Proceedings of the 2005 IEEE International Conference on Robotics and Automation Barcelona, Spain, April 2005
Bio-Material Property Measurement System for Locomotive Mechanism in Gastro-Intestinal Tract Jiwoon Kwon, Sukho Park*, Byungkyu Kim and Jong-Oh Park Microsystem Research Center Korea Institute of Science and Technology, KIST P.O.Box 131, Cheongryang, Seoul 130-650, Korea {jwhj0814, shpark, bkim}@kist.re.kr,
[email protected]
Abstract - Recently, diseases in gastro-intestinal tract have drastically increased. As a result, endoscopic technologies are being developed to diagnose and treat these diseases. Biomaterial property is essential information to develop endoscopic devices especially capsule type endoscope. Because the capsule endoscope is moved by the peristaltic motion, it has some limitations to get the image of the digestive organ. Therefore, locomotive mechanism for capsule is necessary. In order to develop the locomotive mechanism, the information of biomaterial property is required. Especially, the friction force between capsule endoscope and the tissues of the gastro-intestinal tract is very important information. In this paper, we propose the bio-material property measuring system which can supply the information for the design of the locomotive mechanism. By using the proposed measuring system, we evaluate the effects of design parameters such as velocity, diameter size and shape of capsule endoscope that influence the friction force to the capsule endoscope to get the dominant parameters. As a result, we can offer the useful information to design the locomotive mechanism of the capsule endoscope. Index Terms - locomotion; earthworm-like robot; bio-material property; gastro-intestinal tract
I.
INTRODUCTION
In recent years, endoscopic technologies have been developed since disease for digestive system of latest human body increases gradually. Among the endoscopic technologies, a capsule endoscope is highlighted for the patient’s convenience and the possibility of the application in the small intestine. Because the movement of the capsule endoscope only depends on the peristaltic motion, it has some limitations to get the image of the digestive organ. Therefore, the research of the capsule’s locomotion is necessary. As the basic study of the locomotive mechanism, the information of bio-material property is required. Especially, the friction force of the tissues of the gastro-intestinal tract is very important information. In the bio-material properties of the digestive organ, several researches have been reported, which may be summarized as follows: Y. C. Fung investigated mechanical property of living tissues through several animal researches [1]. Also, J. Rosen and B. Hannaford showed similar relationship between general
0-7803-8914-X/05/$20.00 ©2005 IEEE.
1315
visco-elastic material and living tissue [2], and intravascular property detectable device is proposed by Mitsutaka Tanimoto and Fumihito Arai [3]. However, the above mentioned bio-material property measurement systems supply the limited information that is closely dependant on the measuring system. Therefore, they could not supply the information in order to design the locomotive mechanism. On the other hand, the locomotion mechanisms in gastrointestinal tract are proposed as follows: Firstly, rotating wheel has been the most conventional method for the robot locomotion [4]. In this case, the mobility is not sufficient because the wheel has disadvantage when the robot moves on uneven, slippery or flexible environments. Secondly, some of legged robot has been studied, but the mechanism is complex and hard to control [5][6]. In addition, the miniaturization of robots does not mean downsizing the existing macro technologies. In our previous work, we proposed the earthworm-like locomotive mechanism by using shape memory alloy (SMA) spring and silicone bellows [7]. This mechanism is simple and effective. Such mechanism enables the earthworm-like robot to move on any environments. Among the previous researches, we have decided that the earthworm-like locomotive mechanism is appropriate to the locomotion in the GI tract. In this paper, therefore, we propose the bio-material property measuring system which can supply the information for the design of the locomotive mechanism. By using the proposed measuring system, we can derive the dominant design variables and get the proper values of the variables. Therefore, the results of this test can offer the useful information to design the locomotive mechanism of the capsule endoscope.
II.
EARTHWORM-LIKE LOMOTIVE MECHANISM DESIGN IN GASTRO-ITESTINAL TRACT
A. Earthworm’s Locomotion Firstly, the structure and the locomotive mechanism of the earthworm are shown in Fig. 1. An earthworm’s body is made of segments. On each segment, except the first and the last,
there are pairs of tiny bristles called setae that help the worm move through the earth. The worm crawls by elongating to push the fore and by contracting to pull the hind part. Its wall has two kinds of muscles that are used to crawl. Circular muscles, surrounding the body, can make the body shrink or expand. Longitudinal muscles, mounted along the length of the body, can shorten and spread out the length of the worm. If the circular muscle expands, the setae are erected and then they prevent the worm from slipping [8]. The longitudinal muscles play a two-way linear actuator role and the setae play a clamping device role. This mechanism is simple and effective. Such mechanism enables the earthworm to move on any environments.
Figure 2. Principle of Earthworm-like Locomotive Mechanism
III.
BIO-MATERIAL PROPERTY MEASUREMENT SYSTEM
In order to supply the above-mentioned information for the design of the locomotive mechanism, we design and fabricate the bio-material property measurement system. The overall system is shown in Fig. 3. Figure 1. Structure & Locomotive Mechanism of Earthworm
B. Earthworm-like Locomotive Mechanism In previous study, we had proposed the biomimetic microrobot, the earthworm-like locomotive mechanism, in order to create an autonomous mobility. In addition, the newly designed a two-way linear actuator using SMA spring and silicone bellows are fabricated to construct the microrobot. Fig.2 shows the locomotive principle of a proposed microrobot. Silicone bellows acts as a bias spring to provide deformation force. The rear body of the microrobot moves forward when SMA spring is contracted by heat. The front needles clamp a contact surface and the rear needles slide forward. After the contraction of the SMA spring, the spring force of the silicone bellows makes the SMA spring elongate in cooling. Later, the rear needles clamp the contact surface and the front needles slide forward. Finally, the spring force of the bellow is equal to the SMA spring as the initial state. As the robot steps from (a) to (d) repeatedly, the microrobot can move forward. In order to optimize the earthworm-like locomotive mechanism, the bio-material property measurement system is manufactured. By using the measurement system, we measure the friction force between the capsule and the intestinal surface. In addition, we can find the friction effects of various design parameters under elongation and retraction stage.
1316
The system can be simply divided into the following units: test bench unit, motor control units, and data processing & controller unit. Each unit will be explained in great detail in sub-sections. In addition, the schematic diagram of the measuring system is shown in Fig. 4.
Data Processing & Controller
Test Bench
Motor Control
Figure 3. Overall System of the Measuring System
value of the load cell. The encoder has the resolution of 1000 pulses/rev. The motor rotates the ball screw and the lead of the ball screw is 2 mm/rev.
Figure 4. Schemetic Diagram of the Measuring System
A. Test bench unit The test bench unit is designed as Fig. 5 and is manufactured as Fig. 6. In order to measure the friction force between the head of the capsule and the intestinal surface, the test bench unit has two load cells in the directions of the elongation and the retraction, respectively. The one side of the load cell is connected with the mobile head of the capsule by using a tendon and the other side of the load cell is connected with the motor by using ball screw. Height Control Clamping Part Load Cell
Figure 6. Test Bench Unit
C. Data Processing & Controller Unit Finally, as the data processing unit, DAQ board (dSPACE 1103) system is used and it stored the measured friction force data from the load cell amp. Because the signals from the load cells are very low, the signal amplifiers for the load cells are necessary. The amplifier is designed for the conventional load cell and the offset and the output voltage level can be adjusted and the analog signals from the amps are converted as the digital data by using the 16 bits A/D converter in DAQ board. In addition, the motor position controller is included in DAQ board. The controller receives the encoder signal from the motor. The encoder signal is used as the position output and the feedback of the position controller. For the position control algorithm, the desired position trajectory of the motor is designed with a constant desired velocity. Therefore, this system enables to find the effect of the capsule velocity. And the conventional PID controller is used as the control algorithm.
DC Servo Motor
Guide
Ball Screw
Figure 5. Design of Test Bench Unit
When the intestinal specimen is installed in the test bench, according to the difference of the thickness of the specimen, the sensing line of the load cell and the center of capsule head are misaligned. The misalignment can be adjusted by the height control part. In addition, the test bench has the clamping part in order to fix the intestinal tract with the opposite head. By using the clamping part, we can assume that the capability of the clamp is perfect. Finally, by using the motor, the position of the load cell can be controlled with the desired constant velocity. This part will be deal with in the next section. B. Motor Control Unit Secondly, the motor control unit consists of DC servo motors, which have the encoder for measuring the position, the motor drivers, and the power supplier. The encoder signal is connected to the data processing unit and is used as the position
1317
IV. EXPERIMENT In order to acquire the useful information for the locomotive mechanism, the small intestinal specimen without the incision is used and is installed as in Fig. 7. The mobile head which is connected with the two load cells by using a tendon and the clamping head are located inside of the specimen. The one side of the specimen is fixed in the clamping part and the other side is connected to the guide structure of the load cell, which does not affect the sensing part of the load cell. In addition, in order to reduce the effect of the weight of the specimen, the guide structure of the tendon is necessary as shown in Fig. 7. When the load cell moves to the left, the mobile head of the capsule elongates and the specimen is easily slipped on the guide structure. We executed many experiments for variations of velocity, diameter size, shape, stroke and weight. Through these experiments, the friction forces between the head of capsule and the specimen are measured and thus the friction characteristics for the parameter variations can be found.
slip position is about 3~5 mm. After the slippage, the friction force still increases because the tensile force of the specimen behind the mobile head is added on the slip friction. That is, the specimen behind the moving head is extended by the tensile force. These phenomena are shown in Fig. 9. Guide Structure Figure 7. Experimental Setup
Tensile Force
A. Velocity Effect In order to investigate the effect of the velocity of the capsule, the experiments are executed for various velocities as 0.2, 0.4, 0.6, 0.8, and 1.0 mm/sec. The other parameters are used as constant values, which are stroke of 15 mm, the diameter of the head of 13 mm, the weight of 3.5 gf, and the radius of edge of about 1mm. The test results are shown in Fig. 8 (a)-(e) and the comparison of the results is shown in Fig 8 (f). In the given parameters, the friction force is gradually increasing and settling according to increasing the position of the head. The maximum friction force is about 75 ~85 gf. From the results of Fig. 8, we can conclude that the effect of the velocity is negligible.
Figure 9. Tensile Force of Specimen in Elongation
B. Diameter Size Effect The size of the capsule is an important parameter for the design of the earthworm-like robot. In general, as the diameter increases, the friction force will increase. However, the reduction of the diameter has some limitations due to the hardware size of capsule, such as a battery, an image sensor, and an actuator structure. Therefore, the effect of the diameter size of the capsule is investigated and shown in Fig. 10.
Figure 10. Experimental Results of Diameter Size Effect
Figure 8. Experimental Results of Velocity Effect
In addition, when the friction force is about 40 ~ 50 gf, and the position of the head is about 3~5 mm, the friction force has large variation of the slope. This variation of the slope is explained as follows: firstly, when the head initially starts, the head and the specimen are stuck and move together. While the position of the head and the specimen move together, the tensile force of the specimen between the clamping part and the moving head increases. And the tensile force coincides with the frictional force in the load cell. When the tensile force is lager than the stick friction force between the head and the specimen, the slippage between the head and the specimen occurs. The
1318
The diameter sizes of this experimental are 10, 13, and 15 mm. Especially, in the small intestinal specimen, the inner size of the specimen is similar to the diameter of the capsule. Therefore, the diameter size of the capsule has an important influence on the friction force. In accordance with the above expectation, the friction forces increases as the diameter increases. In addition, the maximum friction forces for the diameters 10, 13, 15 mm of the capsule are 45 gf, 80gf, and 115gf, respectively. This value can be helpful information for the design of the actuator size and its actuation force. C. Shape Effect Beside of the size of the capsule, its shape also influences on the friction force. In this paper, the tests with the various shape capsules cannot be executed. As a shape variation factor,
therefore, we choose the radius of the edge of the mobile head. The experimental results are shown in Fig. 11.
signals are also increasing and the maximum force is about 95 ~ 120 gf. This phenomenon is the effects of the tensile force of the specimen and it is already well explained in Fig. 8. Consequently, as the stroke increases, the force of the actuator of the earthworm-like robot should be increased in order to overcome the friction force. E. Weight Effect In order to check the effect of the weight of the capsule, we make the variation of the weight by using the additive mass as shown in Fig. 13.
Figure 11. Experimental Results of Shape Effect
Figure 13. Additive Masses
As the radius varies as R = 0, 1, and 2 mm, we can find that the friction forces severely increase. In order to reduce the friction force, therefore, the tests of the various shapes are necessary. D. Stroke Effect In general, the stroke of the capsule is considered as the displacement, which is able to move during one cycle. Therefore, it is related to the mobility of the capsule and it is regarded that the long stroke of the capsule is better than small one. However, the experiments of the variation of the stroke give us the different viewpoints in Fig. 12.
The weight of the original head is about 2.5gf and added masses of the head are 0.1 gf, 0.5 gf, and 1.0 gf. The results of friction force under variation of weight are very similar with each other. In other words, the difference of the capsule weight does not affect on the friction force. Therefore, the results are not presented in this paper. F. Retraction Friction Force We measure the friction force when the mobile head of the capsule is retracted after the elongation and the retraction forces are measured under variation of the diameter and the shape, which are illustrated as in section IV-B and IV-C. From the results in Fig. 14, the friction forces in the retraction are increased as the diameter decreases and the radius of the edge of the capsule increases. These are inverse results in the elongation in Fig. 10 and Fig. 11.
Figure 12. Experimental Results of Storke Effect
When the stroke of the capsule is about 15 mm, the friction force is shown as Fig.12 (a) and the maximum friction force is about 75 ~ 85 gf. On the other hand, when the stroke is about 25 mm, the friction force signals in Fig. 12 (b) are very similar to those in Fig. 12 (a) until the position reaches at 15 mm. However, as the position is larger than 15mm, the friction force
1319
Figure 14. Experimental Results of Retraction Friction Force
The results are explained as follows: in elongation, the friction forces increase as the diameter increases and the radius of the edge of the capsule decreases. As explained in Fig. 9, the friction forces increase as the tensile force behind the mobile head. However, if the direction of the mobile head is changed, the tensile force acts as the thrusting force of the mobile head. Therefore, the friction forces in the retraction decrease and the magnitudes of the friction forces are very small, compared with those in the elongation.
V.
In the future, by various experiments, we can derive the optimized design of the locomotive mechanism in gastrointestinal tract. ACKNOWLEDGMENT This work was supported by the 21st Century's Frontier R&D Projects, under the contract number MS-02-324-01, sponsored by the Ministry of Science and Technology, Korea.
CONCLUSIONS
REFERENCES
In this paper, we propose the bio-material property measuring system which can supply the information for the design of the locomotive mechanism. Especially, we focus on investigation some parameters to design earthworm-like robot. By using the proposed measuring system, we can derive the dominant design variables, examine their effect, and understand the various phenomena in the earthworm-like locomotion. These phenomena can be explained that, while the robot elongates, the friction forces increase as the diameter increases and the radius of the edge of the capsule decreases. On the other hand, the friction forces decrease under stage of retraction. Consequently, we conclude that the diameter size and the shape of the capsule are very important parameters and thus the parameters should be optimized. In addition, the stroke of the capsule has influence on the friction force. However, the effects of the velocity and the weight are negligible.
[1] [2]
[3]
[4]
[5] [6]
[7]
[8]
1320
Y.C. Fung, Biomechanics-Mechanical Properties of Living Tissues. Berlin, Germany: Springer-Verlag, 1993. J. Rosen, B. Hannaford, “Force Controlled and Teleoperated Endoscopic Grasper for Minimally Invasive Surgery-Experimental Performance Evaluation,” IEEE Trans. Biomed. Eng., Vol.46, No. 10, Oct. 1999. Mitsutaka Tanimoto, Fumihito Arai, “Micro Force Sensor for Intravascular Neurosugery and In-vivo Experiment,” IEEE Micro Electro Mechanical Systems Proceedings., The Eleventh Annual International Workshop, pp.504 - 509, Jan. 1998. Suzumori K., Miyagawa T., Kimura M., Hasegawa Y., “Micro Inspection Robot for 1-in Pipe,” IEEE/ASME Transaction on mechatronics,Vol. 4,No. 3, 1999. Pratt G.A., “Legged Robot at MIT : What’s New Since Raibert,” IEEE Robotics & Automation Magazine, 2000. J. Ryu, Y. Jeong, Y. Tak, B. Kim, B. Kim and J. O. Park, “A Ciliary Motion Based 8-legged Micro robot”, Proc. Of International Symposium on Micro dmechatronics and Human Science, pp.85~91, 2002. Y. Lee, B. Kim, M. Lee, and J. O. Park,”Locomotive Mechanism Design and Fabrication of Biomimetic Microrobot Using Shape Memory Alloy,” IEEE Conf. on Robotics and Automation, 2004. http://www2.worldbook.com/features/cybercamp/ html/walkworm.html
