Development of Pushrim-Activated Power-Assisted Wheelchair

Transcription

1 Development of Pushrim-Activated Power-Assisted Wheelchair Yoon Heo, Ki-Tae Nam, Eung-Pyo Hong, Mu-Sung Mun Korea Orthopedics & Rehabilitation Engineering Center 26, Gyeongin-ro 10beon-gil, Bupyeong-gu, Incheon, Korea, ABSTRACT Recent studies have focused on power-assisted wheelchairs because they have several advantages related to both manual and electric powered wheelchairs. They provide users with exercise for rehabilitation and reduce the user s physical fatigue by providing electric assistance. However, if the assist function is too strong, the effect of exercise may be decreased, and it may cause the weight heavy. Conversely, if it is too weak, the effect of the physical assistance may be insufficient. In this study, the wheelchair driving mechanism is modeled, and the appropriate motor power is chosen based on the computer simulation. For the driving experiments, the wheelchair is developed with in-wheel type electric assist wheels, which can be detached from the wheelchair body, and supports the manual driving mode with a clutch unit. In the experiments, the performance of the developed pushrim-activated power-assisted wheelchair is shown to be close to the proposed design specifications. KEYWORDS PAPAW, driving model, assistance device, torsion sensor, electric wheelchair 1. Introduction A manual wheelchair is an assistance device used by disabled or elderly people, and it has several advantages. In particular, users can exercise by propelling themselves forward with the pushrim; this may also give them the psychological satisfaction of independence, as they are not completely reliant on a machine for their mobility. However, some studies have reported that many manual wheelchair users experience shoulder or finger joint injuries because of the cumulative physical fatigue from self-propulsion [1-2]. Recently, to solve these problems, many researchers have investigated power assistance techniques that can amplify human power by detecting the propulsion movements of the user [3-6]. The pushrim-activated power-assisted wheelchair (PAPAW) is one such assistance device. PAPAWs can detect the pushrim propulsion torque applied by the user with torsion sensors and amplify the user-applied torque with builtin motors, thus reducing the user s physical fatigue during self-propulsion. PAPAWs can be built with lower-powered motors than fully electric powered wheelchairs and also have improved portability, as they can be carried by car. Thus, PAPAWs have advantages related to both manual and electric powered wheelchairs. In the development of a PAPAW, it is very important to determine the appropriate motor power. If the assist function is excessive and high-powered motors are chosen, the effect of exercise may be deteriorated, and the weight of the PAPAW may be too high. This means that its merits of acting like a manual wheelchair will be negated. Conversely, if the assistive power is too weak, the physical assistance of the PAPAW will be insufficient. Therefore, the usability of a PAPAW is highly dependent on appropriate motor selection. In this study, an appropriate motor power is chosen through computer simulation using a wheelchair driving model. For the driving experiments, the model is developed with inwheel type electric assist wheels, which may be detached from the wheelchair body, and ISBN: SDIWC 164

2 supports a manual driving mode with a clutch structure. In the experimental section, the performance of the developed PAPAW is examined in level and sloped driving environments and compared with the design specifications of the wheelchair driving model. In conclusion of this paper, the performance of the developed PAPAW is evaluated, and future studies are presented. Figure 1. Wheelchair driving model. 2. Wheelchair Driving Model As described in the introduction, the determination of appropriate motor power is very important because it is related to the weight of the wheelchair. As the assisted power increases, the wheelchair requires larger motors and higher-capacity batteries. In this study, under the maximum load condition, the required driving power is calculated from the wheelchair driving model, and 50% of the maximum driving power is specified for the assistive power of the PAPAW. In general, the condition under which the maximum power of a motor is required is when the wheelchair is climbing up a slope. If the energy loss of the front wheels is negligibly small, the wheelchair can be modeled using only the rear wheels, as shown in Fig. 1 [7,8 ]. Table 1 gives the wheelchair specifications for the computer simulation when the wheelchair is climbing an incline. The maximum speed on the slope is set to 3 km/h, which is half of the maximum speed on level ground. In Fig. 1, the force when the wheelchair climbs the slope can be represented as Table 1. Specification of the wheelchair model Specification Unit Design value Total weight of wheelchair kgf 30 Maximum load kgf 70 Radius of wheel (R) m 0.3 Maximum speed (V) km/h 3 Acceleration time s 3 Angle of incline ( ) 6 Coefficient of friction ( s ) Figure 2. The power to climb up the slope with respect to the angle of incline. f [ N ] f f (1) where f g is the force exerted by the gravitational acceleration g on the slope with a given incline, f is the frictional force between the ground and the wheels, and f w is the force required to accelerate the wheelchair up the slope with acceleration a. These forces are given as f w w g f [ N] ma (2) f g [ N ] mg sin( ) (3) f [ N ] f mg cos( ) (4) The torque required to drive the wheelchair up the slope can be simply calculated from the product of the force and the wheel radius R: s N s ISBN: SDIWC 165

3 [ Nm] f R (5) Lastly, the power P of the wheel required to climb the slope is the product of the torque and the angular speed of the wheel: P [ ] (6) W Using the above equations and the wheelchair specifications, the power of the wheel required to climb the slope is calculated with respect to the change in the angle of incline; the results are shown in Fig. 2. The final results of the power were calculated with respect to the system efficiency. We assumed that the system efficiency was approximately 65%; motor driver (90%), gear (90%), DC motor (80%). As shown in Fig. 2, the simulation results demonstrate that the power required for the wheelchair to climb a 6 slope is approximately 120 W. Therefore, the power of each wheel should be over 60 W. Initially, it was proposed that the assistive power of the PAPAW should be set to 50% of the maximum power. Therefore, the motor power can be reduced to 30 W for each wheel. The applied motor specifications are presented in Table Electric Assist Wheel Design The developed PAPAW has features of both manual and electric powered wheelchairs. Table 2 The applied motor specification Wheel diameter 24 in Rated power 22 W DC motor Maximum - RM51B power 50 W - SPG motor Rated speed 3725 rpm Weight 0.59 kg Gear ratio 1/4 Timing belt * Gear (19.7:1) Materials Wheel hub Aluminum Wheel rib Carbon Wheel speed Maximum of 6 km/h Wheel mass 6.5 kg Figure 3. Design of the torsion sensor unit. Therefore, the mechanical friction between the power transmission and the wheel hub make it difficult to drive in the manual mode. Accordingly, a clutch unit is designed to physically separate the wheel hub from the power transmission in the manual driving mode. The developed wheel is an in-wheel type electric actuator; a motor, power transmission, signal interface unit, and clutch unit are included in the wheel hub. 3.1 Torsion sensor unit The torsion sensor unit is connected to the pushrim to detect the user-generated propulsion force, and it consists of a potentiometer and three torsion springs. The potentiometer determines the displacement between the pushrim and the torsion springs, which allow the potentiometer to maintain a central position when there is no user input. For the signal interface between the sensor unit and the controller, brushes and a disk type flat slip ring are newly built for the simple sensor signal and DC power transfer. Fig. 3 shows the developed torsion sensor unit. 3.2 Power transmission and clutch unit The power transmission adopted in this PAPAW system consists of a belt and pulley system, which is a simple and low-cost structure that generates less noise than geared mechanisms. In Fig. 4, the developed power transmission structure is presented. The developed PAPAW can be changed to the manual driving mode when the battery is empty or short-range driving is required. ISBN: SDIWC 166

4 Figure 4. The power transmission unit. Figure 7. The configuration of the developed PAPAW control system. Figure 8. The developed electric assist wheels. Figure 5. The clutch unit. Figure 6. The general torque control system A clutch mechanism is designed to physically separate the power transmission unit from the wheel hub by adapting clutch pins that are connected to the timing pulley when the disk type lever is rotated by the user. The developed clutch mechanism is presented in Fig Control System Design As shown in Fig. 5, the general torque control system for the PAPAW consists of low-pass filters (LPFs) [3]. In the LPFs, α represents the assistance ratio, and the time constant τ indicates the assistive power attenuation rate. In general, when the propulsion torque generated by the user is eliminated, the electric Figure 9. Driving experiment on the slope powered wheel will also quickly stop the rotation because of the high mechanical friction in the power transmission. This causes discomfort in the user. Thus, the time constant τ is an important variable in the PAPAW control system, as it generates the virtual inertial torque, which gradually decreases the wheelchair s speed. 5. Experimental Results To validate that the performance of the developed PAPAW meets the design specifications of the wheelchair driving model, the driving experiments were examined with the wheelchair operating in level and inclined driving environments. Fig. 7 shows the configuration of the developed ISBN: SDIWC 167

5 Figure 10. The driving experiment result on the level ground in the manual mode. Figure 12. The driving experiment result on the slope(6 )in the manual mode. Figure 11. The driving experiment result on the level ground in the assist mode. PAPAW. Rotary encoders (Autonics, Korea) were mounted on both wheels to record the driving velocity of the PAPAW. The control system is constructed using a CompactRIO system (National Instruments). Fig 8 shows the developed electric assist wheels, and Fig. 9 shows driving test on a 6 incline. The weight of the user is 63 kg, and the weight of the developed PAPAW is 27 kg, including the battery and two electric assist wheels. In the first experiment, the user drove the developed PAPAW on level ground in both the manual and assist modes. Figs. 10 and 11 show the results of the manual and assist modes, respectively. Figs. 12 and 13 show the Figure 13. The driving experiment result on the slope(6 )in the assist mode. experimental results of the manual and assist modes, respectively, on a 6 incline. These experimental results were recorded at 5 s intervals, and the first three user torques are accumulated for the comparison in the manual and assist mode. The results of the performance comparisons are shown in Figs. 14 and 15. As shown in Figs. 14 and 15, during the level ground driving, the difference between the velocities in the manual and the assist modes is not clear, but the user accumulated torques in the assist mode were 48% lower than the torques in the manual mode. When the PAPAW climbed the 6 incline, only 49% of the accumulated input torques in the manual mode were used in the assist mode, and ISBN: SDIWC 168

6 contactless signal interface technique to be developed in the next study. ACKNOWLEDGEMENT Figure 14. Comparison results of the user accumulated torques on the driving experiments. Figure 15. Comparison results of the wheelchair speed(m/s) on the driving experiments. the velocity was 1.58 times higher than that in the manual mode. Therefore, the experiment results indicate that the developed PAPAW can amplify the usergenerated driving force and that the assistance performance is close to the proposed design specifications. 6. Discussion and Conclusion In this study, the maximum driving power is computed from the wheelchair driving model under the maximum load condition, and 50% of the maximum driving power is specified as the appropriate assist power for the PAPAW. From the driving experiments, the assistive performance of the developed PAPAW is close to the model specifications, which were predefined in the computer simulation. In future studies, subject evaluation will be conducted to gauge user satisfaction with the proposed design specifications. Regarding durability, the strength of the sensor interface using the slip ring and brush is insufficient because it requires frequent maintenance. Thus, it is necessary for a This study was supported by the Senior-friendly Product R&D program funded by the Ministry of Health &Welfare through the Korea Health Industry Development Institute (KHIDI) (HI14C1496). REFERENCES [1] W. Mark Richter, Motion-based Power Assist System for Wheelchairs, US Patent, No A1, [2] C. C. Ou, and T. Chi. Chen, Power-Assisted Wheelchair Design based on a Lyapunov Torque Observer, Int. J. of ICIC, Vol. 8, No. 12, pp , [3] S. h. Oh. and Y. Hori., Fundamental Research on Human-friendly Motion, PHD thesis, The Univ. of Tokyo, Japan, [4] H. Seki and S. Tadakuma, Velocity Pattern Generation for Power Assisted Wheelchair Based on Jerk and Acceleration Limitation, Int. Conf. of IECON Society, [5] H. Seki, T. Sugimoto and S. Tadakuma, Straight and Circular Road Driving Control of Power Assisted Wheelchair Based on Balanced Assisted Torque, Int. Conf. of IECON Society, [6] S. H. Hwang, C. H. Lee and Y. b. Bang, Power- Assisted Wheelchair with Gravity Compensation, 12 th ICCAS, Oct , [7] B. H. Kim, Analysis on Climbing Capability of Wheel Drive Robotic Mechanisms, Journal of KIIS, vol. 18. No 3, pp , 2008 [8] Y. Heo, E. P. Hong and M. S. Mun, Development of power add on drive wheelchair and its evaluation, 9 th ASCC, Jun , ISBN: SDIWC 169