A System Reducing Pressure of Buttock while Seating Using Pneumatic Actuators

— Pressure controlled pneumatic actuator system on the seat of a chair is proposed in this research. In the system, nine pneumatic actuators are allocated on the surface of a chair and inner pressures of them are controlled to alleviate the concentration of contact pressure between skin of a user and surface of the seat. Fundamental characteristics of the developed pneumatic actuator are evaluated experimentally. In addition, construction of the control system of them is introduced. The effectiveness of the adjustment of inner pressure of the actuators to alleviate the concentration of contact pressure is confirmed through experiments.


I. INTRODUCTION
In recent years, due to growing number of aging people, demand for welfare equipment to support elderly people's daily life has increased. As one of the welfare equipment to assist humans to move independently, the wheel chair is attracted attention since it has a wide use for not only the elderly people but also the people with lower body injuries or disabilities. Along with this, it is concerned that occurrence of damages in the buttock of the user who sits in a wheelchair over a long period of time.
Sitting over a long period of time, blood vessels that nourish the skin may be squeezed by the weight of the user and the reaction force from the contact surface of the seat of the wheelchair. This causes the blood flow slowing or stopping, and leads to damage to the surrounding tissue, i.e. pressure sore or pressure ulcer.
There are mainly two effective precautions of sore and ulcer incidence: changing the patient's position in seat every 2 hours; and dispersing the skin pressure concentration at buttock. In the latter precaution, a usual approach is to disperse the body pressure concentration by a mat made of pressure absorbing materials, such as urethane form (1). It is also reported about the evaluation of a cushion for wheelchair (2) and the usage for wheelchair (3). These are so-called passive methods to realize dispersion of the body pressure concentration by the softness and deformation of the mat, not active methods to effect the person from the mat. By this precaution, since the pressure concentration occurs after maintaining the same sitting posture for a long period of time, sitting position change is still necessary.
In this study, we propose a soft actuator system that can provide a desired deformation and give a soft touch to skin by controlling inner pressure. We construct a pneumatic seat with the developed actuators and insert it between buttock and seat surface of a chair. It is expected that it can actively alleviate skin pressure concentration of buttock by changing the shape of a plurality of the actuators by controlling inner pressure. For this purpose, we propose a control algorism to adjust the inner pressure of the pneumatic actuator. To confirm the effectiveness of the proposed approach, experiments to change skin pressure distribution of buttock from the condition that concentration of sitting pressure occurs are conducted.

II. SEAT BASED ON PNEUMATIC ACTUATORS
The proposed seat in this study is constituted with 9 pneumatic actuators arranged in lattice shape as shown in Fig.  1. The deformation can be controlled by adjusting the inner pressure of each actuator independently when a user sits on it. Figure 2 shows a scene when a subject sits in the seat. 20 subjects aged from 22 to 49 subjects tried this seat and all the subjects said that this seat was comfortable to sit on.
The diagram of this system configuration is shown in Fig. 3. The inner pressure of the actuator in this system is controlled by the electro-pneumatic proportional valve. The electropneumatic proportional valve can change inner pressure of the actuator in a range of 0.005 to 0.9 MPa proportional to the operating voltage of signal from the computer. This study is aimed to change the pressure distribution of buttock by changing pressure values at the contact portion between buttock skin and actuators. The change of pressure values at the contact portion is realized by increasing or decreasing the inner pressure of pneumatic actuators. Therefore, it is necessary for the pneumatic actuator to have the ability to adjust the displacement by changing inner pressure while human weight is loaded from the top.

A. Material and shape of a pneumatic actuator
The shape and softness of the pneumatic actuator are very important to give humans appropriate contact force and to make humans feel being comfortable. It is desired for the pneumatic actuator that its contact pressure to skin changes smoothly when its softness and shape are changed by pneumatic control. Among various kinds of materials, silicon rubber is ideal because of its flexibility and hardness. As shown in Fig. 4, silicon rubber with the hardness of 40 is effective because it is not too hard for buttock skin and can transmit force to the skin to generate appropriate displacement with it. The silicone rubber with the hardness of 40 is determined as the material of pneumatic actuator.
Using silicon rubber of hardness 40, a pneumatic actuator with a hemispherical shape is fabricated as shown in Fig. 5. In the case of non-contact, this structure is uniformly expanded radically when inner pressure is applied, and in the case of contacting with the buttock skin, it is expected to generate acting force with no bias against the contact surface.

B. Identificatin of characteristics of pressure-diplacement of a pneumatic actuator
It is important to identify characteristics between pressure and displacement of a pneumatic actuator. In order to identify it, the following experiment is conducted.
The experimental system is shown in Fig.6. The vertical displacement of the pneumatic actuator is measured by an invisible light laser displacement sensor. The invisible light laser sensor is composed of the combination of the sensor head and the amplifier unit. The performances of the senseor are as follows: reference distance: 300 mm, measurement range: 100 mm, responsibility: 10Hz, and the resolution of measurement is 0.5 mm.       Using the electro pneumatic proportional valve, the inner pressure of the actuator is controlled with high accuracy by the compressed air generated by the compressor. As shown in Fig. 7, loaded with a weight in the state of no supplied, i.e. with the inner pressure of 0.0 MPa, the pneumatic actuator is stepwise pressurized from 0 to 50 seconds and depressurized from 50 to 90 seconds by the electro-pneumatic proportional valve with the values shown in Tab. 1 in each period of time.
The displacement of the position of the actuator surface is measured in this process. Where, the displacement is defined as the change of the position of the actuator surface from the original position in the case of no air supplied and no weight loaded.
The mass of weights loaded to the pneumatic actuator changes from 0 to 9.6 kg is given as shown in Tab. 2. The vertical displacement of the pneumatic actuator in each case is measured.
The average weights of the subject participated in this experiment male is 66.13 kg. Assuming the average weight of an adult male of 66.13 kg is loaded on the seat surface of 9 actuators equally, weight on the surface of an actuator is estimated to be 7.24 kg. The load range in this experiment includes the average value.  Figure 8 shows the change of the displacement for the stepwise change of inner pressure of the pneumatic actuator. From this figure, it is found that the displacement increases as the actuator pressure increases and decreases as the actuator pressure decreases at all weight levels. In addition, it is found that, in pressurizing process (from 0 to 50 seconds), the displacement overshoots a little bit to time, and then settled quickly to a steady state. Such overshooting behavior cannot be observed in depressurizing process (from 50 to 90 seconds).
Furthermore, displacement of the actuator surface from initial state reaches to a maximum value of 50 mm when the inner pressure is set to 0.03 MPa under the load of 9.7 kg.
Next, the hysteresis characteristics of the proposed actuator is examined. The displacement of pressurizing and depressurizing process under the load of 7.4 kg to the pneumatic actuator is shown in Fig. 9.
As shown in the figure, the relation between pressure and displacement has slight hysteresis in pressurizing and depressurizing process. The maximum error of displacement is 2.71 mm when the pressure is set to 0.01 MPa. In the case of other weights, almost the same hysteresis characteristics can be found, and the maximum errors are smaller than 2.71 mm.  From these results, the proposed actuator is considered to be possible to obtain almost the same displacement in either pressurizing or depressurizing process. From these results, it is expected that almost desired displacement can be obtained by pressurizing and depressurizing the proposed pneumatic actuator for the people with average or less weight sitting on pneumatic seat.

III. A SYSTEM ON DISPERSION OF SITING PRESSURE WITH FEEDBACK BY SENSORS
When a user sits on the proposed seat, there are various factors, such as the sitting position, the size of the body, the body weight and the sitting posture, etc., influencing on the pressure concentration of buttock skin. Therefore, to alleviate the pressure concentration of buttock skin under various conditions, a controlling system based on the feedback by pressure sensors is proposed.
As shown in Fig. 1, the proposed seat in this study is constituted with 9 pneumatic actuators arranged in lattice shape. The inner pressure of each actuator is controlled independently. The top view of the seat is shown in Fig. 10, and the 9 pneumatic actuators are numbered from 1 to 9 as shown in the figure.
The system constitution of feedback control of sitting pressure is shown in Fig. 11. First, the skin contact pressure between the buttock and the pneumatic seat is measured by a skin pressure distribution measurement system and is sent to a Windows system computer. The data of measured skin pressure is sent to an ART-Linux system computer every 5 [sec] through the TCP / IP communication.
Based on the skin pressure data, the concentration area of skin pressure can be approximated as an elliptic equation, and the center of the approximated ellipse is calculated. According to the relative position of the center of the approximated ellipse to the actuator, two kinds of controlling process are conducted by the computer: as shown in Fig. 12(a), when the position of the center is located directly above the actuator, the relaxation of pressure concentration is performed by reducing the inner pressure of the actuator. Otherwise, when the center of the ellipse is not positioned directly above the actuator as shown in Fig. 12(b), the dispersion of pressure concentration is conducted by increasing inner pressure of the actuators which surround the center of the ellipse.

DISPERSION SYSTEM
The seat pressure dispersion system proposed in this study is evaluated experimentally. A subject is seated in the direction from back to front in Fig. 10. The pressure distribution of the buttock skin is obtained by the skin pressure measurement system.

A. Experimental setup
The experimental setup used in this experiment is shown in Fig. 13. The parts1 in Fig. 13 is a sensor sheet for detecting the skin pressure of the buttock. Pressure-sensitive area of this sensor sheet is 440 × 480 [mm], the number of sensors is 2112, and the resolution is 1 piece/cm 2 . The data of skin pressure measured by this sensor sheet is sent to the computer (parts 4).   In addition, parts 2 in Fig. 13 are the pneumatic actuators which are introduced in Chapter 2. The inner pressure of the actuators is controlled with high accuracy by the computer (parts 5) installed with ART-Linux system using the electropneumatic proportional valve ITV (parts 3).

B. Experimental method
It is important to confirm the validity of the proposed seat pressure dispersion system under various conditions. As an approach for this, skin pressure dispersion system is evaluated when the trunk of the body is inclined to the right or the left, that is, the body weight loads mainly on one side. Here, the experiment is conducted when the trunk is inclined to the right.
At first, a subject is seated obliquely on the pneumatic seat in which the inner pressure of all of the pneumatic actuators is set to 0.01 MPa. The skin pressure distribution is measured once a minute. Based on the measured data, the relaxation and dispersion of pressure concentration are conducted follwoing the proposed algorithm in Chapter Ⅲ .
When the position of the center of ellipse is located directly above one actuator, the inner pressure of that actuator is reduced to 0.0 MPa. When the center of the ellipse is not positioned directly above the actuator, the inner pressure of the actuators which surrounds the center of the ellipse are increased to 0.025 MPa. This pressure control process continues for 11 minutes, and the change of skin pressure distribution is observed.

C. Experimental results
The experimental results are shown in Fig. 14. The original positions of the pneumatic actuators are indicated by circle dotted lines in the figure of the skin pressure distribution of 0 minutes. The skin pressure concentration areas in which the skin pressure value is more than 92.3mmHg are surrounded by elliptic black lines. Continuous pressure over 92.3 mmHg will cause the incidence of pressure sore or pressure ulcer. The areas surrounded by elliptic white solid lines show the areas depressurized from over 92.3 mmHg at 1min before to under 92.3 mmHg. In addition, mark "A" shows the skin pressure concentration areas not located directly above the actuators.
At 0 minutes, it is observed that the pressure concentration occurs at the right side of the buttock and the sacral region. It is considered to be the result of the right inclination of body. From 0 minutes to 1 minutes, it is found that the pressure concentration areas located directly above the actuators of 1, 2, 4, 7, 9 are depressurized to a certain level. From 4 minutes to 5 minutes, the same effect of alleviation of skin pressure can also be found from number 1 and 2 actuators.
Next, we focus attention on the skin pressure concentration which is not located directly above the actuator. From 1 minutes to 2 minutes, it is found that the skin pressure concentration area denoted by mark "A" is depressurized by increasing the skin pressure above number 1 and number 2 actuators.
In addition, from 7 minutes to 8 minutes, it is found that the skin pressure concentration is depressurized by increasing the skin pressure above number 1, number 2 and number 4 actuators.
From these results, it is confirmed that the skin pressure concentration, which is located directly above the actuator, is alleviated by reducing the actuator pressure, and that the skin pressure concentration, which is not located directly above the actuator, is dispersed by increasing the inner pressure of the surrounding actuators. Furthermore, it is also confirmed that the location of skin pressure concentration can be changed with the lapse of time.

V. SUMMARY
In order to alleviate pressure concentration of buttock skin, a pneumatic actuator that can actively disperse skin pressure is developed. Through the analysis of the pressure displacement characteristics of the actuator under given loads, it is confirmed that the displacement can be adjusted by controlling inner pressure of the actuator.
A seat pressure dispersion system is constructed with 9 developed actuators, a skin pressure sensor system and a computer for controlling the pneumatic pressure with the proposed algorithm. It is experimentally confirmed that the pressure concentration can be detected by the sensor system and can be alleviated effectively by controlling inner pressures of the actuators.