors which are attached to a glo v e. A microcontroller is used to synchronize the mo v ement of the actuators with FSRs signals. In addition, we in v estig ate the response of FSRs at dif ferent spatial locations and using dif ferent objects. The paper is or g anized as follo ws, section 2 presents the adopted methodology while section 3 sho ws the used e xperiment al setup. Experimental results and discussion are described in section 4. Our conclusions are dra wn in the final section. 2. METHODOLOGY F or the sak e of clarity , the proposed methodology is di vided into three parts, leap motion, pros thetic hand, and haptic feedback system. Figure 1 sho ws a block diagram of the suggested control system. In the follo wing, we will e xplain each part in detail.                                               L eap Mo t i on C on t ro l l er     P ro ce ssi n g i 3   Use  P roc e ssi n g  to obt a in  fing e r’ s ti p dire c ti ons  a nd c a lcul a te joint   a ng l e s of f in g e rs         M icr oc on t r oll e r   (A r d u in o B oar d )   To c ontrol t he  se rvos  a nd vibro - mot ors a nd  a c quire   F S R s si ng les.        InMoov p ros t het i c h and   a ttac he to   F SR s   Hap tic  glove   (V 1 a nd  V 2 a r e  h a ptic  a c tu a tor s)     Figure 1. Block diagram of the suggested system for controlling InMoo v hand and creating haptic sensation. The solid arro ws refer to control signals while empty arro ws refer to feedback signals. The touchless interaction between haptic glo v e and prosthetic hand is e xpressed by tw o arro ws, the filled arro w to control the prosthetic hand while the empty arro w for haptic feedback 2.1. Leap motion Leap Motion Controller is human-hand tracking de vice equipped with infrared sensors that enable the LMC to respond swift ly to the fingers and hand mo v ements. It intends to re v olutionize the w ay that we are using our computers in f a v or of utilizing the touchless interacti on concept. The leap motion has a compact size which mak es it suitable for dif f erent control applications. In addition, the height of the interaction space is 60 cm abo v e the LMC where the field of the interaction is similar to an in v erted p yramid. This p yramidal shape increases the cameras field of vie w till 150 de grees with area detection capability ranging from 25 mm to 600 mm abo v e the LMC [1]. The first phase of the w ork methodology is to acquire the joint angles of the human- fingers’ mo v ements from LMC using Processing softw are which is an open source frame w ork for visual art programming. Then, the angles v alues are con v erted to appropriate control signals that are fed to actuators for dri ving prosthetic hand fingers 2.2. Pr osthetic hand W e suggest to use v e-finger prost hetic hand in which each finger can be controlled indi vidual ly according to the obtained information from the LMC. There are man y open-source 3D printable models of v e-finger prosthetic hand a v ailable in the internet. A reliable, anthropomorphic, and lo w-cost design to be produced is InMoo v hand [27]. This model has been adopted in this w ork since it is based on sharing and enhancing polic y . Some amendments on the fingers should be made to ensure full inte gration between the A haptic feedbac k system based on leap motion contr oller for ... (Hussam K. Abdul-Ameer) Evaluation Warning : The document was created with Spire.PDF for Python.
5774 r ISSN: 2088-8708 model and the proposed haptic system. These modifications related to the rounded finger geometry where we modify the shape of the fingers to allo w the feedback sensors fix ed on the finger’ s tip easily . Figure 2 sho ws the InMoo v model and the modified printed design.                                               (a)   ( b )           0   ( d )           0   ( c )           0   Figure 2. InMoo v prosthetic hand, (a) Sho ws the 3D printed model while, (b) sho ws the modified model, (c) presents the original finger model, (d) presents the modified inde x finger to allo w FSR fix ed at inde x tip. All models ha v e been printed using Mak erbot replicator 2.3. Haptic system There are dif ferent configurations to create a haptic sens ation. The one that we proposed in this w ork is based on the feedback obtained from reaction forces on the grasped object. These reactions will trigger haptic actuators where a vibro-motor is used for this purpose. Adding haptic feedback to the touchless interaction has the potential for man y applications such as g aming and rehabilitation. 3. EXPERIMENT AL SETUP T o demonstrate the proposed methodology , practical part of this w ork has been di vided into tw o subsections. 3.1. LMC and haptic system LMC has been placed before the user to acquire images of the user’ s hands which the y should be set within 150 de grees from LM and from 25 mm to 600mm directly abo v e the de vice. The captured information is then transferred to personal computer to e xtract information related to figures mo v ements and directions. Processing softw are is used for that purpose in which proper libraries should be added to the Processing to be able to read and interpret the incoming information from LMC. A setback is detected due to lar ge amount of data acquired by LMC that Processing softw are cannot handle smoothly . This lagging problem has been min- imized by reduci ng the frames to be processed and use a personal computer with higher processing capability . A haptic setup is proposed using three force-sensi ng-resisti v e (FSR), type Interlink 402, and tw o vibro-motors, model 1027. The FSR s are laid out at dif ferent locations on the modified InMoo v hand. As mentioned earlier , these sensors are used to generate the signals that trigger the haptic actuators in which the mean v alue of the incoming signals is used to acti v ate the haptic actuators, vibrating motors. T o mimic a realistic grasp feeling, the speed of the haptic actuators is proportional to the mean v alue. 3.2. Pr osthetic hand and sensor calibrations The prosthetic hand used in this w ork w as InMoo v where the fingers’ shape has been modified to be capable of holding the FSR, see Figure 2. The hand w as printed using 3D Mak erBot Replicator and PLA filament. The mo v ement of each finger is actuated by a serv omotor type T o werProMG995. Fi v e serv o motors for al l fingers were used. Serv o horn and fishing line are attac h e d with ea ch serv omotor to control the motion direction of the fingers where the y can be fle x ed or relax ed rel ati v e to the motor rotational direction. A micro- controller type Arduino MEGA i s utilized to control the serv omotors, vibrating motors a nd FSRs. The FSRs ha v e been calibrated using dif ferent range of weight v al u e s and a best fitting polynomial function between the Int J Elec & Comp Eng, V ol. 10, No. 6, December 2020 : 5772 5778 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 5775 output v olt age from each FSR and the used weights w as obtained for the used FSRs. These polynomial func- tions were used to obtain FSRs responses for selecti v e locations on the InMoo v . A schematic diagram of the used components and their wiring is sho wn in Figure 3.                                               (1)   (2)   (3)   (4)   (5)   Figure 3. Schematic diagram of the haptic–hand system, (1) vibrating motor (2) FSR (3) MEGA Arduino (4) serv o motor (5) battery 4. RESUL TS AND DISCUSSION In this w ork, LMC is used to control the motion of a modified v ersion of InMoo v hand and return haptic feedback. In addition, locations of feedback sensors for the suggested haptic system are in v estig ated to pro vide a more realistic touch e xperience. T o v alidate the obtained results, v e subjects were v olunteered to carry out the prosthetic hand control e xperime n t s, in which the tests ha v e been carried out for all participants. T able 1 sho ws the age and gender of the subjects. Although the angles and directions g athered from leap motion functions are dif ferent for each participant, the hand response is perfectly matched for all subjects’ hands mo v ements. Rarely , response lagging is occurred due to limitation i n Processing softw are ability for images acquiring and processing. T able 1. The participants gender and age. Maximum and minimum angles v alues for each finger recorded by the LMC (full straight and straight-fist). Inde x Middle Ring Pink y Thumb Gender Age Min v alue Max v alue Min v alue Max v alue Min v alue Max v alue Min v alue Max v alue Min v alue Max v alue Male 60 -82 22 -87 42 -80 35 -52 20 -60 39 Male 20 -61 23 -80 39 -84 42 -66 35 -98 31 Female 50 -67 15 -82 22 -75 24 -51 19 -88 38 Female 25 -72 32 -81 51 -82 57 -75 20 -68 30 Female 17 -65 23 -82 40 -75 33 -53 30 -80 24 An in v estig ation has been carried out for optimal placement of the FSRs such that these FSRs can return a con v enient haptic feedback. Since the c ylindrical objects are widely util ized to e xamine the grasping reactions [28] , three c ylindrical objects with dif ferent diameters and 12 locations on the prosthetic hand ha v e been selected for this in v estig ation. Figure 4 sho ws the layout of the FSR sensors in which the 12 locations are disseminate along the prosthetic hand and the output responses from FSRs were recoded and presented in Figure 5. It can be noticed that the grasping reactions a t locations 3 till 6 are, in general, tend to decrease compared with other locati o ns . Such performance is link ed to the c ylindrical s hape of the grasped object in which line contact between the grasped object and the upper part of the prosthetic palm is occurred. Inter - estingly , a wide v ariance in the v alue of the reactions at locations 7, 8 and 9 are detected when the objects’ diameters are changed. This finding is related to the object and modified ng e rs geometry where the object is surrounded by the fingers and this encircling let FSRs generate electric v oltage proportional to the size of the grasped object. Ho we v er , increasing the object size up to a critical v alue, which needs further in v estig ating, can cause a superficial result due to missing the contact bet ween the sensors and the grasped object. A similar response for FSRs is occurred when the sensors are positioned at locations 10 and 11 while location 12 returns a less reaction v alue compared with pre vious locations. The reason for this rather contradictory re sult is still not completely clear , b ut we refer that to palm shape of InMoo v . Our findings w ould seem to sho w that the A haptic feedbac k system based on leap motion contr oller for ... (Hussam K. Abdul-Ameer) Evaluation Warning : The document was created with Spire.PDF for Python.
5776 r ISSN: 2088-8708 FSRs locate in the inde x and middle fingers’ tips gi v e better reactions than other locations assuming that the grasped objects ha v e a c ylindrical shape. Ho we v er , dif ferent shapes for tests objects need to be in v estig ated to obtain a comprehensi v e understanding of relation bet ween the grasped objects and the modified prosthetic hand. P artially captured objects can also af fect the grasping reactions that m ay decrease the reaction v alue and also may case slipping of the grasped objects. Thus, the applied grasped force from the real hand should be proportional to the object diameter to pre v ent object sliding. This requires a further study which is be yond the scope of this w ork. The weight of the object af fects the rea ction depending on the pros thetic hand layout. In the case of a horizontal layout, the FSR is influenced by the object weight while in a v ertical layout, the weight does not contrib ute to the grasping reactions. Instead, it has a slipping ef fect which should be balanced by increasing the applied grasping force.                                                 Figure 4. T est locations on the InMoo v to in v estig ate the FSRs responses when a c ylindrical object is used                                                 Figure 5. FSRs responses at dif ferent locations on the InMoo v e hand, Cylindrical objects are used to in v estig ated the c ylindrical grasp 5. CONCLUSION In this w ork LMC is utilized to control anthropomorphic hand with haptic feedback to perform c ylindrical grasping tasks. The proposed system w ork ed fine and smooth response in general w as obtained. Although, the LMC enables decent touchless interaction between the user and the prosthetic hand, geometry of the prosthetic hand requires additional modifications and dynamic analysis to ensure full inte gration with the haptic system. Such in v estig ations will impro v e the potential of using LMC in telesur gery and tele e xamination since accurac y and interaction are utmost k e y f actors. W e in v estig ated the locations of the feedback sensors for the proposed haptic system where 12 locations distrib uted along the prosthetic hand were sel ected. The tip of the middle finger yields a good grasping reaction despite v arying in object’ s shape. Placing the FSR on the hand palm can gi v e acceptable reaction, ho we v er , object size can decrease the contact area and return less ef fecti v e contact reaction. W e recommend a further analysis for optimal sensors placement pre v enting object slipping. Int J Elec & Comp Eng, V ol. 10, No. 6, December 2020 : 5772 5778 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 5777 This may be achie v ed by fusion the date obtained from FSRs and a proximity sensor , where the return signal from the last sensor can detect the occurrence of object slipping while the FSRs maintain the haptic feedback. A CKNO WLEDGMENT W e w ould lik e to thank Dr . Ali H. Al-T imemy for pro viding a printed InMoo v hand. Dr . Al-T imemy is a member of staf f at Al-Khw arizmi colle ge of Engineering, Uni v ersity of Baghdad REFERENCES [1] R. Brouet, R. Blanch, and M.-P . Cani, “Understanding Hand De grees of Freedom and Natural Gestures for 3d Interaction on T abletop, Human-Computer Interaction–INTERA CT , v ol. 8117, pp. 297–314, 2013. [2] B. Buchholz, T . J. Armstrong, and S. A. Goldstein, Anthropometric data for describing the kinematics of the human hand, Er gonomics, v ol. 35, no. 3, pp. 261–273, 1992. [3] J. Lin, Y ing W u, and T . Huang, “Capturing human hand motion in image sequences, Proceedings IEEE Comput. Soc, W orkshop on Motion and V ideo Computing, pp. 99–104, 2002. [4] A. H. Al-T imemy , R. N. Khushaba, G. Bugmann, and J. Es cudero, “Impro ving the performance ag ainst force v ariation of emg controlled multifunctional upper -limb prostheses for transradial amputees, IEEE T ransactions on Neural Systems and Rehabilitation Engineering, v ol. 24, no. 6, pp. 650–661, 2016. [5] C. C. Moldo v an and I. Staretu, “Capturing human hand mo v ements with a webcam to control an anthro- pomorphic gripper , IProcedia Manuf acturing, v ol. 22, pp. 519–526, 2018. [6] M. Y e, et al., A Surv e y on Human Motion Analysi s from Depth Data, IT ime-of-Flight and Depth Imag- ing. Sensors, Algorithms, and Applications, pp. 149–187, 2013. [7] F . W eichert, D. Bachmann, B. Rudak, and D. Fisseler , Analysis of the Accurac y and Rob ustness of the Leap Motion Controller , Sensors, v ol. 13, no. 5, pp. 6380–6393, 2013. [8] G. Marin, F . Dominio, and P . Zanuttigh, “Hand gesture recognition with leap motion and kinect de vices, IEEE International Conference on Image Processing, pp. 1565–1569, 2014. [9] W . Lu, Z. T ong, and J. Chu, “Dynamic Hand Gesture Recognition W ith Leap Motion Controller , IEEE Signal Process. Lett., v ol. 23, no. 9, pp. 1188–1192, 2016. [10] S. Ameur , A. B. Khalif a, and M. S. Bouhlel, A comprehensi v e leap motion database for hand gesture recognition, 7th International Conference on Sciences of Electronics, T echnologies of Information and T elecommunications, pp. 514–519, 2016. [11] A. Elons, M. Ahmed, H. Shedid, and M. T olba, Arabic sign language recognition using leap motion sensor , 9th International Conference on Computer Engineering and Systems, pp. 368–373, 2014. [12] A. Setia w an and R. Pulung an, “Deep Belief Netw orks for Recognizing Handwriting Captured by Leap Motion Controller , IJECE, v ol. 8, no. 6, p. 4693, Dec. 2018. [13] K. N. Y asen, F . L. Malallah, L. F . Abdulrazak, A. M. Darwesh, A. Khmag, and B. T . Shareef, “Hand detection and se gmentation using smart path tracking fingers as features and e xpert system classifier , International Journal of Electrical and Computing Engineering, v ol. 9, no. 6, 2019. [14] Y . Pititeeraphab, P . Choitkunnan, N. Thongpance, K. K ullathum, and C. Pinta virooj, “Robot-arm control system using LEAP motion controller , International Conference on Biomedical Engineering, pp. 109–112, 2016. [15] S. Chen, H. Ma, C. Y ang, and M. Fu, “Hand Gesture Based Robot Control System Using Leap Motion, Intelligent Robotics and Applications, v ol. 9244, pp. 581–591, 2015. [16] I. Staretu and C. Moldo v an, “Leap Motion De vice Used to Control a Real Anthropomorphic Gripper , International Journal of Adv anced Robotic Systems, v ol. 13, no. 3, 2016. [17] F . Despino y , N. Zemiti, G. F orestier , A. Sanchez, P . Jannin, and P . Poignet, “Ev aluation of contactless human–machine interf ace for robotic sur gical training, Int J CARS, v ol. 13, no. 1, pp. 13–24, 2018. [18] T . A. T ra v aglini, P . J. Sw ane y , K. D. W ea v er , and R. J. W ebster III, “Initial Experiments with the Leap Mo- tion as a User Interf ace in Robotic Endonasal Sur gery , Robotics and Mechatronics, v ol. 37, pp. 171–179, 2016. [19] Y . Kim, P . C. W . Kim, R. Selle, A. Shademan, and A. Krie ger , “Experiment al e v aluation of contact-less hand tracking systems for tele-operation of sur gical tasks, IEEE International Conference on Robotics and Automation, pp. 3502–3509, 2014. A haptic feedbac k system based on leap motion contr oller for ... (Hussam K. Abdul-Ameer) Evaluation Warning : The document was created with Spire.PDF for Python.
5778 r ISSN: 2088-8708 [20] M. Sathiyanarayanan and S. Rajan, “Understanding the use of leap motion touchless de vice in ph ysio- therap y and impro ving the healthcare system i n India, 9th International Conference on Communication Systems and Netw orks, pp. 502–507, 2017. [21] S. Nicola, L. Stoicu-T i v adar , I. V irag, and M. Crisan-V ida, “Leap Motion supporting medical education, 12th IEEE International Symposium on Electronics and T elecommunications, pp. 153–156, 2016. [22] A. Skraba, A. K olozv ari, D. K ofjac, and R. Stojano vic, “Wheelchair maneuv ering using leap motion controller and cloud based speech control: Prototype realization, 4th Mediterranean Conference on Em- bedded Computing, pp. 391–394, 2015. [23] G. M. Rosa and M. L. Elizondo, “Use of a gesture user interf ace as a touchless image na vig ation system in dental sur gery: Case series report, Imaging Sci Dent, v ol. 44, no. 2, pp. 155–160, 2014. [24] R. G. Lupu, N. Botezatu, F . Ungureanu, D. Ignat, and A. Moldo v eanu, “V irtual reality based strok e reco v ery for upper limbs using leap motion, 20th International Conference on System Theory , Control and Computing, pp. 295–299, 2016. [25] E. Freeman, S. Bre wster , and V . Lantz, “T actile Feedback for Abo v e-De vice Gesture Interf aces: Adding T ouch to T ouchless Interactions, Proceedings of the 16th International Conference on Multimodal Inter - action, pp. 419–426, 2014. [26] Mingyu Kim, Changyu Jeon, and Jinmo Kim, A Study on Immersion and Presence of a Portable Hand Haptic System for Immersi v e V irtual Reality , Sensors, v ol. 17, no. 5, 2017. [27] A. Bulg arelli , G. T oscana, L. O. Russo, G. A. F arulla, M. Indaco, and B. Bona, A Lo w-Cost Open Source 3d-Printable De xt erous Anthropomorphic Robotic Hand wi th a P arallel Spherical Joint Wrist for Sign Languages Reproduction, International Journal of Adv anced Robotic Systems, v ol. 13, no. 3, 2016. [28] M. Nieuwenhuisen, J. Stueckler , A. Berner , R. Klein, and S. Behnk e, “Shape-primiti v e based object recognition and grasping, 7th German Conference on Robotics, pp. 1–5, 2012. Int J Elec & Comp Eng, V ol. 10, No. 6, December 2020 : 5772 5778 Evaluation Warning : The document was created with Spire.PDF for Python.