Access the full text.
Sign up today, get DeepDyve free for 14 days.
B. Chaparro-Rico, D. Cafolla, M. Ceccarelli, E. Castillo-Castañeda (2017)
Design and Simulation of an Assisting Mechanism for Arm Exercises, 47
S. Chatzis, Dimitrios Korkinof, Y. Demiris (2012)
A Quantum-Statistical Approach Towards Robot Learning by Demonstration
(2018)
Ergo Arm Skate, Ref
(2015)
Development of a mechatronic device for rehabilitation of the upper extremity (shoulder-elbowwrist)
P. Lum, C. Burgar, P. Shor, Matra Majmundar, Machiel Loos (2002)
Robot-assisted movement training compared with conventional therapy techniques for the rehabilitation of upper-limb motor function after stroke.Archives of physical medicine and rehabilitation, 83 7
(2015)
An apparatus for upper body movement
T. Nef, M. Guidali, R. Riener (2009)
ARMin III --arm therapy exoskeleton with an ergonomic shoulder actuationApplied Bionics and Biomechanics, 6
Patricia Leconte, R. Ronsse (2016)
Performance-based robotic assistance during rhythmic arm exercisesJournal of NeuroEngineering and Rehabilitation, 13
(2002)
>e Joint Health Prescription: 8 Weeks to Stronger, Healthier, Younger Joints, Ballantine Books
Ciarán O’Neill, Nathan Phipps, L. Cappello, S. Paganoni, C. Walsh (2017)
A soft wearable robot for the shoulder: Design, characterization, and preliminary testing2017 International Conference on Rehabilitation Robotics (ICORR)
Ying Mao, S. Agrawal (2012)
Design of a Cable-Driven Arm Exoskeleton (CAREX) for Neural RehabilitationIEEE Transactions on Robotics, 28
L. Ombregt (1995)
A System of Orthopaedic Medicine
Hui Bian, Ziye Chen, Hao Wang, T. Zhao (2017)
Mechanical design of EFW Exo II: A hybrid exoskeleton for elbow-forearm-wrist rehabilitation2017 International Conference on Rehabilitation Robotics (ICORR)
Ileana Corona-Acosta, E. Castillo-Castañeda (2015)
Dimensional Synthesis of a Planar Parallel Manipulator Applied to Upper Limb Rehabilitation
S.J. Ball, I. Brown, S. Scott (2007)
A planar 3DOF robotic exoskeleton for rehabilitation and assessment2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society
R. Riener, T. Nef, G. Colombo (2005)
Robot-aided neurorehabilitation of the upper extremitiesMedical and Biological Engineering and Computing, 43
C. Burgar, P. Lum, P. Shor, H. Loos (2000)
Development of robots for rehabilitation therapy: the Palo Alto VA/Stanford experience.Journal of rehabilitation research and development, 37 6
(2016)
Device for arm motion assistance
J. Annisa, S. Mohamaddan, M. Jamaluddin, A. Aliah, A. Omar, H. Helmy, A. Norafizah (2014)
Development of upper limb rehabilitation robot prototype for home setting
H. Krebs, N. Hogan, M. Aisen, B. Volpe (1998)
Robot-aided neurorehabilitation.IEEE transactions on rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society, 6 1
(2016)
Design and test of therapy exercise for human arms
P. Maciejasz, J. Eschweiler, K. Gerlach-Hahn, Arne Jansen-Troy, S. Leonhardt (2014)
A survey on robotic devices for upper limb rehabilitationJournal of NeuroEngineering and Rehabilitation, 11
(2003)
Rehabilitation Techniques in Sports
Hindawi Journal of Healthcare Engineering Volume 2018, Article ID 9303282, 15 pages https://doi.org/10.1155/2018/9303282 Research Article Experimental Characterization of NURSE, a Device for Arm Motion Guidance 1,2 2 2 Betsy Dayana Marcela Chaparro-Rico , Daniele Cafolla , Marco Ceccarelli , and Eduardo Castillo-Castaneda Instituto Polit´ecnico Nacional-CICATA Quer´etaro, Cerro Blanco 141, Colinas del Cimatario, 76090 Santiago de Quer´etaro, QRO, Mexico Laboratory of Robotics and Mechatronics (LARM), University of Cassino and Southern Lazio, Via Di Biasio 43, 03043 Cassino, Italy Correspondence should be addressed to Betsy Dayana Marcela Chaparro-Rico; betsychaparro@hotmail.com Received 24 February 2018; Accepted 20 May 2018; Published 3 July 2018 Academic Editor: Antonio Ferna´ndez-Caballero Copyright © 2018 Betsy Dayana Marcela Chaparro-Rico et al. )is 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. )is paper presents an experimental characterization of NURSE, a device for arm motion guidance. )e laboratory setup and testing modes are presented to explain the experimental procedure. Two exercises for the upper limb exercise are used to test the NURSE behaviour, and successful results are presented. Trajectories and linear accelerations are tested when the device performs the two exercises without and with load. In addition, torque and power consumption are considered to check the NURSE behaviour. therapist to keep the same quality of motions during long- 1. Introduction therapy sessions. In addition, the motion cannot be con- Every year, 15 million people worldwide suffer a stroke; trolled, and a feedback of the patient evolution is difficult to more than 70 percent must deal with mobility impairment obtain. While there remain a number of tasks that only human therapists can perform, many rehabilitation exercises and cognitive disabilities [1]. Additionally, the arm mobility can be affected by neurological, muscle, and joint diseases are mainly highly repetitive. )is is where robotic systems [2]. Lymphatic and vascular disorders can also reduce arm are useful since they can reproduce the same task countless mobility [3]. On the contrary, the arm mobility can be also times, with precision and accuracy without fatigue or loss of affected by traumatic and overuse injuries of the shoulder, attention [12]. It has been proved that use of robotic systems elbow, and wrist [3, 4]. In such a case, exercises are necessary benefits the rehabilitation process [13, 14]. In addition, the to recover a suitable range of motion by strengthening, use of robotic systems reduces the recovery time by 30% [13]. flexing, and extending the muscles and the joints [5]. Several devices have been developed for arm motion as- However, the number of trained human therapists who can sistance. However, there are several issues to solve in the provide this support is limited, while the demand is growing, existing robotic devices such as they are costly and they have particularly in elderly people [6, 7]. )e required exercises bulky structures very difficult to adjust to the patient arm. )e existing devices for arm motion assistance can be for an assistive therapy should be designed by a specialist according to the medical diagnosis, and it can be vary from classified into three groups: nonactuator devices, exo- a specialist to another [3, 8, 9]. However, all exercises start skeletons, and end-effector devices. )e nonactuator devices from the basic movements of the human arm seen in are frequently used by rehabilitation centers since they have [3, 4, 8, 10, 11]. During a traditional exercise, the specialist significant lower costs, are easier to use, and are inherently assists the limb motion. However, it is difficult for the safe. An example is the handboard to trace the number 8 2 Journal of Healthcare Engineering exercises. In [25], a portable end-effector device is pro- [15], and the mechanism is composed of a roller skate for the arm support, a table with a guide with an 8 shape, and posed for arm exercises on an inclined plane. )e device is composed of two actuators that are actuated by cables and pieces of different weights to apply resistance to the motion during the therapy. However, the mechanism a hand grip device. )e device trajectories are limited by offers only one exercise, and the arm motion is not con- four guides that constrain the end-effector movement trolled. Another example is the skateboard [16, 17]. along straight lines, and the device cannot perform other )e skateboard is a known mechanism composed of types of exercises. In [26], MIT-MANUS is presented, a board with wheels that allow movements on a horizontal a commercial and known device for arm therapy that has plane. )e patient should perform the movements by been developed in the early 1990s. MIT-MANUS is principally composed of a five-bar mechanism and himself/herself. Since the skateboard is a cheap mecha- nism, it is widely used. However, by using the skateboard, a modular end-effector. )e robotic arm helps in the shoulder and elbow motion on a horizontal plane, and the the therapy motion cannot be controlled. However, the nonactuator devices do not have movement correction. modular end-effector allows the movements of the wrist joint. Currently, MIT-MANUS has a clinical version that is Within the exoskeletons group, ArmeoPower can be found [18]. ArmeoPower has six degrees of freedom to perform named “InMotion ARM ” as pointed out in [27]. How- 3D motions and a graphical interface for virtual in- ever, the device has a reduced workspace in terms of the teraction. However, ArmeoPower is difficult to wear, range of possible motions. Furthermore, the device is not costly, and has a bulky frame. Another exoskeleton is portable and it requires to be operated by highly trained MEDARM [19], which can assist the arm motion on personnel. Another end-effector device named “REAplan” a horizontal plane; it is actuated by cables and has 3 de- is presented in [28]. )e device is based on the Cartesian grees of freedom. )e MEDARM exoskeleton is adjustable mechanism with a handle that is moved on a horizontal for users of different sizes. However, MEDARM needs plane to assist the arm motion. However, REAplan has a bulky frame structure, is difficult to transport and a bulky and heavy structure so that is difficult to transport. construct, and has been proposed only to assist the right In addition, it has a reduced workspace in relation to the arm, and it is difficult to align the exoskeleton joints with required link sizes. However, the end-effector devices the human arm joints. Another exoskeleton named present advantages with respect to the exoskeletons such as “CAREX” is proposed in [20]. )e exoskeleton is actuated they present a simple structure and control and they are by seven cables and has five degrees of freedom. However, easy to adjust to the patient. CAREX needs a very huge structure to support the seven As seen in the above examples, the main issues to motors that move the cables. In addition, the cables can be consider about the existing devices for arm motion are that dangerous for the subject since they move close to his/her the devices with a large workspace are very difficult to head. Other exoskeletons with similar disadvantages to transport, construct, and wear as seen in [18–23]; the ArmeoPower, MEDARM, and CAREX can be seen in existing portable devices cover a small workspace [16, 24–28] [21–23]: in [21], an exoskeleton is proposed to assist just and offer few types of exercises [24, 25]; and the widely used the shoulder motion, but it is difficult to wear; in [22], an basic mechanisms do not have motion control during the exoskeleton is proposed to assist the elbow and wrist therapy or they perform a single trajectory as seen in [15, 17]. joints, but it has a bulky frame structure and it is not In order to solve the above issues, NURSE (cassiNo- comfortable to use since the frame must be placed in the qUeretaro uppeR-limb aSsistive dEvice) was developed as an middle of the patient’s legs so that the arm gets a the proper alternative solution for arm motion assistance with ad- position; in [23], an exoskeleton named “ARMin III” is vantages over the existing devices. NURSE is an end-effector proposed, but like ArmeoPower, it is difficult to wear and device composed of a mechanism, a controller, and a user has a bulky frame. As seen in [18–23], the main issues in interface. NURSE is based on a mechanism of 2 degrees of the exoskeletons are that the expostulations have joint axes freedom whose workspace is amplified by using a panto- fully determined as well as physiological movements, but graph. NURSE can assist the arm motion during a re- robot axes have to be aligned with anatomical axes and are habilitation therapy and the arm motion of elderly people very difficult to transport, construct, and wear. In addition, during an exercise. )e main advantages of NURSE are the exoskeletons are very difficult to adapt to different presented in this paper together with the experimental anthropometric sizes. An example of an end-effector de- characterization. vice can be seen in [16]. )e device is based on a planar parallel mechanism 3RRR. )e device can assist the arm 2. Exercises for Arm Motion Guidance motion on a horizontal, vertical, or inclined plane by performing several trajectories within its workspace. In order to assist the arm motion during a therapy, two However, the device has large links and presents stiffness exercises for upper limb rehabilitation and exercise have problems. In [24], an end-effector device is proposed to been designed by the authors as reported in [29] (Figure 1). assist the arm motion. )e forearm of the user is supported )e considered exercises can be used in patients recovering by an end-effector device, and the device can assist the from injuries and neurological, muscular, and joint diseases. shoulder/elbow flexion and extension without other tra- Moreover, they can also be used for the arm exercise by jectories. )e disadvantage of this end-effector device is elderly people. Figure 1(a) shows exercise no. 1 that has been that it covers a small workspace and offers few types of designed to treat the shoulder. )e exercise consists of Journal of Healthcare Engineering 3 TP A A B X X (a) (b) Figure 1: e two considered exercises for upper limb rehabilitation and exercise: (a) exercise no. 1 to treat the shoulder joint; (b) exercise no. 2 to treat both shoulder and elbow joints. 1500 1800 1600 1700 1500 1600 1700 1800 X (mm) X (mm) (a) (b) Figure 2: Reference trajectories generated by regression analysis (in black) and the trajectories acquired from the 12 subjects: (a) trajectories for exercise no. 1 to treat the shoulder joint; (b) trajectories for exercise no. 2 to treat both the shoulder and elbow joints. TP Wheel End-effector Wheel (a) (b) Figure 3: NURSE: (a) a prototype; (b) the tracing point (TP) on the end-eƒector and wheels. Y (mm) Y (mm) 4 Journal of Healthcare Engineering Top Front camera camera Current sensing Control unit IMU Laptop (a) Current sensing Motor control boards Emergency switches Power supplies (b) Figure 4: Experiment layout: (a) overview of the lab setup; (b) control area details. performing a horizontal shoulder ‡exion by tracing the hand is guided by a specialist to perform a desired exercise, it trajectory in red dotted lines with a tracing point (TP) from is assumed that a device for motion assistance should the point A to the point B (Figure 1(a)). Figure 1(b) shows perform the path of the same exercise. A Kinect vision exercise no. 2 that has been designed to treat both the system [31] was used to carry out the data collection of the arm motion from 12 subjects that performed the above shoulder and elbow joints. e exercise consists of tracing the number 8 with the TP. Exercise no. 2 starts and ends in exercises. e subjects performed each exercise during 12 the same point. Since the path to trace the number 8 is repetitions. From the collected trajectories, a reference complex, it is also used as a reference trajectory to evaluate trajectory was generated for each exercise by using re- the behaviour of robots that perform human tasks [30]. gression analysis as reported in [29]. Figure 2 shows the e procedure for the motion design of the considered trajectories generated for each exercise and the trajectories exercises is explained in [29]. e reached coordinates of the acquired by the Kinect vision system. It is important to TP with respect to an XY reference frame were used for the notice that other arm exercises have also been designed design motion. Since in an assistive therapy, the patient’s in [29]. Control area Mechanism area Journal of Healthcare Engineering 5 (a) Figure 5: Markers for image processing. Start Actuators Interface Device homing turned on initialization (b) Trajectory selection Running test Data acquisition (c) Is data No collection Figure 7: Some snapshots of test no. 1 without load during rep- ok? etition no. 1 together with the trajectory obtained by image pro- cessing (in red): (a) the šrst sample position; (b) the second sample Yes position; (c) the third sample position. Postprocessing stage Device End Figure 6: Experiment ‡ow chart. Table 1: Experiment to test NURSE. –200 –150 –100 –50 0 50 100 150 200 Test Description Inputs Outputs X (mm) no. Repetition no. 1 Repetition no. 3 Perform exercise no. Repetition no. 2 Programmed trajectory 1 X , Y , a , a , p, τ , and m m x y 1 X, Y Perform exercise no. τ Figure 8: Comparison between the programmed trajectory and the TP trajectories from three repetitions during test no. 1 without load. Y (mm) 6 Journal of Healthcare Engineering 0.2 0.2 0.15 0.15 0.1 0.1 0.05 0.05 0 0 –0.05 –0.05 0 51015 0 51015 Time (s) Time (s) Repetition no. 1 Repetition no. 1 Repetition no. 2 Repetition no. 2 Repetition no. 3 Repetition no. 3 (a) (b) Figure 9: Acquired linear acceleration during test no. 1 without load for the three repetitions seen in Figure 8: (a) X linear acceleration; (b) Y linear acceleration. 3000 3000 1000 1000 0 0 –1000 –1000 05 10 15 05 10 15 Time (s) Time (s) Repetition no. 1 Repetition no. 1 Repetition no. 2 Repetition no. 2 Repetition no. 3 Repetition no. 3 (a) (b) Figure 10: Computed torques during test no. 1 without load for the three repetitions seen in Figure 8: (a) Motor 1; (b) Motor 2. an end-eƒector has been designed for a comfortable grasping 3. Laboratory Setup and Testing Modes of the user. Figure 3(b) shows the tracing point TP on the NURSE has been conceived and designed to solve all the NURSE end-eƒector. e linkage structure is composed of issues that have been mentioned in Introduction, giving the aluminum bars that have a thickness of 6 mm and a width of possibility to perform exercises useful for physical therapy or 25 mm. e mechanism structure weighs 2.6 kg, and it šts rehabilitation, for treatments of injuries or diseases, for into a box of 35 × 45 × 30 cm. More details of the mechanical prevention of injuries or diseases, or for physical exercising design of NURSE are explained in [32, 33]. e mechanism [32] (Figure 3(a)). can guide both right and left human arms on a plane within e proposed device is composed of a linkage structure a large workspace to follow whatever desired trajectory that is driven in planar movements by two actuators. Two [32, 33]. e planar linkage structure is characterized by wheels are used to support the NURSE structure (Figure light links for compact design, low-power consumption, and 3(b)). e used wheels have omnidirectional balls of stainless easy portability. e movements that can be performed by steel, and they can support a load of 25 kg each. In addition, NURSE involve the shoulder and elbow of a human arm in Torque (N-mm) Linear acceleration (m/s ) Torque (N-mm) 2 Linear acceleration (m/s ) Journal of Healthcare Engineering 7 (a) 0 28 10 12 14 16 Time (s) Repetition no. 1 Repetition no. 2 Repetition no. 3 Figure 11: Computed power consumption of test no. 1 without load for the three repetitions seen in Figure 8. (b) Load (c) Figure 12: A zoomed view of a NURSE end-eƒector with a load of 520 g. Figure 13: Some snapshots of test no. 1 with a load of 520 g during repetition no. 1 together with the trajectory obtained by image an independent way or in a coordinated motion. Since processing (in red): (a) the šrst sample position; (b) the second NURSE can perform several trajectories of diƒerent sizes, it sample position; (c) the third sample position. can be used by people of any age, anthropomorphic sizes, and anthropometric sizes, including children and elderly people as pointed out in [32, 33]. To test the performances and the behaviour of NURSE, some experiments have been carried out at LARM labora- tory in Cassino. A specišc layout has been designed to allow a satisfactory acquisition of the needed data (Figure 4). In Figure 4(a), it is possible to notice that the area can be divided in two subareas, namely, the mechanism area and the control area. e mechanism area includes the NURSE together with two cameras. One camera has been installed on the top of –200 –150 –100 –50 0 50 100 150 200 NURSE being planar to its workspace, while the other camera X (mm) has been installed in front of NURSE. Furthermore, an IMU Repetition no. 1 Repetition no. 3 (inertial measurement unit) sensor has been placed on the TP. Repetition no. 2 Programmed trajectory e control area consists of a laptop in which an interface sends the positions for the NURSE motors according to a selected Figure 14: Comparison between the programmed trajectory and arm exercise, the control unit, and the current-sensing modules the TP trajectories from three repetitions during test no. 1 with as in Figure 4(b). Each actuator is connected to a control a load of 520 g. Power consumption (watts) Y (mm) 64 8 Journal of Healthcare Engineering 0.2 0.2 0.15 0.15 0.1 0.1 0.05 0.05 0 0 –0.05 –0.05 0 5 10 15 0 5 10 15 Time (s) Time (s) Repetition no. 1 Repetition no. 1 Repetition no. 2 Repetition no. 2 Repetition no. 3 Repetition no. 3 (a) (b) Figure 15: Acquired linear acceleration during test no. 1 with a load of 520 g for the three repetitions seen in Figure 14: (a) X linear acceleration; (b) Y linear acceleration. 3000 3000 2000 2000 1000 1000 0 0 –1000 –1000 0 5 10 15 0 5 10 15 Time (s) Time (s) Repetition no. 1 Repetition no. 1 Repetition no. 2 Repetition no. 2 Repetition no. 3 Repetition no. 3 (a) (b) Figure 16: Computed torques during test no. 1 with a load of 520 g for the three repetitions seen in Figure 14: (a) Motor 1; (b) Motor 2. board that will generate the trajectories to reproduce the e placed IMU sensor on the TP can be used to measure selected exercise from the interface. e current-sensing module the angular displacement in terms of roll (θ), pitch (Φ), and is composed of two current sensors, and each sensor is yaw (ψ) and to acquire the linear acceleration along X, Y, and Z-axes as shown in Figure 5. connected to each motor. Finally, one emergency switch turns oƒ the motor amplišer, while the second turns oƒ the e two current sensors based on the Hall eƒect are used entire system. to compute the power consumption and check the behaviour e top camera has been used to track the movement of of each actuator. the TP to validate if the programmed trajectory is satisfac- e experiments are carried out following the ‡ow chart torily reproduced by NURSE; to do so, some markers (red shown in Figure 6. Before running a test, the device is set color circles) have been placed on the structure for the motion manually in the home position, the actuator is turned on, and tracking by image processing (Figure 5). e front camera the interface is initialized. After that, the exercise to be per- allows for an overview of the working area. formed is selected, the exercise trajectory is sent to the motor Torque (N-mm) Torque (N-mm) Linear acceleration (m/s ) Linear acceleration (m/s ) Journal of Healthcare Engineering 9 (a) 0 2 4 6 8 10 12 14 16 Time (s) Repetition no. 1 Repetition no. 2 Repetition no. 3 Figure 17: Computed power consumption during test no. 1 with a load of 520 g for the three repetitions seen in Figure 14. control board, and the test runs reproducing the desired task. (b) While the experiment is running, the data are acquired from the cameras and the sensors. When the exercise ends, the data are collected and checked to evaluate if there is any data discrepancy due to sensors or video acquisition failure. In such a case, the test is repeated; otherwise, the postprocessing stage starts and the device characterization is carried out to evaluate the performance of NURSE. Table 1 shows parameters of the tests that have been carried out in order to characterize the NURSE behaviour. e references trajectories of exercise nos. 1 and 2 in Figure 2 are used to carry out the tests. In test no. 1, NURSE performs exercise no. 1 during three repetitions. In test no. 2, NURSE (c) performs exercise no. 2 during three repetitions. Both tests Figure 18: Some snapshots of test no. 2 without load during are carried out without load and with a load of 520 g by using repetition no. 1 together with the trajectory obtained by image a velocity of 396 /s. e used load of 520 g is equivalent to processing (in red): (a) the šrst sample position; (b) the second 30% of the average weight of the forearm together with the sample position; (c) the third sample position. hand, and it has been considered enough for lab experi- ments. In both tests, the positions of the TP are programmed in the control (X, Y) as inputs. shows the trajectory obtained from the marker on the TP. After the acquisition, the positions of the TP (X , Y ) m m Figure 8 shows the trajectory programmed in the device and are obtained by image processing. e positions of the TP the trajectories obtained from the marker on the TP during are used to validate if the device is able to perform the repetition nos. 1, 2, and 3. As shown in Figure 8, the trajec- programmed trajectory. e linear accelerations of the TP tories obtained from the marker on the TP are close to the (a , a ) are acquired by the IMU sensor, and they can be used x y programmed one with a maximum deviation of 10 mm. is to evaluate the smoothness of the motion as an important deviation is related with the accuracy of the home position aspect for user safety. Using the acquired motor’s current, it since it is set manually. However, the repeatability deviation is possible to compute the torque of each motor (τ , τ ) and 1 2 between the trajectories performed by NURSE has a maxi- the power consumption (p) of NURSE to evaluate if the mum value of 3 mm for test no. 1 without load. actuators struggle while replicating the task. Figure 9 shows the linear accelerations acquired from the TP during test no. 1 when the device is unloaded for the three repetitions as seen in Figure 8. e linear accelerations 4. Test Results in X have a maximum value of 0.058 m/s and a minimum Test no. 1 has been carried without load during three repe- value of −0.015 m/s , and linear accelerations in Y have titions. Figure 7 shows three snapshots of the video while the a maximum value of 0.059 m/s and a minimum value of test is carried out during repetition no. 1. In addition, Figure 7 −0.015 m/s . e linear acceleration values in X and Y are Power consumption (watts) 10 Journal of Healthcare Engineering 800 0.2 0.15 0.1 0.05 –0.05 0 5 10 15 20 –150 –100 –50 5 100 150 200 Time (s) X (mm) Repetition no. 1 Repetition no. 1 Repetition no. 3 Repetition no. 2 Repetition no. 2 Programmed trajectory Repetition no. 3 Figure 19: Comparison between the programmed trajectory and (a) the TP trajectories from three repetitions during test no. 2 without load. 0.2 0.15 negligible, and it shows that the movement of the TP is smooth. e spikes in the linear acceleration are due to the 0.1 backlash of the wheels. Figure 10 shows the acquired motor torques during test 0.05 no. 1 when the device is unloaded for the three repetitions as seen in Figure 8. Motor 1 reaches a maximum magnitude of 2,071 N-mm, and Motor 2 reaches a maximum magnitude of 2,119 N-mm. e torque values conšrm that commercial servomotors can be used for NURSE motion. In addition, –0.05 the torques curves show a symmetrical behaviour between Motor 1 and Motor 2. 0 5 10 15 20 Figure 11 shows the power consumption of NURSE Time (s) without load during test no. 1. e power consumption Repetition no. 1 reaches a maximum value of 23.130 W. e power con- Repetition no. 2 sumption values conšrm that NURSE works with low- Repetition no. 3 power consumption when it is unloaded. Similarly, test no. 1 has been carried out with a load of (b) 520 g during three repetitions. Figure 12 shows a zoomed view of a NURSE end-eƒector with the load of 520 g. Some Figure 20: Acquired linear acceleration during test no. 2 without snapshots of the test with the acquired trajectory from the load for the three repetitions seen in Figure 19: (a) X linear ac- TP during repetition no. 1 are shown in Figure 13. Figure 14 celeration; (b) Y linear acceleration. shows the trajectories acquired from the TP during the three repetitions and the programmed one. When NURSE is loaded in test no. 1, the deviation between the trajectories have a maximum value of 0.041 m/s and a minimum value of acquired from the TP and the programmed one has a max- −0.039 m/s . erefore, the linear acceleration values in X and imum value of 13 mm. e deviation when NURSE is loaded Y are also negligible when the device is loaded. us, NURSE is 3 mm greater than the deviation when NURSE is unloaded. can reproduce the exercise of test no. 1 when it is loaded as However, this diƒerence is negligible, and it can also be re- smoothly as when it is unloaded. lated with the accuracy of the home position as mentioned When the device is loaded during test no. 1, the torque of above. It is important to notice that the repeatability deviation the Motor 1 reaches a maximum magnitude of 2,926 N-mm between the trajectories performed by NURSE has a maxi- and Motor 2 has a maximum magnitude of 3,217 N-mm mum value of 4.5 mm for test no. 1 with load. (Figure 16). As seen in Figure 16, the torque increases e linear accelerations acquired from the TP during test around 1,098 N-mm when the device is loaded with respect no. 1 with a load of 520 g are shown in Figure 15. e linear to the torque when it is unloaded as seen in Figure 10. accelerations in X have a maximum value of 0.036 m/s and However, the torque values conšrm that NURSE can also be a minimum value of −0.045 m/s , and linear accelerations in Y moved by commercial motors in the loaded condition. Y (mm) 2 2 Linear acceleration (m/s ) Linear acceleration (m/s ) 00 Journal of Healthcare Engineering 11 3000 3000 2000 2000 1000 1000 0 0 –1000 –1000 –2000 –2000 –3000 –3000 0 5 10 15 20 0 5 10 15 20 Time (s) Time (s) Repetition no. 1 Repetition no. 1 Repetition no. 2 Repetition no. 2 Repetition no. 3 Repetition no. 3 (a) (b) Figure 21: Computed torques during test no. 2 without load for the three repetitions seen in Figure 19: (a) Motor 1; (b) Motor 2. e power consumption when NURSE is loaded has a maximum value of 30.35 W for test no. 1 (Figure 17). us, it increased 7.220 W with respect to the power consumption when NURSE is unloaded. erefore, the NURSE low-power consumption characteristic remains. Test no. 2 has been carried out without load during three repetitions. In Figure 18 are shown some snapshots of NURSE when it is performing repetition no. 1 together with the trajectory acquired from the TP. Figure 19 shows the trajectories acquired from the TP during repetition nos. 1, 2, 10 and 3 and the programmed one. As seen in Figure 19, the trajectories performed by the NURSE to trace the number 8 are close to the programmed one with a maximum deviation of 16 mm. e deviation in test no. 2 is greater than the 0 5 10 15 20 deviation in test no. 1 since the 8 shape has more changes in Time (s) direction and is being more complex to perform than the trajectory for horizontal shoulder ‡exion. On the contrary, Repetition no. 1 the backlash of NURSE wheels can aƒect the motion more Repetition no. 2 Repetition no. 3 when it has several changes of direction than when it maintains a same direction. However, the repeatability Figure 22: Computed power consumption during test no. 2 deviation between the trajectories performed by NURSE has without load for the three repetitions seen in Figure 19. a maximum value of 8.22 mm for test no. 2 without load. Despite the fact that the wheels backlash can aƒect the trajectory shape during test no. 2, the linear accelerations torques reached by the motors without load during test no. 2 acquired from the TP show that the motion remains smooth remain in the same range than the torques in test no. 1 as seen in Figure 20, where the linear accelerations in X have without load. erefore, it shows that when NURSE is a maximum value of 0.062 m/s and a minimum value of unloaded, it requires a similar force to perform the trajectory for horizontal shoulder ‡exion than it requires to perform −0.008 m/s and linear accelerations in Y have a maximum 2 2 value of 0.095 m/s and a minimum value of −0.015 m/s . As the number 8. e latter is conšrmed also by the power seen in Figure 20, the linear acceleration values during test consumption that presents a maximum value of 24.510 W no. 2 have remained in the same range than the linear (Figure 22). e power consumption during test no. 2 accelerations during test no. 1. without load increases only 1.380 W with respect to the value Figure 21 shows the torque required by Motor 1 and in test no. 1 without load. erefore, NURSE maintains low- Motor 2 during test no. 2 without load. Motor 1 reaches power consumption while tracing the number 8. a maximum torque of 2,264 N-mm, and Motor 2 reaches Similarly, test no. 2 has been carried out during three a maximum torque of 1,840 N-mm. As seen in Figure 21, the repetitions by using a load of 520 g. Figure 23 shows some Torque (N-mm) Torque (N-mm) Power consumption (watts) 12 Journal of Healthcare Engineering (a) (b) (c) Figure 23: Some snapshots of test no. 2 with a load of 520 g during repetition no. 1 together with the trajectory obtained by image processing (in red): (a) the šrst sample position; (b) the second sample position; (c) the third sample position. –150 –100 –50 0 50 100 150 200 X (mm) Repetition no. 1 Repetition no. 3 Repetition no. 2 Programmed trajectory Figure 24: Comparison between the programmed trajectory and the TP trajectories from three repetitions during test no. 2 with a load of 520 g. 0.2 0.2 0.15 0.15 0.1 0.1 0.05 0.05 0 0 –0.05 –0.05 0 5 10 15 20 0 5 10 15 20 Time (s) Time (s) Repetition no. 1 Repetition no. 1 Repetition no. 2 Repetition no. 2 Repetition no. 3 Repetition no. 3 (a) (b) Figure 25: Acquired linear acceleration during test no. 2 with a load of 520 g for the three repetitions seen in Figure 24: (a) X linear acceleration; (b) Y linear acceleration. Linear acceleration (m/s ) Y (mm) Linear acceleration (m/s ) Journal of Healthcare Engineering 13 4000 4000 2000 2000 0 0 –2000 –2000 0 5 10 15 20 0 5 10 15 20 Time (s) Time (s) Repetition no. 1 Repetition no. 1 Repetition no. 2 Repetition no. 2 Repetition no. 3 Repetition no. 3 (a) (b) Figure 26: Computed torques during test no. 2 with a load of 520 g for the three repetitions seen in Figure 24: (a) Motor 1; (b) Motor 2. snapshots of NURSE when it is performing repetition no. 1. 50 As seen in Figure 24, the trajectories acquired from the TP during repetition nos. 1, 2, and 3 are close to the pro- grammed one with a maximum deviation of 21 mm. However, the repeatability deviation between the trajec- tories performed by NURSE has a maximum value of 11.94 mm for test no. 2 with load. Although the wheels backlash introduces deviation in the motion to perform the trajectories, the linear accelerations acquired from the TP show that NURSE motion continues to be smooth with the linear accelerations in X having a maximum value of 2 2 0.062 m/s and a minimum value of −0.056 m/s and linear accelerations in Y having a maximum value of 0.075 m/s and a minimum value of −0.026 m/s (Figure 25). As seen 0 in Figures 9, 15, 20, and 25, the linear accelerations ac- 0 5 10 15 20 quired from the TP are maintained around the same range. Time (s) erefore, it can be said that NURSE can reproduce the Repetition no. 1 trajectories with a smooth motion during test nos. 1 and 2 Repetition no. 2 with and without load. Repetition no. 3 e torque of Motor 1 reached a maximum magnitude Figure 27: Computed power consumption during test no. 2 with of 3,527 N-mm, and the torque of Motor 2 reached a load of 520 g for the three repetitions seen in Figure 24. a maximum magnitude of 3,464 N-mm, Figure 26. In test no. 2, the torque increases 310 N-mm with respect to the torque when the device is loaded in test no. 1 (Figure 16). 5. Conclusions It can be said that NURSE needs more force when per- forming the exercise of test no. 2 than when performing the NURSE, a device for arm motion assistance, is presented exercise of test no. 1 both in loaded conditions. However, with an experimental characterization. NURSE can assist the the torque values are in a range that always can be reached motion of both right and left human arms during a re- by commercial servomotors. e power consumption has a habilitation therapy or during the arm exercise for elderly maximum value of 37.080 W as seen in Figure 27. ere- people. e NURSE behaviour has been characterized by fore, the power consumption increased 6.730 W with re- performing tests of several exercises for upper limb re- spect to the obtained values during test no. 1 with a load as habilitation or training, whereas in this paper, two signiš- seen in Figure 17. However, NURSE continues to have low- cant ones have been discussed. e tests have successfully power consumption also to trace the number 8 in loaded been carried out without and with load by looking at tra- conditions. jectory tracking, linear acceleration, torque, and power Torque (N-mm) Torque (N-mm) Power consumption (watts) 14 Journal of Healthcare Engineering http://www.who.int/healthinfo/global_burden_disease/2004_ consumption. )e examined trajectories during the tests report_update/en/. show that NURSE is able to perform a trajectory near to the [8] D. Knudson, Fundamentals of Biomechanics, Springer, Chico, programmed one with a minimum deviation of 16 mm when CA, USA, 2nd edition, 2007. it is unloaded and a maximum deviation of 21 mm in the [9] W. E. Prentice, Rehabilitation Techniques in Sports Medicine, loaded condition. )e trajectories performed by NURSE McGraw-Hill Education, Boston, MA, USA, 4th edition, 2003. during reported test no. 1 have a satisfactory maximum [10] I. Kapandji, >e Physiology of the Joints, Vol. 3, Churchill repeatability deviation of 3 mm when it is unloaded and Livingstone, New York, NY, USA, 2008. 4.5 mm when it is loaded. )e trajectories performed by [11] L. Ombregt, A System of Orthopaedic Medicine, Churchill NURSE during reported test no. 2 have a satisfactory Livingstone Elsevier, Edinburgh, Scotland, 3rd edition, 2013. maximum repeatability deviation of 8.22 mm between them [12] R. Riener, T. Nef, and G. Colombo, “Robot-aided neuro- when it is unloaded and 11.94 mm when it is loaded. )e rehabilitation of the upper extremities,” Medical and Biological linear accelerations during test nos. 1 and 2 have been Engineering and Computing, vol. 43, no. 1, pp. 2–10, 2005. [13] C. G. Burgar, P. S. Lum, P. C. Shor, and H. F. Machiel Van der successfully measured within a satisfactory range of mini- 2 2 Loos, “Development of robots for rehabilitation therapy: the mum −0.008 m/s and maximum 0.095 m/s with a smooth Palo Alto VA/Stanford experience,” Journal of Rehabilitation NURSE motion. )e NURSE motors operated with a max- Research and Development, vol. 37, no. 6, pp. 663–673, 2000. imum torque of 3,527 N-mm occurring during test no. 2 [14] P. S. Lum, C. G. Burgar, P. C. Shor, M. Majmundar, and with load as a feasible result for commercial servomotors. M. Van der Loos, “Robot-assisted movement training com- NURSE worked with low-power consumption without and pared with conventional therapy techniques for the re- with a load. )e maximum power consumption has been habilitation of upper-limb motor function after stroke,” 37.080 W and it has been reached during test no. 2 with load. Archives of Physical Medicine and Rehabilitation, vol. 83, )e experimental results show that NURSE is capable of no. 7, pp. 952–959, 2002. reproducing successfully different exercises with a smooth [15] S. L. Mediatric, Figure 8 System, Ref. NC99110, 2016, http://www. motion and a proper low-power consumption. mediatric.com/productos.aspx?tipo�1&pag�77&idprod�632, in Spanish. [16] I. P. Corona-Acosta and E. Castillo-Castaneda, “Dimensional Conflicts of Interest synthesis of a planar parallel manipulator applied to upper limb rehabilitation,” in Multibody Mechatronic Systems, )e authors declare that they have no conflicts of interest. Mechanisms and Machine Science, H. M. E. Ceccarelli, Ed., vol. 25, pp. 443–452, Springer, Cham, Switzerland, 2015. [17] AliMed, Ergo Arm Skate, Ref. 8591, 2018, https://www.alimed. Acknowledgments com/ergo-arm-skate.html. [18] Hocoma, ArmeoPower, 2017, https://www.hocoma.com/ Betsy Dayana Marcela Chaparro-Rico thankfully acknowl- solutions/armeo-power/. edges CONACYT for the financial support for her Ph.D. [19] S. J. Ball, I. E. Brown, and S. H. Scott, “A planar 3DOF robotic studies at Instituto Politecnico ´ Nacional in Mexico with exoskeleton for rehabilitation and assessment,” in Proceedings a program for a double Ph.D. degree at the University of of the 29th Annual Conference of the IEEE EMBS, pp. 4024– Cassino and Southern Lazio. 4027, Cite´ Internationale, Lyon, France, August 2007. [20] Y. Mao and S. Kumar-Agrawal, “Design of a cable-driven arm exoskeleton (CAREX) for neural rehabilitation,” IEEE References Transactions on Robotics, vol. 28, no. 4, pp. 922–931, 2012. [21] C. T. O’Neill, N. S. Phipps, L. Cappello, S. Paganoni, and [1] National Stroke Association, Mobility after Stroke, National Stroke Association, Centennial, CO, USA, 2018, https://www. C. J. Walsh, “A soft wearable robot for the shoulder: design, characterization, and preliminary testing,” in Proceedings of stroke.org/sites/default/files/resources/NSA_Mobility_brochure.pdf. [2] I. P. Corona-Acosta, “Development of a mechatronic device the 2017 International Conference on Rehabilitation Robotics (ICORR), pp. 1672–1678, QEII Centre, London, UK, July for rehabilitation of the upper extremity (shoulder-elbow- ´ 2017. wrist),” Master thesis, Instituto Politecnico Nacional- CICATA UNIDAD QUERETARO, Santiago de Queretaro, ´ [22] H. Bian, Z. Chen, H. Wang, and T. Zhao, “Mechanical design of EFW Exo II: a hybrid exoskeleton for elbow-forearm-wrist QRO, Mexico, 2015, in Spanish. [3] T. M. Skirven, A. L. Osterman, J. Fedorczyk, and P. C. Amadio, rehabilitation,” in Proceedings of the 2017 International Conference on Rehabilitation Robotics (ICORR), QEII Centre, Rehabilitation of the Hand and Upper Extremity, Vol. 2, Elsevier Health Sciences, Toronto, ON, Canada, 6th edition, 2011. London, UK, July 2017. [23] T. Nef, M. Guidali, and R. Riener, “ARMin III—arm therapy [4] S. J. Hall, Basic Biomechanics, McGraw-Hill, New York, NY, USA, 6th edition, 2012. exoskeleton with an ergonomic shoulder actuation,” Applied [5] J. Rippe, S. McCarthy, and W. M. Abbott, >e Joint Health Bionics and Biomechanics, vol. 6, no. 2, pp. 127–142, 2009. Prescription: 8 Weeks to Stronger, Healthier, Younger Joints, [24] J. Annisa, S. Mohamaddan, M. S. Jamaluddin et al., “De- Ballantine Books, New York, NY, USA, 2002. velopment of upper limb rehabilitation robot prototype for [6] P. Maciejasz, J. Eschweiler, K. Gerlach-Hahn, A. Jansen-Troy, home setting,” in Proceedings of the 5th Brunei International and S. Leonhardt, “A survey on robotic devices for upper limb Conference on Engineering and Technology (BICET 2014), pp. 1–6, Bandar Seri Begawan, Brunei, November 2014. rehabilitation,” Journal of NeuroEngineering and Re- habilitation, vol. 11, no. 3, pp. 1–29, 2014. [25] D. Campolo, F. Widjaja, and J. Klein Hubert, “An apparatus for upper body movement,” US Patent Application No. [7] World Health Organization (WHO), >e Global Burden of Disease: 2004 Update, WHO, Geneva, Switzerland, 2008, US20150302777A1, USPTO, Alexandria, VA, USA, 2015. Journal of Healthcare Engineering 15 [26] H. I. Krebs, N. Hogan, M. L. Aisen, and B. T. Volpe, “Robot- aided neurorehabilitation,” IEEE Transactions on Re- habilitation Engineering, vol. 6, no. 1, pp. 75–87, 1998. [27] Bionik Laboratories, Inmotion Arm, Bionik Laboratories, Tor- onto, ON, Canada, 2018, http://bionikusa.com/healthcarereform/ upper-extremity-rehabilitiation/inmotion2-arm/. [28] P. Leconte and R. Ronsse, “Performance-based robotic as- sistance during rhythmic arm exercises,” Journal of Neuro- Engineering and Rehabilitation, vol. 13, no. 1, pp. 1–16, 2016. [29] B. D. M. Chaparro-Rico, E. Castillo-Castaneda, M. Ceccarelli, and D. Cafolla, “Design and test of therapy exercise for human arms,” in Proceedings of the 5th International Workshop on Medical and Service Robots (MESROB 2016), Graz, Austria, July 2016. [30] S. Chatzis, D. Korkinof, and Y. Demiris, “A quantum- statistical approach toward robot learning by demonstra- tion,” IEEE Transactions on Robotics, vol. 28, no. 6, pp. 1371–1381, 2012. [31] Microsoft, User’s Manual Microsoft: Kinect SDK, 2015, https:// msdn.microsoft.com. [32] B. D. M. Chaparro-Rico, D. Cafolla, M. Ceccarelli, and E. Castillo-Castañeda, “Device for arm motion assistance,” Italian Application Patent No. IT102016000107499, October [33] B. Chaparro-Rico, D. Cafolla, M. Ceccarelli, and E. Castillo- Castaneda, “Design and simulation of an assisting mechanism for arm exercises,” in Advances in Italian Mechanism Science, Mechanisms and Machine Science, G. Boschetti and A. Gasparetto, Eds., vol. 47, pp. 115–123, Springer, Cham, Switzerland, 2017. 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 of Healthcare Engineering – Hindawi Publishing Corporation
Published: Jul 3, 2018
You can share this free article with as many people as you like with the url below! We hope you enjoy this feature!
Read and print from thousands of top scholarly journals.
Already have an account? Log in
Bookmark this article. You can see your Bookmarks on your DeepDyve Library.
To save an article, log in first, or sign up for a DeepDyve account if you don’t already have one.
Copy and paste the desired citation format or use the link below to download a file formatted for EndNote
Access the full text.
Sign up today, get DeepDyve free for 14 days.
All DeepDyve websites use cookies to improve your online experience. They were placed on your computer when you launched this website. You can change your cookie settings through your browser.