A Novel Powered Gait Orthosis using Pneumatic Muscle Actuator

ICCAS2003 October 22-25, Gyeongju TEMF Hotel, Gyeongju, Korea

1. INTRODUCTION

To enable the SCI patients to stand and walk themselves is one of the goals in the rehabilitation engineering. The resciprocationg gait orthosis (RGO) was proposed for the paraplegic individual who is unable to stand or walk[1,2,3,4,5].

RGO is to couple the right and left sides of the orthosis by specially designed hip joints and pelvic section. The function of RGO is that hip extension in one joint results in hip flexion on the contralateral side. However it needs many energy expenditure because patients use pelvis and trunk movement excessively for gait.

In this study, we propose a novel powered gait orthosis(PGO) attached to artificial muscles using pneumatic actuator in hip joint. The PGO gait can walk more easely than RGO because the PMA assists hip flexion power in heel off. Role of artificial muscles assist hip flexion. To evaluat the PGO for paraplegic patient , we analyse three dimensional gait motion and energy consumption test. We show energy efficiency of the paraplegic during ambulation is improved.

2. Air muscle

The Air Muscle[6] is an extraordinary actuator that is small, light, and simple. It is soft, has no stiction, is easily controllable and exceptionally powerful. The Air Muscle consists of a rubber tube covered in braided plastic mesh shell which shortens in length like a human muscle when inflated with compressed air at low pressure.

An Air Muscle has a power-to-weight ratio as high as 400:1, vastly outperforming both pneumatic cylinders and DC motors that can attain a ratio of only about 16:1. It has been in continuous development for advanced robotics work by Shadow since 1982, and is now available for use in a variety of applications as a powerful, lightweight actuator. Air Muscles are normally operated using compressed air in the 0-70psi (0-5 bar) range. Figure 1 is a specification of air muscle in this experiment.

Fig. 1 A specification of air muscle

3. Driving System

The concept of the PGO driving system is to couple the right and left sides of the orthosis by specially designed hip joints and pelvic section. Hip extension in one joint results in hip flexion on the contralateral side, resulting in a more normal and efficient type (reciprocating) gait. The flexion power of the hip using air muscle is the ideal ?motor? which drives the orthosis. Also, the pelvic band of the PGO torso sections is made of high strength aluminum. It is designed to be very strong and rigid. This unique pelvic band design, keeps the legs tracking so one does not interfere with the other.

This permits easier walking, less gait training time and less energy consumption.

In figure.2, driving system of powered gait orthosis(PGO) consists of the orthosis, compressed air system that supply air in air muscle, sensors, control system. A compressed air supply system is composed of a air compressor, 2-way solenoid valve(MAC, USA), accumulator, pressure sensor.

Role of this system provide air muscle with the compressed air at hip joint constantly. According to output signal of foot switch sensor, air muscle assists the flexion of hip joint during PGO gait

Fig. 2. A block diagram of controller and compressed air supply system in PGO

.

4. Control system

We made the gait step table by output signals from foot switches with figure 3. Microprocessor by this table controls air flow rate that is supplied in air muscle using solenoid

A Novel Powered Gait Orthosis using Pneumatic Muscle Actuator

Korea Orthopedics & Rehabilitation Engineering center, Incheon, (Tel : +82-32-500-0598; E-mail: kangsj@iris.korec.re.kr) Abstract :

One of the main goals in the rehabilitation of SCI patients is to enable the patient to stand and walk themselves. We are developing high-thrust powered gait orthosis(PGO) that use air muscle actuator(shadow robot Co., UK) to be assisted gait and rehabilitation purposes of them. We made of PD controller and measured hip joint angle by its load and the pressure to control air muscle of PGO. As a results, maximum flexion angle of hip joint is 20°, and angular velocity is 30.4±2.5°/sec, and then delay time of system was average 0.62±0.03s. As the hip flexion angle and the pelvic angle is decreased during the gait with PGO, the patient can walk faster. By using the PGO, the energy consumption can also be decreased. therefore, the proposed PGO can be a very useful assitive device for the paraplegics to walk.

Keywords: power ed gait orthosis , PD controller ,Air muscle, Rehabilitation, orthosis, Reciprocating Gait Orthosis

1500

ICCAS2003 October 22-25, Gyeongju TEMF Hotel, Gyeongju, Korea

valve.

Role of two foot switches that is attached on fore and back of sole of a foot operates air muscle according to gait cycle.

The foot switches are set metatarsal and calcaneus anatomically. The flexion/extension of hip joint with gait cycle classified by 4 step in figure 3. When the foot sensors of heel and metatarsal does not contact on floor, because it was identify it by swing phase of gait. This assist the flexion of hip joint to contract air muscle. At heel contact, the hip joint become fully extension, release the air in air muscle.

Fig. 3. A classification of gait step during the duration of gait

We were used by select icroprocessor(PIC16C73) for the control air muscle and the detection of real-time gait cycle.

Control routine of this system consisted of initial system routine and main service routine. Main service routine includes with A/D converting routine, PD control routine, and PWM output routine. A control variable of this system is setted by 100 ms in resolution the flexion angles of hip joint. Figure 4 is flowchart of the control logic program

Fig. 4. A flowchart of control program

In figure 5, PD[7](angle position) control system of PGO made of the feedback system that adjust pressure in air muscle comparing with difference of the flexion angle of hip joint measure by bend type angle sensor(Abrams Gentile Co., USA).

Fig. 5 A block diagram of position control system

5. Gait Analysis Subjects

Three subjects(aged 33.25±11.5 years) who were two adult normal male and one paraplegic male participated in this study.

Subjects were recruited from the laboratory staffs and patient of the hospital. Their heights and weights ranged from 170±1.7㎝, 60.5±7.5kg, respectively.

Experimental procedures

Subjects were performed the gait analysis five times per one month after gait training on PGO during about three months.

In order to identify any kinematic value of subjects, we were used six infrared CCD cameras and spheric reflective markers of 25mm in diameter. Seventeen spheric reflective markers were mounted in rigid arrays secured to each body segments : the anterior superior Iiliac spine, sacrum, great trochanter, medial thigh, knee joint and ankle joint, medial tibia, foot. The analog position signals of each body-fixed markers were converted to digital form, fed on line to a computer, and stored on a hard disc. The raw data of the marker positions were passed through a filter, and analyzed with the built-in software in the three-dimensional motion analyzer system( Vicon 370; Oxford Metrics Inc., UK). To obtained kinetic data each lower limb joint of subjects, we measured ground reaction force using two force plate( 900×600㎜, 600×400㎜ in size; Kistler Co., Swiss).

Also, the energy expenditure of PGO and RGO gait was measured continuously by a light-weight, portable, and telemetric energy consumption measuring system( K4, KOSMED Co., Italy). The system measured the metabolic variables including oxygen uptake(VO2), carbon dioxide(CO2) production, heart rate(HR), and minute ventilation. The system consisted of a processing unit, a receiver unit, and a battery. A soft face -mask was used to sample the exhaled air. The exhaled gas entered a microdynamic chamber that analyzed the oxygen concentration by an electrochemical sensor, and the CO2 concentration by an infrared electrode. A cardiac belt transmitted HR signals to the processing unit, which contained a transmitter that sent signal to the receiver unit.

Fig. 6 A photograph of measurement of the energy expenditure

1501

ICCAS2003 October 22-25, Gyeongju TEMF Hotel, Gyeongju, Korea

6. Results and discussion

The aim of this study ultimately is verifying that PGO gait is more efficient than RGO for paraplegic, because the air muscle assists hip flexion power in heel off.

In figure 7 and 8 the gait characteristics of the paraplegic wearing PGO and RGO are compared with that of the normal person. At the heel contact in RGO and PGO gait, hip flexion angles are approximately 10 deg. while that of normal gait is approximately 30 deg. The smaller hip flexion angles of the PGO and RGO gaits are due to the smaller step length. In normal gait, the duration of stance phase and swing phase are 60% and 40% of the total gait cycle respectively. In RGO gait, the duration of the swing phase is reduced since the hip joint is flexed by the rotation of the pelvic band. The hip flexion angle fluctuates during the stance phase as the movement of the body center is unstable. On the other hand, in the PGO gait since the hip joint is flexed by the air muscle and not by the rotation of the trunk, the movement of the center of the body appears to be more stable than that in the RGO gait.

The ratio of the duration of the swing phase in PGO gait is 68±8% showing improvement from the RGO gait in which the duration of the swing phase is 79±4%. The gait speed is 61±3step/min for RGO gait and 77±2step/min in PGO gait respectively.

TO (RGO)

79±

%4

TO (PGO)

68±

%8

TO (Normal)

6 0%

HC TO (RGO)

79±

%4

TO (PGO)

68±

%8

TO (Normal)

6 0%

HC

Fig. 7 Hip flexion/extension angle

During gait, the pelvic tilt is normally related to the balance of the body and the amount of energy consumption. The pelvic tilt during the normal gait reduces the vertical movement of the hip joint and minimizes the vertical motion of the body center. In Figure 8, the pelvic tilts during the PGO and RGO gaits becomes larger than that of the normal gait. The reason is that since there are no knee flexion during the PGO/RGO gait, larger pelvic tilt than that of the normal gait is needed for toe off. The fact that pelvic tilt during the PGO gait is relatively smaller than that of the RGO gait means that the movement of the body center is more stable in PGO gait because of the air muscle. .

Fig. 8 pelvic tilt

The energy consumed during the gait can be estimated by measuring the oxygen consumption rate. The more the energy is consumed, When the muscle movement becomes larger during the gait, more oxygen is consumed and so is the energy.

In Figure 9, the oxygen consumption rates are 8.65±3.3(ml/min/Kg) in RGO gait and 7.2±2.5(ml/min/Kg) in PGO gait.

Fig 9. Hip flexion angle

7. Conclusion

In this paper a PGO using PMA's for SCI patients is proposed.

As the hip flexion angle and the pelvic angle is decreased during the gait with PGO, the patient can walk faster. By using the PGO, the energy consumption can also be decreased.

therefore, the proposed PGO can be a very useful assitive device for the paraplegics to walk.

ACKNOWLEDGMENTS

1502

ICCAS2003 October 22-25, Gyeongju TEMF Hotel, Gyeongju, Korea

This study wa supported by a grant of the Korea Health 21 R&D Project, Ministry of Health & Welfare, Republic of Korea. (02-PJ3-PG6-EV03-0004)

REFERENCES

[1] R. C. Baker and B. Charlie, “Nonlinear unstable systems,” International Journal of Control, Vol.

23, No. 4, pp. 123-145, 1989.

[2] Douglas R, Larson PF, DAmbrosia R, McCall RE. ,“The LSU reciprocation gait orthosis”, Orthopedics.,Vol.6,pp 834-839, 1983

[3] Solomonow M, Best R, Aguilar E, et al.

“Reciprocating gait orthosis powered with electrical muscle stimulation (RGO II): Part I.

Performance evaluation of 70 paraplegic patients.”, Orthopedics., Vol 20, pp315-324, 1997.

[4] Kobetic R, Triolo RJ, Marsolais EB. ,”Muscle selection and walking performance of multichannel FES systems for ambulation in paraplegia”, IEEE Trans Rehab Eng, Vol. 5, No.

1, pp 23-29, 1997.

[5] Marsolais EB ,Polando G, Lehneis HR, Tashman S., ”Der gang von paraplegikern mit dem hybrid system aus FNS/orthesen Paraplegic walking with hybrid FNS/Orthotic system”, Orthopadie Technik, Vol. 49, pp 372-376, 1998.

[6] Klute, G. K. , J. M. Czerniecki, and B.

Hannaford, “ McKibben Artificial Muscles : Pneumatic Actuators with Biomechanical Intelligence.” Proceeding of the IEEE/ASME 1999 International Conference on Advanced Intelligent Mechatronics (AIM’99), Atlanta, GA, Sep 19-22,1999

[7] Chou, C.P. , B. Hanaford, “Measurement and Modeling of McKibben Pneumatic Artificial Muscles.”, IEEE Transactions on Robotics and Automation, Vol. 12, No. 1, pp.90-102, 1996.

[8] R. K. Mudi, N.R. Pal, “ A robust self-tuning scheme for PI- and PD type fuzzy controllers”, IEEE Trans. Fuzzy Syst., Vol. 7, No. 1, Feb., 1999

1503