an exoskeletal robot for human shoulder joint motion ... · shoulder joint. in thispaper, we also...

IEEE/ASME TRANSACTIONS ON MECHATRONICS, VOL. 8, NO. 1, MARCH 2003 125

An Exoskeletal Robot for Human Shoulder JointMotion Assist

Kazuo Kiguchi, Member, IEEE, Koya Iwami, Makoto Yasuda, Keigo Watanabe, Member, IEEE, andToshio Fukuda, Fellow, IEEE

Abstract—We have been developing exoskeletal robots in orderto assist the motion of physically weak persons such as elderly per-sons or handicapped persons. In our previous research, a prototypeof a two degree of freedom exoskeletal robots for shoulder joint mo-tion assist have been developed since the shoulder motion is espe-cially important for people to take care of themselves in everydaylife. In this paper, we propose an effective fuzzy-neuro controller, amoving mechanism of the center of rotation (CR) of the shoulderjoint of the exoskeletal robot, and intelligent interface in order torealize a practical and effective exoskeletal robot for shoulder jointmotion assist. The fuzzy-neuro controller enables the robot to as-sist any person’s shoulder motion. The moving mechanism of theCR of the robot shoulder joint is used to fit the CR of the robotshoulder joint to that of the physiological human shoulder jointduring the shoulder motion. The intelligent interface is realized byapplying a neural network and used to cancel out the effect thehuman subject’s arm posture change. The effectiveness of the pro-posed method has been evaluated by experiment.

Index Terms—Electromyogram (EMG) signals, exoskeleton,human motion assist, soft computing.

I. INTRODUCTION

RECENT progress in robotics and mechatronics technologybrings a lot of benefits not only in industries, but also in

welfare and medicine. We have been developing exoskeletalrobots [1]–[3] in order to assist the motion of physically weakpersons such as elderly persons or handicapped persons. Itis important that such physically weak people are able totake care of themselves in the aging society. The exoskeletalrobots [4]–[7], which are sometimes called as exoskeletons,power suits, man amplifiers, man magnifiers, or power assistsystems, have been mainly studied for the purpose of militaryor industry use from the early 1960s. Since the design conceptis different, these robots were not suitable for physically weakpersons using in everyday life. On the other hand, activeorthotic systems [8], [9], which are similar to the exoskeletalrobots, also have been studied for the purposed of welfareand medicine from the 1960s. In order to use these systems,however, the users had to learn how to control the systemsbecause of the primitiveness of their controllers. In this paper,we propose a two degrees of freedom (DOF) exoskeletal robot

Manuscript received November 1, 2002; revised December 21, 2002. Rec-ommended by Technical Editor T. Nakamura. The work was supported by theTateisi Science and Technology Foundation, Japan under Grant 1021006.

K. Kiguchi, K. Iwami, M. Yasuda , and K. Watanabe are with the Departmentof Advanced Systems Control Engineering, Saga University, Saga 840-8502,Japan (e-mail: [email protected]).

T. Fukuda is with the Department of Micro System Engineering, Nagoya Uni-versity, Chikusa-ku, Nagoya 464-8603, Japan.

Digital Object Identifier 10.1109/TMECH.2003.809168

and its control method for automatic shoulder motion assistsince human shoulder joints are involved in a lot of motion ineveryday life. The proposed exoskeletal robot is a modifiedversion of the previously proposed 2-DOF exoskeletal robotprototype [3]. The architecture of the robot and the controllerare newly designed in this paper. The proposed exoskeletalrobot is automatically activated based on the human subject’selectromyogram (EMG) signals which directly reflect themuscle activity levels of human subject. The EMG signals areimportant information to understand how the human subjectintends to move. Consequently, the EMG signals can be usedas input information for the robotic systems [10]–[12]. Forthe exoskeletal robot in this study, seven kinds of the EMGsignals from the shoulder muscles of the human subject aswell as the shoulder joint angles are used as input information.Thus the exoskeletal robot is able to assist the motion of thehuman subject effectively by applying his/her EMG signalsas main input signals to the robot. Even though the EMGsignals contain very important information, it is not very easyto predict the shoulder motion from the EMG signals in a shorttime since many muscles are involved in the motion [13], [14].Furthermore, it is difficult to obtain the same EMG signal forthe same motion even from the same person since the EMGsignal is a biologically generated signal. Moreover, the levelof the EMG signals might be much different between persons.Therefore, the robot controller must have on-line adaptationability to the physiological condition of each human subject ifwe apply the exoskeletal robots to several persons [15], [16].In order to cope with this problem, a fuzzy-neuro controller,which is able to adapt itself to the physiological condition ofeach human subject on-line, is proposed for the controller ofthe exoskeletal robot. The physiological control of the robotcan be realized with this control method.

On the other hand, the mechanism of the prototype of the ex-oskeletal robot in our previous study was too simple in com-parison with that of the human shoulder. In addition to this,it is impossible to set the center of rotation (CR) of the robotshoulder joint is the same as that of the human shoulder joint(glenohumeral joint) since it is located inside of the human body.Furthermore, human shoulder complex provides 7-DOF for thearm movement since shoulder complex consists of the scapula,clavicle, and humerus and moves conjointly [17]. Consequently,the CR of the human shoulder joint is dislocated according tothe shoulder motion. Since the upper arm of the human sub-ject is almost fixed to the arm holder of the exoskeletal robotin our system, the subject must move his/her body instead ofhis/her upper arm to adjust the location of the CR of the shoulder

1083-4435/03$17.00 © 2003 IEEE

126 IEEE/ASME TRANSACTIONS ON MECHATRONICS, VOL. 8, NO. 1, MARCH 2003

Fig. 1. Architecture of the exoskeletal robot.

joint. Consequently, the human body in the exoskeletal robotwas sometimes forced to move back and forth or left and rightaccording to the shoulder motion since the CR of the robotshoulder joint was fixed and different from that of the humanshoulder joint.

In this paper, we also propose a mechanism of the moving CRof the shoulder joint of the 2-DOF exoskeletal robot for shouldermotion assist in order to cancel out the ill effects caused bythe position difference of the CR between the robot shoulderjoint and the human shoulder joint during the shoulder motion.The proposed mechanism makes the CR of the robot shoulderjoint mechanically move in accordance with the shoulder jointmotion. The linkwork mechanism has been applied to realizethe proposed mechanism. By the effect of the proposed shouldermechanism, the generated shoulder motion of the human subjectbecomes smooth like the physiological shoulder motion.

In our previous study [3], the arm posture of the human sub-ject was supposed to be the same at all times. Even though thehuman subject tries to perform the same shoulder motion, how-ever, the amount of the EMG signals from the shoulder musclesvaries [20] if the arm posture is changed since the dispositionof the shoulder muscles are changed [18], [19]. Therefore, thearm posture has to be taken into account in order to carry out thereliable shoulder motion assist. In this paper, we propose intel-ligent interface between the human subject and the fuzzy-neurocontroller to cancel out the effect caused by subject’s arm pos-ture difference. In this method, the fuzzy-neuro controller is ad-justed instantly by the intelligent interface in accordance withthe human subject’s arm posture. The intelligent interface is re-alized by applying a neural network.

The effectiveness of the proposed exoskeletal robot and itscontrol method has been evaluated by experiment with humansubjects.

II. EXOSKELETAL ROBOT

The architecture of the exoskeletal robot is shown in Fig. 1.The exoskeletal robot consists of a frame, two main links, anarm holder, two dc motor [Harmonic Drive System Company],

drive wires, wire tension sensors (strain gauges), and the mech-anism of the moving CR of the shoulder joint. The exoskeletalrobot worn by a human subject is supposed to assist the subject’sshoulder joint motion (flexion-extension and abduction-adduc-tion motions as shown in Fig. 2) by manipulating the subject’supper arm with the arm holder, which is fixed on the slider onthe link-2. The manipulation of the subject’s upper arm is car-ried out by controlling the arm holder motion with dc motorsvia driving wires. The inside of the arm holder is covered by anair cushion in which air pressure is adjustable to fit any size ofupper arm. The flexibility of the air cushion softens the motiondifference of the robot and the human subject caused by the dif-ference of the CR of the shoulder joints. Considering the factthat many physically weak persons use a wheel chair, the heavyparts of the proposed exoskeletal robot (i.e., the dc motors) areinstalled in the chair, and the other parts of the exoskeletal robotare directly attached to the human subject.

Human shoulder joint (glenohumeral joint) consists ofmany muscles such as deltoid, biceps, triceps, pectoralismajor, infraspinatus, and teres major, and moves in 3-DOF(flexion-extension, abduction-adduction, and internal-externalrotation). However, human shoulder complex provides 7-DOFfor the arm movement since shoulder complex consists of thescapula, clavicle, and humerus and moves conjointly [17]. Inthe case of shoulder flexion motion, the clavicle rotates about ananteroposterior axis at the sternoclavicular joint. The rotationresults in the elevation and translation (to medial direction) ofthe acromioclavicular joint with regard to the sternoclavicularjoint [21]. The translation of the sternoclavicular joint results inthe translation of the glenohumeral joint to the same direction.Therefore, the CR of the glenohumeral joint is dislocatedaccording to the shoulder motion.

Since the human upper arm is almost fixed in the arm holderof the robot, the relative distance between the arm holder and theCR of the human shoulder joint is almost constant. Therefore,the distance between the arm holder and the CR of the robotshoulder joint must be moderately adjusted in accordance withthe shoulder motion, in order to cancel out the ill effects causedby the position difference of the CR between the robot shoulderand the human shoulder.

KIGUCHI et al.: EXOSKELETAL ROBOT FOR HUMAN SHOULDER MOTION ASSIST 127

(a)

(b)

(c)

Fig. 2. Assisted shoulder joint motion.

The proposed mechanism of the moving CR of the robotshoulder, which consists of links and a slider as shown in Fig. 3,is installed between the link-1 and the link-2 of the robot.The motion of the proposed mechanism is depicted in Fig. 4.The joint between the link-1 and the link-2 (i.e., the shoulderjoint of the exoskeletal robot) is supposed to be located atjust behind the armpit of the human subject. The proposedmechanism makes the CR of the robot shoulder joint movebehind (farther position from the arm holder) in accordancewith the shoulder vertical flexion angle in the case of verticalflexion motion, and move inward (closer position to the armholder) in accordance with the shoulder horizontal extension

Fig. 3. Proposed linkwork mechanism (abducted position).

(a)

(b)

Fig. 4. Motion of the proposed linkwork mechanism.

angle in the case of horizontal extension motion. The linkworkmechanism has been applied to realize the proposed mecha-nism. In the case of shoulder vertical flexion-extension motion,the link-2 is vertically rotated with respect to the joint betweenthe link-1 and link-2. As the link-2 rotates vertically, the ad-ditional link (the link for the slider) is rotated with respect toanother joint (joint-2). Note that the joint-2 is a universal joint.The other end of the link for the slider is attached on the slideron the link-2. Since the radius of the link-2 and the link for the


Fig. 5. Location of electrodes.

slider are different, the slider moves along the link-2 accordingto the shoulder flexion angle. In the case of shoulder horizontalflexion-extension motion, the link-1 is rotated about its axis ac-cording to the horizontal flexion-extension angle. As the link-1rotates, joint-3 is rotated with respect to the axis of the link-1.The rotation of the joint-3 causes the movement of the posi-tion of the joint-2 along the lateral-medial direction as shown inFig. 4. As the position of the joint-2 moves along the lateral-me-dial direction, the slider moves along the link-2 since the link forthe slider is connected to the joint-2.

Usually, the limitation of human shoulder movable range is180 in flexion, 60 in extension, 180 in abduction, and 75in adduction. Considering the practical application to everydaylife, the shoulder motion limitation of the proposed robot is 0in extension and adduction, 90in flexion, and 90 in abduc-tion this system. The maximum angular velocity of the motor islimited by the hardware for safety. The maximum torque of therobot (i.e., the maximum current of the motor) is also limited byboth the hardware and software for safety. Furthermore, there isan emergency stop switch beside the robot.

III. CONTROL OF THEROBOT

By adjusting the amount of force generated by the shouldermuscles, the shoulder motion can be moderately controlled. Themuscle activity level can be described by the EMG signal. Con-sequently, human intention of shoulder motion can be estimatedby observing the EMG signals of the shoulder muscles.

Fuzzy-neuro control, combination of fuzzy control andadaptive neuro control, is applied to control the exoskeletalrobot. The initial fuzzyIF-THEN control rules of the fuzzy-neurocontrol are designed based on the analyzed human subject’sshoulder motion patterns in the experiment [3] and the experi-mental results in another research [13], [14] assuming that thearm posture of the human subject is in standard posture (i.e.,shoulder rotation angle is neutral [0], elbow flexion/extensionangle is neutral [0], and arm pronation/supination angle isneutral [0 ]).

The skin surface EMG signals of shoulder muscles, whichimply the human subject’s intention, and the shoulder joint an-gles are used as input signals of the robot controller in orderto control the robot as the human subject intended. The loca-tion of electrodes on shoulder muscles is shown in Fig. 5. Theelectrodes are located on the anterior, posterior and middle part

Fig. 6. Architecture of the fuzzy-neuro controller.

Fig. 7. Teaching equipment.

of deltoid, biceps, triceps, pectoralis major (lateral part), teresmajor, pectoralis major (clavicular part), and trapezius and thoseare connected to ch.1, ch.2, ch.3, ch.4, ch.5, ch.6, and ch.7, re-spectively.

The input variables of the fuzzy-neuro control are the meanabsolute value (MAV) [22] of EMG of seven kinds of muscles.The equation of the MAV is written as

(1)

where is the voltage value at th sampling and is thenumber of samples in a segment. The number of samples is setto be 100 and the sampling time is set to be 0.5 ms in this study.

Four kinds of fuzzy linguistic variables (ZO: zero, PS: posi-tive small, PM: positive medium, and PB: positive big) are pre-pared for each MAV of EMG of main muscles (ch. 2, 4-6). Threekinds of fuzzy linguistic variables (ZO: zero, PS: positive small,


Fig. 8. Change of membership function.

and PB: positive big) are prepared for each MAV of EMG ofthe other muscles (ch. 1, 3, and 7). Another three kinds of fuzzylinguistic variables (LA: low angle, MA: medium angle, HA:high angle) are prepared for each shoulder joint angle. The out-puts of the fuzzy-neuro control are the torque command to gen-erate the desired shoulder motion of the exoskeletal robot. Thetorque command for the exoskeletal robot joints is then trans-ferred to the force command for each driving wire. The relationbetween the torque command for the exoskeletal robot joints andthe force command for driving wires is written as the followingequation:

(2)

where is the torque command vector for the exoskeletal robotjoints, is the force command vector for the driving wires,and is the Jacobian which relates the exoskeletal robot jointvelocity to the driving wire velocity. Force control is carried outto realize the desired force () in driving wires by the drivingmotors.

In the fuzzy-neuro controller, 32 kinds of fuzzyIF-THEN rulesare prepared to generate the desired torque of the exoskeletalrobot. The architecture of the fuzzy-neuro controller is depictedin Fig. 6. Here means sum of the inputs, means multi-plication of the inputs. Two kinds of nonlinear functions (and ) are applied to express the membership function of thefuzzy-neuro controller.

(3)

(4)

(5)

(6)

Fig. 9. Effect of the arm posture change.

where is a threshold value and is a weight. The processof the fuzzy-neuro controller is the same as that of ordinal sim-plified fuzzy controllers. Consequently, the output of the fuzzy-neuro controller is calculated with the following equation:

(7)

where represents the output vector, denotes the degree offitness of the rule, and is the weight for the rule.

When the exoskeletal robot is attached to another human sub-ject, or when physiological condition of the human subject ischanged a lot, on-line adaptation of fuzzy-neuro controller iscarried out by adjusting each weight of the fuzzy-neuro to min-imize the amount of muscle activity and motion error which isgiven by the teaching equipment shown in Fig. 7. The angle oflink-1 of the teaching equipment is supposed to correspond to


Fig. 10. Architecture of the intelligent interface.

that of link-1 of the exoskeletal robot, and the angle of link-2of the teaching equipment is supposed to correspond to thatof link-2 of the exoskeletal robot. In the adaptation process,the human subject indicates his/her desired shoulder motion bydemonstrating the same motion with the teaching equipmentusing his/her wrist. In this study, both the antecedent part and theconsequent part of the fuzzyIF-THEN control rules are supposedto be adjusted to fit physiological condition of each human sub-ject by using the back-propagation learning algorithm in onlinemanner. The evaluation function for the fuzzy-neuro controllertraining is written as:

(8)

where is the desired shoulder angle indicated by the teachingequipment, is the measured shoulder angle,is a coefficientwhich changes the degree of consideration of the muscle activityminimization, is the desired muscle activity level inch.i, and is the measured muscle activity level in ch.i.The assistance level of the robot can be moderately adjusted bychanging the desired muscle activity levels. Note that certain de-sired muscle activity levels are prepared for each shoulder mo-tion (i.e., vertical shoulder flexion/extension motion and hori-zontal shoulder flexion/extension motion) considering physio-logical muscle allocation.

The inputs to the fuzzy-neuro controller are instantly adjustedby the modification coefficients outputted from the intelligentinterface, which is explained in the next section, in accordancewith the human subject’s arm posture. The definition of mem-bership functions of input variables to the fuzzy-neuro controlleris adjusted immediately by multiplying the input variables bythe modification coefficients. This operation makes the same

Fig. 11. Experimental setup.

effect as changing the membership functions wider or narrower[23] as shown in Fig. 8.

IV. I NTELLIGENT INTERFACE

Human shoulder joint consists of many muscles and movesin 3-DOF (flexion-extension, abduction-adduction, and in-ternal-external rotation). The muscle activity level can bedescribed by the EMG signal. Even though the human subjecttries to perform the same shoulder motion, the amount ofthe EMG signals from the shoulder muscles varies if the armposture is changed because of physiological reason [20]. Thedisplacement of the shoulder muscles is geometrically changedif the arm posture is changed.


(a)

(b)

Fig. 12. Experimental results for Trajectory 1 with assist of the exoskeletalrobot (standard arm posture).

In this study, intelligent interface is proposed to take intoaccount the subject’s arm posture. The intelligent interfacemodifies the controller inputs by multiplying modification co-efficients according to the subject’s arm posture. A neural net-work is used to realize the intelligent interface. Preliminaryexperiment, which investigates the effect of subject’s arm pos-ture (the effect of shoulder vertical flexion/extension angle,shoulder horizontal flexion/extension angle, shoulder internalrotation angle, elbow flexion/extension angle, and forearmpronation/supination angle) with respect to the amount of EMGsignals of shoulder muscles, was performed to prepare thetraining data of the neural network. An example of the effectthe arm posture change is shown in Fig. 9. The neural networkmakes a nonlinear mapping between subject’s arm posture andmodification coefficients for the controller inputs by off-linelearning. The architecture of the neural network is depicted in

(a)

(b)

Fig. 13. Experimental results for Trajectory 1 without assist of the exoskeletalrobot (standard arm posture).

Fig. 10. This neural network realizes the intelligent interfacebetween the human subject and the fuzzy-neuro controller.The neural network consists of three layers (input layer,hidden layer, and output layer). The input layer consists offive neurons, the hidden layer 50 neurons, and the outputlayer eight neurons. Sigmoid function is used for neuronsin the hidden layer and the output layer. The input variablesto the neural network are shoulder flexion/extension angle,shoulder horizontal flexion/extension angle, shoulder internalrotation angle, elbow flexion/extension angle, and forearmpronation/supination angle. The outputs from the neural net-work are the modification coefficients for input variables tothe fuzzy-neuro controller (i.e., EMG signals of the shouldermuscles).

The neural network has been trained with the training data(91,000 data set) obtained from the preliminary experimental


(a)

(b)

Fig. 14. Experimental results for Trajectory 2 with assist of the exoskeletalrobot (standard arm posture).

results using the back-propagation learning algorithm in off-linemanner.

V. EXPERIMENT

Experiment has been carried out with a health humanmale subject (22 years old) to evaluate the effectiveness ofthe proposed exoskeletal robot and its control method. Theexperimental setup is depicted in Fig. 11. The amplified EMGsignals are sampled at a rate of 2 kHz and the signals from thewire tension sensors are also sampled at a rate of 2 kHz andlow-pass filtered at 8 Hz.

For the first experiment, the target following experimentshave been carried out with and without assist of the exoskeletalrobot in order to verify the controllability of the exoskeletalrobot. When the experiment without assist of the exoskeletalrobot was performed, the human subject wore the exoskeletalrobot to measure the shoulder angle. We had experimentallyverified that wearing the robot did not affect the EMG signals

(a)

(b)

Fig. 15. Experimental results for Trajectory 2 without assist of the exoskeletalrobot (standard arm posture).

of the human subject. This experiment has been performedwith the standard arm posture of the human subject (withoutchanging the arm posture of the human subject). The initialcontrol rules of the fuzzy-neuro controller were designedassuming the arm posture of the human subject was in standard.In the experiment, the target trajectory (Trajectory 1: bothvertical flexion-extension and horizontal flexion-extensiontrajectory, Trajectory 2: vertical flexion-extension trajectory atthe horizontal flexion angle 30) of shoulder is displayed on themonitor, and a human subject is supposed to make his shoulderangles follow it. The generated shoulder trajectory is supposedto be very close to the target trajectory if the exoskeletal robotis well controlled, and the EMG levels of shoulder muscles aresupposed to be lower if the exoskeletal robot effectively assiststhe shoulder motion of the human subject. The experimentalresults (EMG signals at the anterior and middle part of deltoid)of the human subject for the Trajectory 1 with and without assistof the exoskeletal robot are shown in Figs. 12 and 13, and thosefor the Trajectory 2 with and without assist of the exoskeletal


(a)

(b)

Fig. 16. Experimental results for Trajectory 2 without assist of the exoskeletalrobot (changed arm posture).

robot are shown in Figs. 14 and 15, respectively. These resultsshow that the human subject can follow the target trajectorieswith and without support of the exoskeletal robot. This meansthe exoskeletal robot does not constrain the subject’s shouldermotion. One can also see that the EMG levels of shouldermuscles become lower when the shoulder motion of the subjectis assisted by the exoskeletal robot. Here, muscle activity level(MAV) at the anterior part of deltoid was reduced to 18% and15% for the Trajectory 1 and 2, respectively. Muscle activitylevel (MAV) at the middle part of deltoid was reduced to 15%and 60% for the Trajectory 1 and 2, respectively. These resultsshow the effectiveness of the proposed robot in human shouldermotion assist in the case when the arm posture of the humansubject is in standard.

When the arm posture of the human subject is changed(shoulder horizontal flexion/extension angle: 30, shoulder in-ternal rotation angle: 0, elbow flexion angle: 90, and forearmpronation/supination angle: 0) from the standard posture, the

(a)

(b)

Fig. 17. Experimental results for Trajectory 2 with assist of the exoskeletalrobot (changed arm posture) without the proposed interface.

EMG levels of shoulder muscles during the shoulder verticalflexion-extension motion at the horizontal flexion angle 30(Trajectory 2) are changed as shown in Fig. 16. If we applythe exoskeletal robot for motion assist with the fuzzy-neurocontroller designed for the standard arm posture without theproposed interface (neural network) to this case, the exoskeletalrobot does not work very efficiently as shown in Fig. 17. Inthis case, the EMG levels of shoulder muscles are not so muchimproved and the target following becomes worse than thosewithout assist of the exoskeletal robot. Here, muscle activitylevels (MAV) at the anterior and middle part of deltoid werereduced to only 57% and 58%, respectively. However, if thefuzzy-neuro controller is modified by the proposed interface,the results become better as well as those in Fig. 14 as shownin Fig. 18. Here, muscle activity levels (MAV) at the anteriorand middle part of deltoid were reduced to 18% and 31%, re-spectively. These results show the importance and effectivenessof the proposed interface.


(a)

(b)

Fig. 18. Experimental results for Trajectory 2 with assist of the exoskeletalrobot (changed arm posture) with the proposed interface.

VI. CONCLUSION

In order to realize a practical and effective exoskeletal robotfor shoulder joint motion assist, an effective fuzzy-neuro con-troller, a moving mechanism of the CR of the shoulder joint ofthe exoskeletal robot, and intelligent interface have been pro-posed in this paper. A moving mechanism of the CR of theshoulder joint of the exoskeletal robot has been proposed tofit the CR of the robot shoulder joint to that of the physio-logical human shoulder joint during the shoulder motion. Aneffective fuzzy-neuro controller has been also proposed to au-tomatically control the robot with EMG signals of the humanshoulder muscles. Furthermore, an intelligent interface betweenthe human subject and the fuzzy-neuro controller has been pro-posed to cancel out the ill effects caused by subject’s arm pos-ture in this paper. The effective control of the exoskeletal robotfor human shoulder motion assist can be expected by realizingthe intelligent interface. A nonlinear map between subject’s arm

posture and modification coefficients for the fuzzy-neuro con-troller inputs was examined in the preliminary experiment andthen off-line trained by a neural network to realize the intelli-gent interface. The effectiveness of the proposed method hasbeen verified by the experiment with healthy human subjects.We would like to apply the proposed system to elderly personsand handicapped persons for the future research.

REFERENCES

[1] K. Kiguchi, S. Kariya, K. Watanabe, K. Izumi, and T. Fukuda, “An ex-oskeletal robot for human elbow motion support – Sensor fusion, adap-tation, and control,”IEEE Trans. Syst., Man, Cybern., pt. B, vol. 31, pp.353–361, June 2001.

[2] , “Application of multiple fuzzy-neuro controllers of an exoskeletalrobot for human elbow motion support,” inProc. 32nd Int. Symp.Robotics (ISR2001), 2001, pp. 1563–1568.

[3] K. Kiguchi, K. Iwami, T. Saza, S. Kariya, K. Watanabe, K. Izumi, andT. Fukuda, “A study of an exoskeletal robot for human shoulder mo-tion support,”Proc. IEEE/RSJ Int. Conf. Intelligent Robots and Systems(IROS2001), pp. 2111–2116, 2001.

[4] R. S. Mosher and B. Wendel, “Force-reflecting electrohydraulic servo-manipulator,”Electro-Technol., pp. 138–141, Dec. 1960.

[5] W. Cloud, “Man amplifiers: Machines that let you carry a ton,”PopularSci., vol. 187, no. 5, pp. 70–73, 1965.

[6] R. S. Mosher, “Handyman to Hardiman,” Soc. Auto. Eng., publicationMS670 088, 1967.

[7] H. Kazerooni and S. L. Mahoney, “Dynamics and control of robotic sys-tems worn by humans,”Trans. ASME, J. Dyn. Syst., Meas. Control, vol.113, no. 3, pp. 379–387, 1991.

[8] V. L. Nickel, A. Karchak Jr., and J. R. Allen, “Electrically poweredorthotic systems,”J. Bone and Joint Surgery, vol. 51-A, no. 2, pp.343–351, 1969.

[9] N. Benjuya and S. B. Kenney, “Hybrid arm orthosis,”J. Prosthetics andOrthotics, vol. 2, no. 2, pp. 155–163, 1990.

[10] O. Fukuda, T. Tsuji, A. Ohtsuka, and M. Kaneko, “EMG-based human-robot interface for rehabilitation aid,” inProc. IEEE Int. Conf. Roboticsand Automation, 1998, pp. 3942–3947.

[11] D. Nishikawa, W. Yu, H. Yokoi, and Y. Kakazu, “EMG prosthetic handcontroller using real-time learning method,” inProc. IEEE Int. Conf.Systems, Man, and Cybernetics, 1999, pp. I-153–I-158.

[12] H. Huang and C. Chen, “Development of a myoelectric discriminationsystem for a multi-degree prosthetic hand,” inProc. IEEE Int. Conf.Robotics and Automation, 1999, pp. 2392–2397.

[13] B. Laursen, B. R. Jensen, G. Nemeth, and G. Sjogaad, “A model pre-dicting individual shoulder muscle forces based on relationship betweenelectromyographic and 3D external forces in static position,”J. Biome-chanics, vol. 31, no. 8, pp. 731–739, 1998.

[14] A. T. C. Au and R. F. Kirsch, “EMG-based prediction of shoulder andelbow kinematics in able-bodied and spinal cord injured individuals,”IEEE Trans. Rehab. Eng., vol. 8, pp. 471–480, Dec. 2000.

[15] T. Tsuji, Y. Kawaguchi, and M. Kaneko, “An adaptive training methodfor human-robot systems using neural networks” (in Japanese),J. Robot.Soc. Jpn., vol. 18, no. 5, pp. 683–389, 2000.

[16] S. Fujii, D. Nishikawa, and H. Yokoi, “Development of prosthetic handusing adaptive control method for human characteristics,” inIntell.Auton. Syst.. Amsterdam, The Netherlands: IOS Press, 1998, pp.360–367.

[17] V. M. Zatsiorsky, Kinematics of Human Motion. Champaign, IL:Human Kinetics, 1998.

[18] H. G. Coury, S. Kumar, and Y. Narayan, “An electromyographic study ofupper limb adduction force with varying shoulder and elbow postures,”J. Electromyography and Kinesiol., vol. 8, pp. 157–168, 1998.

[19] H. Graichen, K.-H. Englmeier, M. Reiser, and F. Eckstein, “Anin vivotechnique for determining 3-D muscular moment arms in different jointpositions and during muscular activation,”Clinical Biomechan., vol. 16,pp. 389–394, 2001.

[20] R. L. Lieber, “Skeletal muscles is a biological example of a linearelectro-active actuator,” presented at the SPIE 6th Annu. Int. Symp.Smart Structures and Materials, San Diego, CA, Mar. 1999.

[21] Z. Dvir and N. Berme, “The shoulder complex in elevation of the arm:A mechanism approach,”J. Biomechan., vol. 11, pp. 219–225, 1978.


[22] B. Hudgins, P. Parker, and R. N. Scott, “A new strategy for multifunctionmyoelectric control,”IEEE Trans. Biomed. Eng., vol. 40, pp. 82–94, Jan.1993.

[23] K. Kiguchi and T. Fukuda, “Intelligent position/force controller for in-dustrial robot manipulators application of fuzzy neural networks,”IEEETrans. Ind. Electron., vol. 44, pp. 753–761, Dec. 1997.

Kazuo Kiguchi (S’92–M’93) received the B.E.degree in mechanical engineering from NiigataUniversity, Niigata, Japan in 1986, the Master ofApplied Science degree in mechanical engineeringfrom the University of Ottawa, Ottawa, ON, Canada,in 1993, and the Ph.D. degree in engineering fromNagoya University, Nagoya, Japan, in 1997.

From 1986 to 1989, he was a Research Engineerwith Mazda Motor Company and from 1989 to 1991,with MHI Aerospace Systems Company. From 1994to 1999, he was with the Department of Industrial and

Systems Engineering, Niigata College of Technology, Niigata, Japan. Currently,he is an Associate Professor in the Department of Advanced Systems ControlEngineering, Graduate School of Science and Engineering, Saga University,Saga, Japan. His research interests include biorobotics, intelligent robots, ma-chine learning, application of soft computing for robot control, and applicationof robotics in medicine.

Dr. Kiguchi is a Member of the Robotics Society of Japan, IEEE Roboticsand Automation Society, IEEE Systems, Man, and Cybernetics Society, IEEEEngineering in Medicine and Biology Society, IEEE Industrial Electronics So-ciety and IEEE Computer Society, the Japan Society of Mechanical Engineers,the Society of Instrument and Control Engineers, the Japan Society of ComputerAided Surgery, International Neural Network Society, Japan Neuroscience So-ciety, the Virtual Reality Society of Japan, the Japanese Society of Prostheticsand Orthotics, and the Japanese Society for Clinical Biomechanics and RelatedResearch. He received the J. F. Engelberger Best Paper Award at WAC2000.

Koya Iwami was born on November 23, 1978. Hereceived the B.E. degree from Saga University, Saga,Japan, in 2001, where he is currently pursuing theM.E. degree in the Department of Advanced SystemsControl Engineering.

Makoto Yasuda was born on February 3, 1979. Hereceived the B.E. degree from Saga University, Saga,Japan, in 2002, where he is currently pursuing theM.E. degree in the Department of Advanced SystemsControl Engineering.

Keigo Watanabe (S’83–M’90) received the B.E.and M.E. degrees in mechanical engineering fromthe University of Tokushima, Tokushima, Japan, in1976 and 1978, respectively, and the D.E. degree inaeronautical engineering from Kyushu University,Fukuoka, Japan, in 1984.

From 1980 to March 1985, he was a ResearchAssociate at Kyushu University. From April 1985to March 1990, he was an Associate Professor atthe College of Engineering, Shizuoka University,Shizuoka, Japan. From April 1990 to March 1993, he

was an Associate Professor, and from April 1993 to March 1998, he was a FullProfessor in the Department of Mechanical Engineering, Saga University, Saga,Japan. Since April 1998, he has been with the Department of Advanced Sys-tems Control Engineering, Graduate School of Science and Engineering, SagaUniversity. He has published more than 360 technical papers in transactions,journals, and international conference proceedings, and is the author or editorof 18 books, includingAdaptive Estimation and Control(Englewood Cliffs,NJ: Prentice-Hall, 1991),Stochastic Large-Scale Engineering Systems(NewYork: Marcel Dekker, 1992) andIntelligent Control Based on Flexible NeuralNetworks(Norwell, MA: Kluwer, 1999). He is an Active Reviewer of manyjournals and transactions, and an Editor-in-Chief ofMachine Intelligence andRobotic Control, and an Editorial Board Member of theJournal of Intelligentand Robotic Systemsand theJournal of Knowledge-Based Intelligent Engi-neering Systems. His research interests are in stochastic adaptive estimationand control, robust control, neural network control, fuzzy control, and geneticalgorithms and their applications to machine intelligence and robotic control.

Dr. Watanabe is a Member of the Society of Instrument and Control Engi-neers, Japan Society of Mechanical Engineers, Japan Society for Precision En-gineering, Institute of Systems, Control and Information Engineers, the JapanSociety for Aeronautical and Space Sciences, Robotics Society of Japan, andJapan Society for Fuzzy Theory and Systems.

Toshio Fukuda (M’83–SM’93–F’95) graduatedfrom Waseda University, Tokyo, Japan, in 1971 andreceived the M.S. and Dr. Eng. degrees from theUniversity of Tokyo, Tokyo, Japan, in 1973 and1977, respectively. He also studied at the GraduateSchool of Yale University, New Haven, CT, from1973 to 1975.

In 1977, he joined the National MechanicalEngineering Laboratory, Tsukuba, Japan, and from1979 to 1980, became a Visiting Research Fellowat the University of Stuttgart, Stuttgart, Germany. In

1982, he joined the Science University of Tokyo, Tokyo, Japan, and in 1989,joined Nagoya University where he is currently a Professor in the Departmentof Micro System Engineering, Department of Mechano-Informatics andSystems, engaging in the research of intelligent robotic systems, cellularrobotic systems, mechatronics and micro- and nanorobotics. He is an author ofsix books, editor of five books, and has published over 1,000 technical papersin micro- and nanosystems, robotics, mechatronics, and automation.

Dr. Fukuda has been a Fellow of the Society of Instrument and ControlEngineers (SICE), since 1995. He received the IEEE Eugene MittlemannAward in 1997, the Banki Donat Medal from the Polytechnic University ofBudapest, Budapest, Hungary in 1997, the Medal from the City of Sartillo,Sartillo, Mexico in 1998, and the IEEE Millennium Medal, in 2000. He wasthe Vice President of the IEEE Industrial Electronics Society (IES) from 1990to 1999, IEEE Neural Network Council Secretary since 1992, Vice Presidentof the International Fuzzy Systems Association (IFSA) since 1997, Presidentof IEEE Robotics and Automation Society President from 1998 to 1999,Editor-in-Chief of the IEEE/ASME TRANSACTIONS ONMECHATRONICS from2000 to 2002, Director of the IEEE Division X from 2001 to 2002, and theFounding President of the IEEE Nanotechnology Council since 2002.

an exoskeletal robot for human shoulder joint motion ... · shoulder joint. in thispaper, we also...

Documents