Comprehensive Example Analysis on Main Technology of Exoskeletal Robot System for Upper Limb Rehabilitation

Abstract:

Many medical treatments need to confront problem of upper limbs’ rehabilitation. Major function of exoskeletal robot system for upper limbs rehabilitation is to assist patient to carry out upper limbs’ rehabilitation training. Main technology of exoskeletal robot system for upper limbs rehabilitation includes design of mechanical structure of exoskeletal robot, design of control system of exoskeletal robot and implemention of data and information transmission between exoskeletal robot and upper limbs of human body. Currently implemention of data and information transmission rely mainly on methods of acquiring sEMG signal and force feedback. Reviewing and analyzing the specific technical examples will be important way in improving and upgrading the technology in future. Current development trend of main technology is to advance degree of comfort, accuracy and effect in rehabilitation training. Further development trend of main technology henceforth is people-oriented, strengthening effect, improving efficiency, optimizing structure and reducing energy consumption that follow claims of medical treatment.

• Periodical: Applied Mechanics and Materials (Volume 623)
• Pages: 211-218
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.623.211
• Online since: August 2014
• Authors: Ming Shi Chen*, Liang Zhao
• Keywords: Control System, Data Transmission, Example, Exoskeletal Robot System, Mechanical Structure, Medical Treatment, Upper Limb Rehabilitation

References

[10] Other rehabilitation robot systems like Haptic Walker.

[11] and Haptic Arm Exoskeleton.

[12] adopt impedance control method made up of position control loop and force control loop to provide rehabilitation training in horizontal plane to patient. Impedance control strategy makes robot smooth to human body, which not only meets requirement of patient's security, but also helps robot executing various patterns of rehabilitation training through adjusting relevant parameters of control law. Various existing force feedback data gloves encounter following problems. First, all of them can only bring active force to bear on direction of finger pulp of manpower fingertips and can't realize force feedback from direction of manpower dorsal digital. Second, in process of force feedback, force from driver acts on knuckles of fingers, which affect telepresence of constraint space. Third, in process of force feedback knuckles of manpower fingers should be fixed with gloves, so patients will feel strain and telepresence of free space will be disrupted while manpower fingers' bending and stretching. Last, most existing exoskeletal force feedback data gloves' mechanical structures adopt rope gearing which must be equipped with spring tube, so that problems of friction and delay that make system control complicated are brought. 4. 2. Data and Information Transmission Relying on Method of acquiring sEMG signal At the present stage research on sEMG signal can be divided into two aspects in principle. One hand is to analyze and research physiology information contained in sEMG signal and to set up relationship between physiological process and biochemical process inside muscle and change of EMG, which are applied to diagnosis of neuromuscular disorder and assessment of motor function in clinical medicine, ergonomic analysis of muscular work in field of ergonomics, and fatigue assessment and analysis on sports technique in sports science[13-15]. The other hand is to recognize various limb movements through recognizing correspondent sports information contained in correspondent sEMG signal. The method is widely used in field of HCI, clinical rehabilitation and so on. Typical methods of analyzing on sEMG signal include time-domain analysis, frequency-domain analysis and time-frequency analysis. For example, eigenvalues such as AEMG and RMS in time-domain analysis and MPF and MF in frequency-domain analysis can be ROC of evaluation of muscle fatigue[16, 17]. In recent years field of nonlinear analysis on sEMG signal is being more and more researched and developed. Here's an example of automatic passive training program of exoskeletal robot system for upper limbs rehabilitation basing on sEMG signal recognition as in figure 11.

[18] In order to control rehabilitation robot to execute predetermined trajectory movement training for affected extremity, the robot system need to understand exercise intention of hemiplegic patient whose motor function of unilateral limb is damaged by recognizing sEMG signal of upper limb movement on uninjured side. The automatic passive training program basing on sEMG signal recognition includes 5 steps as following: Step 1 is determining occurrence of action in which 4-channel sEMG signal is monitored all along and synchronally sampled after confirming effective movement. Step 2 is analyzing 4-channel sEMG and actualizing feature extraction in which feature vector of sEMG of each channel is extracted from a huge amount of 4-channel sEMG signals of upper limb movement on uninjured side through effective analytical approach and dimension-reduction is actualized. Step 3 is encoding the set of action goal, making feature vector of sEMG of specific action and code of the specific action as input and output respectively and supervising and training pattern classifier with teaching signal until pattern classifier is convergent in which supervised learning of pattern classifier is actualized. Step 4 is inputting extracted feature vector of 4-channel sEMG into trained pattern classifier in order to recognize current type of movement in which pattern recognition is actualized. Step 5 is executing actions in which exoskeletal robot system for upper limbs rehabilitation executes predetermined trajectory passive rehabilitation training according to result of action recognition. Fig. 11. An example of automatic passive training program basing on sEMG signal recognition 5. DISCUSSION AND CONCLUSION There are still some problems in the practice application to be solved. Design of connection and fixation between exoskeleton and upper limbs of human body currently need to be improved more reasonably, by which problems as poor blood circulation or unnatural muscle movement in fixation position would be to avoid and positioning accuracy of exoskeleton would be improved to a large extent. In the case of patient's slow movement current exoskeletal robot systems for upper limbs rehabilitation have well speed of response in experimental test results, but in the case of patient's rapid or complex movement the systems generally can't reach the same speed of response, which needs improvement from more excellent design and technology of exoskeletal robot. And in data and information transmission for obtaining patient's intention of motion, method of acquiring sEMG signal is easy to be disturbed and restricted by acquiring environment, and method of force feedback can't deal with patient's rapid or complex movement as its own hysteresis quality. It's in need of developing some more perfect technology for data and information transmission of exoskeletal robot systems for upper limbs rehabilitation in the future. Moreover, there is still very large promotion space in flexibility, security, environmental protection and degree of comfort of the exoskeletal robot. Further development trend of main technology of exoskeletal robot system for upper limbs rehabilitation henceforth is people-oriented, strengthening effect, improving efficiency, optimizing structure and reducing energy consumption that follow claims of medical treatment. REFRENCES.

[1] Molewijk A C, Stiggelbout A M, Otten W, et al., Implicit normativity in evidence-based medicine: A plea for integrated empirical ethics research, Health Care Analysis, Vol. 11, No. 1, 2003, pp.89-92.

[2] Gary G E, Pinson L A, Evidence-based medicine and psychiatric practice, Psychiatric Quarterly, Vol. 74, No. 4, 2003, pp.387-99.

DOI: 10.1023/a:1026091611425

[3] Kiguchi K, Tanaka T, Watanabe, et al., Designed and control of an exoskeleton system for human upper-limb motion assist Advanced Intelligent Mechatronics (AIM 2003), 2003 IEEE/ASME International Conference, 2003, pp.926-31.

DOI: 10.1109/aim.2003.1225466

[4] Kousidou S, Tsagarakis N G and Caldwell D G, Evaluation of a 'soft' exoskeleton for rehabilitation and physiotherapy of the upper limb, Autonomous Robots, Vol. 15, No. 1, 2003, pp.21-23.

[5] Sugar T G, He Jiping, Koeneman E J, et al., Design and control of RUPERT: A device for robotic upper extremity repetitive therapy, IEEE Trans. on Rehabilitation Engineering, Vol. 15, No. 3, 2007, pp.336-46.

DOI: 10.1109/iembs.2005.1616077

[6] Chen W B, Human upper limb kinematics and anthropomorphic robot kinematic design and motion planning 2012 Doctoral Dissertation, Huazhong University of Science and Technology, (2012).

[7] Xing K X, The research on some key technologies of hand rehabilitation robot system, 2010 Doctoral Dissertation, Huazhong University of Science and Technology. (2010).

[8] Kousidou S, Tsagarakis N G, Caldwell D G, et al., Assistive exoskeleton for task based physiotherapy in 3-dimensional space, Proc. 1st IEEE International Conference on Biomedical Robotics and Biomechatronics, 2006, pp.266-71.

DOI: 10.1109/biorob.2006.1639097

[9] Tsagarakis N G, Caldwell D G, Development and control of a 'soft-actuated' exoskeleton for use in physiotherapy and training, Automonous Robots, Vol. 15, No. 1, 2003, pp.21-33.

[10] Krebs H I, Hogan N, Aisen M L, et al., Robot-aided neurorehabilitation, IEEE Transactions on Rehabilitation Engineering, Vol. 6, No. 1, 1998, pp.75-87.

DOI: 10.1109/86.662623

[11] Schmidt H, Hesse S, Werner C, et al., Upper and lower extremity robotic devices to promote motor recovery after stroke-recent developments, Proc. 26th IEEE International Conference on Engineering in Medicine and Biology Society, 2004, pp.4825-28.

DOI: 10.1109/iembs.2004.1404335

[12] Gupta A, Malley A K, Design of a haptic arm exoskeleton for training and rehabilitation, IEEE Transactions on Mechatronics, Vol. 11, No. 3, 2006, pp.280-89.

DOI: 10.1109/tmech.2006.875558

[13] Carary M A, Carnaby G D, Groher M E, et al., Functional benefits of dysphagia therapy using adjunctive sEMG biofeedback, Dysphagia, Vol. 19, No. 3, 2004, pp.160-64.

DOI: 10.1007/s00455-004-0003-8

[14] Yang H C, Wang J, Wang D M, Characteristics of surface electromyography single change based on isokinetic muscle fatigue, China Sport Science, Vol. 25, No. 7, 2005, pp.16-20.

[15] Zhang H H, Wang J, Yang H C, Surface electromyographic dynamics during biceps brachii eccentric and concentric contractions, Journal of Zhejiang University(Engineering Science), Vol. 40, No. 11, 2006, pp.1937-41.

[16] Merletti R, Conte L R, Orizio C, Indices of muscle fatigue, Journal of Electromyography and Kinesiology, Vol. 1, No. 1, 1991, pp.20-33.

DOI: 10.1016/1050-6411(91)90023-x

[17] Farina D, Merletti R, Comparison of algorithms for estimation of EMG variables during voluntary isometric contractions, Journal of Electromyography and Kinesiology, Vol. 10, No. 5, 2000, pp.337-49.

DOI: 10.1016/s1050-6411(00)00025-0

[18] Li Q L, Study on sEMG based exoskeletal robot for upper limbs rehabilitation, 2009 Doctoral Dissertation, Harbin Institute of Technology, (2009).