Sitting Assistance that Considers User Posture Tolerance

Daisuke Chugo

1,2

, Masayu Koyama

, Masahiro Yokota

, Shohei Kawazoe

, Satoshi Muramatsu

Sho Yokota

, Hiroshi Hashimoto

, Takahiro Katayama

, Yasuhide Mizuta

and Atsushi Koujina

Graduate School of Science and Technology, Kwansei Gakuin University, Sanda, Hyogo, Japan

School of Science and Technology, Kwansei Gakuin University, Sanda, Hyogo, Japan

School of Science and Information and Telecommunication Engineering, Tokai University, Hiratsuka, Kanagawa, Japan

Faculty of Science and Engineering, Toyo University, Kawagoe, Saitama, Japan

Master Program of Innovation for Design and Engineering, Advanced Institute of Industrial Technology, Tokyo, Japan

Service Robot Division, Rt. Works Co., Ltd, Higashinari, Osaka, Japan

hashimoto@aiit.ac.jp, contact@rtworks.co.jp

Keywords: Sitting Assistance, Posture Tolerance, Robotic Walker.

Abstract: This paper proposes a novel sitting assistance robot, which considers the posture tolerance of its user. The

standing and sitting motion are different essentially because the standing is lifting motion against gravitational

force and sitting is posture coordination to sitting position according to the gravity. Therefore, the robot should

lead the patient’s posture within a stable range during sitting and the required performance is different from

standing assistance. However, in previous studies, conventional assistive robots used the sitting motion which

is “reverse” motion of standing. Furthermore, these robots helped patients by using a fixed motion reference

pathway in spite of their original intention, and as the results, these robots failed to assist by confliction

between their intended motion and reference path. Therefore, we propose a novel sitting assistance robot,

which allows its user to move their body within a prescribed degree of posture tolerance during the process

of moving from a standing to a sitting position. Our key findings cover two fundamental research topics. One

is the investigation into posture tolerance during a sitting motion. The other topic is a novel assistance control

algorithm that considers the investigated posture tolerance by combining position control and force control.

A prototype assistive robot, based on the proposed idea was fabricated to help patients sitting down safely

according to their original intention.

1 INTRODUCTION

Activities such as standing, walking, and sitting may

be the most serious and important activities in the

day-to-day lives of elderly people as they lack

physical strength (Alexander et al., 1999; Hughes et

al., 1996). In a typically bad case, an elderly person

who does not have enough physical strength will not

be able to stand up and sit down, and as the result,

they may then be restricted to life in a wheelchair life

or become bedridden (Cabinet Office, Government of

Japan, 2016). Furthermore, once an elderly person

falls into such lifestyle, the decrease in their physical

strength becomes more pronounced due to the lack of

exercise (Hirvensalo et al., 2000). For increasing their

QOL (Quality of Life), they need a personal assistive

robot which enables them to perform daily activities

alone easily even if their physical strength reduces by

aging.

In previous works, many researchers have been

developed assistance devices for a standing motion

(Nagai et al., 2003; Funakubo et al., 2001).

However, these devices are specialized in only a

“standing assistance” and they do not discuss on a

sitting motion. Some previous researchers say a

sitting motion is only “reverse” motion of standing

(Ehara et al., 1996). However, standing and sitting

motion are different essentially because standing is

lifting motion against gravitational force and sitting

is posture coordination to sitting position according

to the gravity. In general, a sitting motion has high

risk for falling down compared with a standing

motion for elderly people (Yoneda, 1998) and it is

difficult to realize a sitting assistance using only

“reverse” motion of standing. Furthermore, a sitting

assistance requires the posture coordination within

stable range, not a force assistance as a standing

assistance because sitting motion follows a gravity

direction. Therefore, the robot should assist the

Chugo, D., Koyama, M., Yokota, M., Kawazoe, S., Muramatsu, S., Yokota, S., Hashimoto, H., Katayama, T., Mizuta, Y. and Koujina, A.

Sitting Assistance that Considers User Posture Tolerance.

DOI: 10.5220/0007956104890496

In Proceedings of the 16th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2019), pages 489-496

ISBN: 978-989-758-380-3

489

patients according to their intended sitting motion,

and should help only when their posture are

unsuitable and have high risk for falling down.

In this paper, we propose a novel sitting

assistance robot, which considers the variation in the

range of movements of a patient's body when sitting

from a standing position. To achieve this objective,

we initially investigated the posture tolerance during

the process of sitting down. In this range, patients

can sit down, stably and safely, using their own

physical strength. Secondly, we extend a published

assistance algorithm, (Chugo et al., 2014) which

combines position control and force control, to adapt

to the parameters of the prior investigation into

posture tolerance. Using the proposed algorithm, our

robotic device only assists them to sit down when

necessary.

2 SYSTEM CONFIGURATION

2.1 System Overview

Previously, we have developed robotic walkers which

have a standing/sitting assistance function, (Chugo et

al., 2014; Chugo et al., 2017) and Fig. 1 shows our

recent latest prototype (Chugo et al., 2017). The

design of our proposed assistive device is based on a

powered walker. It has a standing/sitting support

manipulator, which moves the user in an

upward/downward direction. Fig. 1(a)-(b) shows the

default position of the walker’s manipulator at a

standing and walking heights. Its width is 54 cm and

can pass through a typical toilet door with a standard

width of 60 cm (JIS - Japan Industrial Standard –

1526:1997) as Fig. 1(c). Its height is 71.0–96.5 cm

and fits users whose heights are in the range of 145–

160 cm.

To lift down a user, our proposed walker uses a

linear motion DC motor and a gas spring. This motor

can generate a force in the up/downward direction,

whereas the gas spring helps this force in the

up/downward direction. In total, our system can lift a

weight of 40 kg, which is enough to assist someone

in standing or sitting. Using the gas spring, our system

can use a smaller actuator, which means that its

design can be fairly inexpensive. Furthermore, the gas

spring prevents the device from moving suddenly

when the power is down.

The wheels on each side of the walker have an

actuator and an electric parking brake as Fig. 1(a).

The actuated wheels are located in the almost same

position as that of a user and four caster wheels help

in maintaining its balance so that the user can turn

around on the spot when they walk using the device.

The actuator has enough power to control its own

position references, but when a user wants to fix its

position over a long time, they should make use of

the parking brake, because it is a mechanical brake

and its energy efficacy is better than that of the

control scheme containing the actuators.

(a) Side view (b) Front view

Figure 1: Our assistive walker.

2.2 User Interface

A handle, armrest, and controller are provided on the

top of the walker, as shown in Fig. 2(a). There are

force sensors inside the armrests and touch sensors on

the handles. When a user wants to move, they have to

put their arm on the arm-rest and grip the handles.

Using the two sensors, our device judges whether the

user is ready to movement. A gripping switch is

provided on each handle, as shown in Fig. 2(b). This

switch has two input steps that can be changed by the

strength used for the grip. Usually, in emergency

situations, elderly people tend to release the control

switch or push it strongly because of the fear of falling

(Maki et al., 1991). Therefore, we use the two-step

switch in such conditions, as shown in Fig. 2(b), and

our device provides assistance for standing/sitting

only in the case of the first step, whereas in the case

540mm

Gas Spring

Linear

Actuator

965mm

500mm

Actuated Wheel

Front Caster

Rear Caster

ICINCO 2019 - 16th International Conference on Informatics in Control, Automation and Robotics

490

of the second step, our device regards the user as

being in an emergency situation.

Figure 2: Our assistive walker.

3 SITTING MOTION

TORERANCE

3.1 Difference between Sitting Motion

and Standing Motion

From previous works, a sitting motion is same to

“reverse” motion of standing (Ehara et al., 1996). In

our preliminary experiment, we assist a sitting

operation with this reverse motion using our

prototype. Subjects are 6 young people and 2 elderly

people. As the result, all subjects feel fear of falling

and a reverse motion seems to be unsuitable for a

sitting assistance. Thus, in this paragraph, we

analyze a standing motion and a sitting motion

which the nursing specialist recommends.

For analysis, we assume a standing and a sitting

motions are symmetrical and we discuss the motion

as movement of the linkages model on 2D plane

(Nuzik et al., 1986). We measure the angular values

among the linkages, which reflects the relationship

of body segments using a motion capture system.

The angular value is derived using the body

landmark as shown in Fig. 3(a). Furthermore, we

measure the position of the center of pressure (COP)

using a force plate system (MG-100, ANIMA Corp.,

Japan) as the index of body stability. The

coordination is shown in Fig. 3(b). Subjects are 6

young healthy people and they operate both motions

based on the recommended motion by the nursing

specialist.

Fig. 4 shows the angular value of each joint and

Fig. 5 shows the position of COP during a sitting

motion. In Fig. 3(b), the Y-axis shows the angular

values of the pelvis and trunk, knee and ankle,

whereas the X-axis shows the movement pattern

(Chugo et al., 2014), which is the ratio of the

standing motion, as shown by (1). Fig. 4(a) is sitting

and Fig. 4(b) is reverse tracks of standing for easy to

analysis.

(1)

In equation 1,

is the time required for completion

of the sitting down operation and

is the present time.

From Fig. 4(a), in a sitting motion, the subject

lowers his trunk at one motion (10-60[%] movement

pattern). On the other hand, in Fig. 4(b), the subject

keeps his trunk around 10-25[%] movement pattern.

In a sitting motion, the subject inclines his trunk and

lowers it earlier than in case of a standing motion.

Furthermore, in a sitting motion, the subject inclines

his trunk larger than in case of a standing motion.

These features are same to previous reports (Dubost

et al., 2005).

(a) Side view (b) Its coordination

Figure 3: Experimental Setup. (I), (II) and (III) are force

plates. We change the height of chair (IV), according to the

subjects.

(a) Sitting motion (b) Its coordination

Figure 4: Angular values of each joint during a motion

recommended by nursing specialists.

From Fig. 5, the tracks of COP in both motions

are different. In a sitting motion, the position of COP

moves slowly than in case of a standing motion. From

100

120

140

160

180

200

0 25 50 75 100

Movement Pattern (%)

Angle (deg

)

P elvis/ Trunk

Knee

Ankle

100

120

140

160

180

200

0 25 50 75 100

Movement Pattern (%)

Angle (deg

)

Pe lvis /Trunk

Knee

Ankle

Sitting Assistance that Considers User Posture Tolerance

491

tracks of the knee angle in Fig. 4(a), around 60[%]

movement pattern, we can verify that the subject sits

down the target chair. In Fig. 5, the position of COP

moves slowly, especially, around 60[%] movement

pattern. This means the subject puts his hip on the

target chair and moves his weight from his foot to his

hip.

Figure 5: Position of COG during a motion.

In general, a sitting motion does not require the

physical strength as a standing motion, because

sitting is lifting down body movement to from

standing to sitting position according to the gravity.

From these results, we can assume the subject may

lower his trunk with rough path plan and coordinate

his body balance by inclination of his body according

to the process of a sitting motion. Therefore, the

required conditions for sitting assistance are follows.

 The robot should allow the patients sitting by

their intended motion in the safety range.

 The robot should help the patient if the

patient’s posture are the outside of safety range

and have high risk for falling down.

3.2 Sitting Posture Conditions

In our previous works, we investigate the posture

condition from the viewpoint of body dynamics

during standing (Yokota et al., 2019). However,

sitting and standing are completely different motion

and required conditions for sitting assistance are also

different as the previous paragraph. Considering

above, the sitting posture should be fulfilled by three

conditions from the viewpoint of body dynamics.

 Stability condition: The patient should be able

to keep their body balanced in this posture. This

study defines the condition as follows: the

position of the center of gravity (COG) should

be located within the range of the patients

footprint, while keeping the body balanced

during sitting down.

 Muscle condition: The patient should be able to

control their body motion in this posture. In

general, the output force generated by muscles,

changes according to the human posture

because the positional relationship between the

muscles and bones changes with the adopted

posture (Chugo et al., 2014). This means an

unsuitable posture cannot generate a sufficient

upper direction output force for proceeding

through with the sitting motion. This study

defines this condition as follows: the output

force of the muscles listed in Fig. 6 should not

exceed the muscle's maximum output during

sitting.

 Landing condition: The patient should be able

to control the sitting posture at landing the

seating surface. When landing, the patient

should reduce lowering velocity enough

because strong impact between buttocks and

the seating surface has high risk of injuring

(Yamamoto et al., 2015). This study defines

this condition as follows: the output force of the

muscles should not exceed the muscle's

maximum output when stopping the sitting

motion just before landing the seating surface.

This paper investigates the tolerance level, which

fulfills these three conditions through computer

simulation studies using OpenSim, a human motion

dynamics simulator package. In this simulation, we

used a 3DGait-Model2392 (Opensim documentation,

2018) as human model and modified its body

parameters to fit a typical Japanese elderly person

(Okada et al., 1996). The sitting motion was based on

the references recommended by nursing specialists

(Kamiya, 2005) as shown in Fig. 4(a).

(a) muscles (b) coordination

Figure 6: Human Model.

Generally, the sitting motion consists of three

phases, as shown in Fig. 7(a). Thus, we set a variation

30[deg] range on the reference posture at the

end of each phase (Postures (A)-(C) in Fig. 7(b)) in

the computer simulation. Note posture (D) is the final

posture and therefore we did not set a variation on this

phase.

-0.5

-0.4

-0.3

-0.2

-0.1

0.1

0 20406080100

Movement Pattern (%)

Position of COP (m)

Sitting

Reverse of Standing

ICINCO 2019 - 16th International Conference on Informatics in Control, Automation and Robotics

492

(a) sitting phase (b) references

Figure 7: Simulation Setup.

3.3 Sitting Posture Conditions

Fig. 8 shows the acceptable position of point P,

identified in Fig. 6(b), derived from the computer

simulation. Figs. 8(a), 8(b) and 8(c) show the

acceptable tolerance at 30[%], 50[%] and 70[%]

movement during the pattern of the sitting motion.

Acceptable tolerance fulfills three required

conditions, stability condition, muscle condition and

landing condition.

The sitting motion will be realized within the

tolerance shown in Fig. 8(d) and in this range of

motion, the patient can physically achieve final sitting

posture. The reference tracks are the sitting motion by

Fig. 4(a) and dashed lines shows the tolerance. In

general, a sitting motion has high risk for falling

down to forward direction for elderly people (Maki et

al., 1991) and inclining to forward direction should be

avoided during sitting motion. In Fig. 8(d), there are

the tolerance in backward direction and this result fits

the knowledge of previous study (Dubost et al., 2005).

From this result, it is important to consider the

patient's capable muscle output force in the sitting

posture.

4 ASSISTANCE CONTROL

ALGORITHM

To allow patients to move their intended motion

during sitting down, our controller uses a

combination of damping control and position control.

Damping control can change the strength of assistive

power, thus, it can allow for an offset from the

reference pathway of motion, allowing the patient to

move freely during the sitting down process.

(a) 30% (b) 50%

Figure 8: Simulation Results.

Considering these characteristics, damping

control should be used in the tolerance discussed in

previous section. By contrast, position control is

useful for maintaining body posture, however, its

pathway is fixed. Thus, it is useful when the patient's

posture exceeds the acceptable range.

In our previous work (Chugo et al., 2017), we

proposed an assistance control algorithm based on

the voluntary movement of the patient. We know

from previous research (Yokota et al., 2019) that the

motion of the human body consists of voluntary

movements, which generate the total body motion,

and a posture adjustment action, which keeps the

body stable during motion. This means the robot

should only provide a force that assists the physical

activity in response to the voluntary movement of

the patient, and our proposed algorithm only assisted

the patient when physical strength was required for

doing a voluntary movement. However, the

previously reported algorithm did not consider the

variation in the range of movements during human

motion, so this paper extends this control algorithm

as follows:

 First, we defined the body movement vector

as (2). This shows the velocity direction of

point P (Fig. 4(a)), which is the COG of the

upper body. The position of P

()

ref ref

is a

motion reference point based on the sitting

motion recommended by nursing specialists.

Details regarding the generation of this

100

120

140

160

180

200

0 255075100

Movement Pattern (%)

Angle (deg

)

Pelvis/Trunk

Knee

Ankle

Phase1 Phase2 Phase3

(A)

(B)

(C)

(D)

0.5

0.6

0.7

0.8

0.9

0 0.05 0.1 0.15

Y Position [m]

X Position

[

]

Stability Condition is not fulfilled

Muscle Condition is not fulfilled

Landing Condition is not fulfilled

All conditions are fulfilled

0.5

0.6

0.7

0.8

0.9

-0.1 -0.05 0 0.05 0.1

Y Position [m]

X Position [m]

Stability Condition is not fulfilled

Muscle Condition is not fulfilled

Landing Condition is not fulfilled

All conditions are fulfilled

0.5

0.6

0.7

0.8

0.9

-0.15-0.1-0.05 0 0.05

Y Position [m]

X Position [m]

Stability Condition is not fulfilled

Muscle Condition is not fulfilled

Landing Condition is not fulfilled

All conditions are fulfilled

0.5

0.6

0.7

0.8

0.9

-0.08 -0.03 0.02 0.07 0.12

Y Position [m]

X Position [m]

Reference Tracks

Tolerance at 70%

Tolerance at 30%

Tolerance at 50%

Tolerance Range

Sitting Assistance that Considers User Posture Tolerance

493

reference point are given in our previous paper

(Chugo et al., 2014).

()

() () ()

0, , , , 1

ref

ref ref ref ref

pp p p

ref

ref ref ref ref

pp p p

xxsx

yysy

  



  



(2)

Furthermore, our robot has control references for

each actuator as detailed in (3), which realize the

designed sitting motion (2).

ref



is the motion

reference for a powered walker and

ref



is for an

assistance manipulator.

() () ()

0, , , , 1

ref ref ref

ref

rbt rbt rbt

ref

rbt

ref ref ref

ref

rbt rbt rbt

rbt

xxsx

yysy













(3)

 Second, we assumed the subject applies all

forces

user

at position P because the armrest

and the handle of our assistive robot are

connected rigidly. We can calculate

user

from

the force applied to the armrest

armrest

and the

handle

using force sensors in the robot's

body (Fig. 9(a)) as (4).

()

user armrest handle

=− +fff

(4)

(a) Applied force and body

movement vector

(b) Body motion and its

range

Figure 9: Voluntary movement during sitting.

 Third, we assumed the patient also applies a

force for doing a voluntary movement of their

own intention, therefore

user

shows a

voluntary component. At the same time, our

controller calculates a motion reference

ref

this posture (Fig. 9(a)) and refers its

investigated tolerance (gray area at Fig. 9(b)).

Our controller evaluates if

user

is within the

tolerance at this posture, the patient's motion

fulfills the both conditions as discussed in

section two.

 Finally, our robot controls two actuators by (5).

()

upref

rbt

upref

rbt

ref ref

rbt handle handle rbt rbt

ref ref

rbt armrest armrest rbt rbt

xBf f Kxx

yBf f Kyy

−−−−



(5)

where

upref

rbt

is the updated reference value that our

robot actually uses for delivering sitting assistance.

()

rbt rbt

is the actual position of the powered

walker and the assistance manipulator of our robot.

and

K in (5) are constants used to coordinate the ratio

between the damping and position controls.

0handle

and

armrest

are the forces the patient applies to the

assistance system before the patient sits.

In order to apply the damping control only when

the patient's motion fulfills both the stability

condition and the muscle condition, the coefficient

that validates the damping control mode is calculated

as (6).

B will be larger value if

user

locates on

the center of the tolerance and in this situation, it

fits

ref

. By contrast, the position control is always

useful because it helps the patient maintain a stable

posture during motion. Therefore, we set the

coefficient,

K which validates the position control

mode, to be constant. The values of

b and K were

determined experimentally.

()

ref

p user

ref

p user

B b if conditions are fuifilled

B if conditions are not fuifilled



⋅



⋅







(6)

Using these ideas, our controller sets the ratio of

the damping control mode to a larger value if the

patient's trajectory fits the expected reference

pathway. Thus, the patient can move freely as

intended if their posture is not largely different from

the reference posture.

ICINCO 2019 - 16th International Conference on Informatics in Control, Automation and Robotics

494

5 EXPERIMENT

5.1 Validation of Sitting Motion

Simulation

To confirm the accuracy of sitting motion simulation,

we compare actual EMG results with its simulation

results. We measure the surface electromyograms on

several body segments, motion data by motion

capture system and ground reaction force by force

plates during sitting motion. Motion and ground

reaction force data are used to realize the sitting

motion in the simulation. The subject is young man

(22 years old) who do not have physical handicap

(height and weight are 174[cm], 60.5[kg]).

Fig. 10 shows the muscle activities about vastus

medialis, which are acquired as accurate EMG and

calculated by simulation. Muscle activities expressed

in percentage. The simulation results have a strong

correlation with EMG results and these results show

that simulation results are trustworthy.

Figure 10: Muscle activity by EMG and simulation results.

5.2 Experiments with Our Prototype

We implemented our proposed idea to the prototype

(Fig. 1) and conducted a practical experiment with it.

To confirm the efficiency of our sitting assistance, we

tested three cases.

 Case1: Using only position control, without our

proposed idea.

 Case2: Using our proposed idea.

We used ten subjects and each subject attempted

all three cases, five times each. Subjects were elderly

whose care level are 1 or 2. As seen in Fig. 11, our

prototype succeeds to assist the sitting motion

according to the intended motion of the subject. In

case1, the subject clings our robot during sitting

motion. On the other hand, in case 2, our robot

follows the patient’s movement to backward direction

and the subject can incline her trunk.

(a) Without proposed controller (Case1)

(b) With proposed controller (Case2)

Figure 11: Sitting motion with our assistive walker (Subject

A). The therapist stands near the subject for safety reason

and he does not assist the subject.

Fig. 12 shows the position of position P during

sitting motion. In case 1, the subject's position fits the

reference trajectory and this means our robot does not

allow the patient to move freely as intended. All

subjects has large upset because the robot applies

assistance force for fitting the reference trajectory

accurately.

On the other hand, in case 2, the subject's position

does not fit the reference trajectory but is within the

investigated tolerance range. This mean our robot

evaluates that the patient's motion fulfills both the

stability, muscle and landing conditions, and accepts

the body motion of the patient even though it does not

fit the reference pathway. As the result, the robot does

not apply any unnecessary assistive force and

succeeds in allowing the subject to use intended

sitting motion.

(a) case1

without proposed method

(b) case 2

with proposed method

Figure 12: The position of P (defined as Fig. 6(b)) during

sitting motion. Without our proposed idea (a), the position

of COG does not fit the reference and our robot tries to fit

it. Therefore, there are large upsets. In contrast with case1,

with proposed idea (b), its position moves within the

tolerance range and our robot allows the patient to move

his/her intended motion.

100

0 20406080100

Muscle Activity ]%]

Movement Patten [%]

EMG

Estimated Value

0.5

0.6

0.7

0.8

0.9

-0.08 -0.03 0.02 0.07 0.12

Y Position [m]

X Position [m]

Reference Tracks

Subject A

Subject B

Subject C

Subject D

Subject E

Tolerance Range

0.5

0.6

0.7

0.8

0.9

-0.08 -0.03 0.02 0.07 0.12

Y Position [m]

X Position [m]

Reference Tracks

Subject A

Subject B

Subject C

Subject D

Subject E

Tolerance Range

Sitting Assistance that Considers User Posture Tolerance

495

6 CONCLUSIONS

This paper proposes a novel sitting assistance device,

which allows patients to move through intended

movement. To realize this, we investigated the

motion tolerance of the sitting posture, which fulfills

both body balance and muscle force condi-tions.

Furthermore, we proposed a novel assistance control,

which maintains body stability whilst using intended

body motion of its user during sitting.

ACKNOWLEDGEMENTS

This work is supported in part by a Grant-in-Aid for

Scientific Research C (16K01580) from the Japan

Society for the Promotion of Science and the

Matching Planner Program (VP29117940231) from

Japan Science and Technology Agency, JST.

This study received a generous support on the

nursing sitting assistance way from Prof. Tsuyoshi

Uematsu, Mr. Yoichi Kondo (physical therapist) and

staffs on Kunitachi Nursing Station, Tokyo, Japan.

And also we received a kind support on experimental

procedure with our prototype from Mr. Kazuyuki

Hareyama (occupational therapist) and staffs on

Sankei Building Well Care Co., Ltd., Tokyo, Japan.

REFERENCES

N. B. Alexander, A. B. Schultz, and D. N. Warwick, 1991.

Rising From a Chair: Effects of Age and Functional

Ability on Performance Biomechanics, In J. of

Geometry: MEDICAL SCIENCES, vol. 46, no. 3, pp.

M91–98.

M. A. Hughes, M. L. Schenkman, 1996. Chair rise strategy

in the functionally impaired elderly, J. of

Rehabilitation Research and Development, vol. 33, no.

4, pp. 409–412.

Cabinet Office, Government of Japan, 2016. KOUREISHA

HAKUSHO (The whitepaper on the aged society),

ISBN: 4865790519, pp.25, (in Japanese).

M. Hirvensalo, T. Rantanen and E. Heikkinen, 2000.

Mobility difficulties and physical activity as predictors

of morality and loss of independence in the community-

living older population. In J. of the American Geriatric

Society, Vol.48, pp.493-498.

K. Nagai, I. Nakanishi and H. Hanabusa, 2003. Assistance

of self-transfer of patients using a power-assisting

device. In Proc. of the IEEE Int. Conf. on Robotics and

Automation, pp.4008-4015.

A. Funakubo, H. Tanishiro and Y. Fukui, 2001. Power

Assist System for Transfer Aid. In J. of the Society of

Instrument and Control Engineers, Vol.40, No.5,

pp.391-395.

Y. Ehara and S. Yamamoto, 1996. Analysis of Standing up

motion, Ishiyaku Pub. Inc, pp.65-73.

J. Yoneda, 1998. Study of Difficulties in Sitting and

Standing in the Elderly for Fall Prevention, In J. of

Yamaguchi Medical Association, Vol.47, No.3, pp.47-

59, 1998.

D. Chugo, T. Yamada, S. Muramatsu, Y. Sakaida, S.

Yokota and H. Hashimoto, 2014. A Standing

Assistance based on a Load Estimation considering

with a Muscle Arrangements at the Human Leg, In

Proc. of IEEE Int. Conf. on Robotics and Biomimetics,

pp. 1517–1522.

D. Chugo, S. Kawazoe, S. Yokota, H. Hashimoto, T.

Katayama, Y. Mizuta and K. Atsushi, 2017. Pattern

Based Standing Assistance Adapted to Individual

Subjects on a Robotic Walker, In Proc. of 26th IEEE

Int. Symp. on Robot and Human Interactive

Communication, pp.1216-1221.

E. Maki, P. J. Holliday, and A. K. Topper, 1991. Fear of

falling and postural performance in the elderly, In J.

Gerontology, vol. 46, no. 4, pp. 123–131.

S. Nuzik, R. Lamb, A. Vansant, and S. Hirt, 1986. Sit-to-

Stand Movement Pattern, A kinematic Study, In

Physical Therapy, vol. 66, no. 11, pp. 1708–1713.

V. Dubost, O. Beauchet, P. Manckoundia, 2005. Decreased

trunk angular displacement during sitting down: an

early feature of aging, In Physical Therapy, Vol.85,

No.5, pp.404-412.

M. Yokota, S. Kawazoe, D. Chugo, S. Yokota, H.

Hashimoto, T. Katayama, Y. Mizuta and K. Atsushi,

2019. Standing Assistance Control considering with

Posture Tolerance of its User, In Proc. of 2019

IEEE/SICE Int. Symp. on System Integration

, pp.543-

547.

Y. Yamamoto and N. Kado, 2015. Analysis of the sitting-

down movement, In J. Kansai Phys. Ther., Vol.15,

pp.13-16, 2015.

Opensim documentation gait 2392 and 2354 models (2018),

Available at Apr., 2019. In https://simtk-

confluence.stanford.edu/display/OpenSim/Gait+2392

+and+2354+Models

H. Okada, M. Ae, N. Fujii and Y. Morioka, 1996. Body

Segment Inertia Properties of Japanese Elderly, In

Biomechanisms, No.13, pp.125-139.

K. Kamiya, 2005. Development and evaluation of life

support technology in nursing, In Proc. of 7th RACE

Symp., Research into Intelligent Artifacts for the

Generalization of Engineering, The University of

Tokyo, pp.116-121.

ICINCO 2019 - 16th International Conference on Informatics in Control, Automation and Robotics

496