![](bg8.png)
prototype with an eye mask for removing the
influences by visual information.
As the result, the user can go straight with our
system as Fig.12(b). On the other hand, without our
assistance system, it is difficult to go straight by the
inclination as Fig.12(a). Fig.13 shows the difference
between the velocities of right and left wheel during
the experiment. When the difference is zero, the
wheelchair goes straight and when the difference is
the positive value, the wheelchair turns to the right
direction (the gravity direction) as Fig.12(a). From
Fig.13, the wheelchair goes straight with our
assistance. Fig.14 shows the average value of Fig.13.
In Fig.14(b), the maximum velocity difference is
almost same and we can verify that our controller
controls the wheels to realize the straight direction.
From these results, we can verify our system can
assist to fit the trajectory which its user wants.
(a) Without assistance
(b) With assistance
Figure 12: Test run on the slope (Subject A).
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
0510
Time [sec]
Velocity [m/sec]
A
B
C
D
E
F
-0.2
-0.1
0
0.1
0.2
0.3
0.4
0.5
0510
Time [sec]
Velocity [m/sec]
A
B
C
D
E
F
(a) Without assistance (b) With assistance
Figure 13: The velocity difference between a right and a
left wheel. Positive value means the wheelchair turns to
the right direction (The gravity direction).
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
ABCDE F
Subject
Velocivt [m/sec]
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
ABCDE F
Subject
Velocivt [m/sec]
(a) Without assistance (b) With assistance
Figure 14: The average value of the difference between
the velocities of a right and a left wheel. Positive value
means the wheelchair turns to the right direction (The
gravity direction).
5 CONCLUSIONS
In this paper, we develop a novel assistance control
for a passive-type wheelchair for healthy users who
have enough upper body strength. For realizing the
assistive wheel control, we develop the estimation
scheme for the user’s intention and our system
controls its wheels based on the estimated results.
Using our system, the user can move with the
wheelchair easily to the direction he wants.
In our future work, we will improve the user’s
motion estimation scheme. In the experiments in
section 4.2.1, the errors of the estimated trajectory
tend to be large when the subject changes the motion
rapidly. From our experiments, these motions are
characteristic and we will classify them considering
with the character of the wheelchair movement
during these motions.
ACKNOWLEDGEMENTS
This work is supported in part by Kawanishi
Memorial ShinMaywa Education Foundation.
REFERENCES
National Consumer Affairs Center of Japan, 2002.
Considering the safety of self-propelled manual
wheelchairs, In NCAC News, Vol.14, No.3, pp.4.
Japan Institute of Construction Engineering, 2008. A
Sidewalk guideline for smooth mobility, 2nd edition,
ISBN: 4802893922. (in Japanese)
Yamaha Motor Co., Ltd., 2012. JW
II, http://www.yamaha-motor.jp/wheelchair/unit/jw2/
(Available at Feb., 2013)
D. P. Miller and M. G. Slack, 1995. Design and Testing of
a Low-Cost Robotic Wheelchair Prototype, In
Autonomous Robots, Vol.2, pp.77-88.
N. I. Katevas, N. M. Sgouros, S. G. Tzafestas, G.
Papakonstantinou, P. Beattie, J. M. Bishop, P.
Tsanakas and D. Koutsouris, 1997. The Autonomous
Mobile Robot SENARIO: A Sensor-Aided Intelligent
Navigation System for Powered Wheelchairs, In IEEE
Robotics and Automation Magazine, Vol.4, No.4,
pp.60-70.
Y. Murakami, Y. Kuno, N. Shimada and Y. Shirai, 2001.
Collision avoidance by observing pedestrian’s faces
for intelligent wheelchairs, In Proc. of IEEE/RSJ Int.
Conf. on Intelligent Robots and Systems, pp.2018-
2023.
A. Goswani, M. A. Peshkin and J. Colgate, 1990. Passive
robotics: An exploration of mechanical computation,
In Proc. of IEEE Int. Conf. on Robotics and
Automation, pp.279-284.
8[de
]
ICINCO2013-10thInternationalConferenceonInformaticsinControl,AutomationandRobotics
266