Pnömatik yapay kas tabanlı ve çalışma alanı değiştirilebilen esnek robotik kavrayıcı geliştirilmesi
Development of a pneumatic artificial muscle-based flexible robotic gripper with changeable workspace
- Tez No: 964450
- Danışmanlar: PROF. DR. AKIN OĞUZ KAPTI
- Tez Türü: Yüksek Lisans
- Konular: Makine Mühendisliği, Mechanical Engineering
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2025
- Dil: Türkçe
- Üniversite: Sakarya Üniversitesi
- Enstitü: Fen Bilimleri Enstitüsü
- Ana Bilim Dalı: Makine Mühendisliği Ana Bilim Dalı
- Bilim Dalı: Makine Tasarım ve İmalat Bilim Dalı
- Sayfa Sayısı: 159
Özet
Pnömatik yapay kaslar, hava basıncına maruz kaldığında kasılan ve biyolojik kasların doğal hareketlerine benzer şekilde tek yönlü çekme kuvveti uygulayan eyleyicilerdir. Her ne kadar pnömatik yapay kaslar, 1950'lerde McKibben yapay kasları adı altında keşfedilmiş olsa da günümüze kadar popülerlik kazanamamış, endüstriyel uygulamalarda geleneksel yöntemlerin yerini alamamıştır. Bugünlerde malzeme biliminin gelişiminden ivme kazanarak popüler hale gelen pnömatik yapay kaslar, bilimsel araştırmalarda tercih edilen bir alan olmuştur. Eğilen, burulan, kasılan gibi birçok çeşidi bulunan yapay kaslar, çalışma mantığı olarak aynı gözükseler de kendi içlerinde farklı kuvvet ve farklı kullanım alanları oluşturmaktadırlar. Bu çalışmada eğilen pnömatik yapay kas temelli esnek robotik kavrayıcı ele alınmıştır. Eğilen tür yapay kaslar, içerilerinde bulunan odacıklara basınç uygulandığında şişerek genişlemesi, uygulanan basınç kaldırıldığında ise eski haline geri dönmesi ile elde edilen çift yönlü eğilme hareketi oluşturulmasına olanak tanır. Eğilen pnömatik kasların bu özelliğinden faydalanarak, çalışma alanı genişletilebilen, kırılgan ve deforme olabilen cisimlere zarar vermeden farklı şekil ve boyutlardaki nesneleri kavrayabilen esnek robotik kavrayıcı geliştirilmiştir. Yapay kasın üretileceği malzemeyi belirlemek için farklı malzemelerden çekme testi numunesi üretilmiştir. Numunelerin üretimi için 3D yazıcı kullanılarak kalıplar oluşturulmuş ve silikon dökümler yapılmıştır. Ardından standartlara uygun çekme testleri yapılarak bu testler doğrultusunda belirlenmiş silikon malzeme ile yapay kasların üretimi yapılmıştır. Geliştirme sürecinde bilgisayar destekli tasarım metotları ile 3 farklı eğilen pnömatik yapay kas tasarlanmış ve eğilme performanslarını incelemek amacıyla sonlu elemanlar analizleri yapılmıştır. Tasarlanan kasları oluşturmak için 3D yazıcı kullanarak imal edilecek olan kalıplar kurgulanıp üretilmiştir. Üretilen yapay kaslar belirli testlere tabi tutularak farklı basınçlardaki kuvvet ve dönme açısı verileri elde edilmiştir. Deneysel çalışmadaki sonuçlar incelenerek kavrayıcıda kullanılacak olan nihai yapay kas belirlenmiştir. Nihai yapay kasın belirlenmesinin ardından silikon kasları kontrol edecek bir sistem tasarlanmış ve üretilmiştir. Bu bağlamda robotik kavrayıcının; 4 adet eğilen pnömatik yapay kasa sahip olması, çalışma alanının yatay olarak 50x50 mm ile 200x200 mm arasında, dikey olarak ise 330 mm kadar değişebilmesi, kütlesi 600 g'a kadar olan farklı yapılarda ve geometrilerdeki cisimleri kavrayabilmesi hedeflenmiştir. Kavrayıcının kontrolü için görüntü işleme algoritması kullanılmıştır. Sistem bir kamera yardımıyla manipüle edeceği nesnenin boyutlarını algılayarak ön kavrama pozisyonunu ayarlayabilmiştir. Tutuş pozisyonu optimum konuma getirildiğinde parmaklara basınç verilerek kavrama işlemi başlamış ve nesne zeminden ayrılıp yukarı doğru kaldırılmıştır. Esnek robotik kavrayıcı ile yapılan deneyler sonucunda; kavrayıcının maksimum 0,8 bar'da 600 g kütle kaldırabildiği, 600 g'dan fazla kütleler için step motorların yetersiz kaldığı, 12 mm'den ince nesnelerin kavranamadığı, nesne inceldikçe basıncın düşmesi ile kavrama veriminin arttığı ve yalnızca kare ve dikdörtgen kesitli nesnelerin manipüle edildiği gözlemlenmiştir.
Özet (Çeviri)
Pneumatic artificial muscles are actuators that contract when subjected to air pressure and exert a unidirectional pulling force, similar to the natural movements of biological muscles. Although pneumatic artificial muscles were discovered in the 1950s under the name McKibben artificial muscles, they have not gained popularity until today and have not replaced traditional methods in industrial applications. Nowadays, pneumatic artificial muscles, which have become popular by gaining momentum from the development of material science, have become a preferred field in scientific research. Artificial muscles, which have many types such as bending, twisting and contracting, create different forces and different usage areas within themselves, even though they seem to be the same in terms of working logic. In this study, a flexible robotic gripper based on a bending pneumatic artificial muscle will be discussed. Bending type artificial muscles allow the creation of bidirectional bending motion, which is achieved by inflating and expanding when pressure is applied to the chambers inside them, and returning to its original state when the pressure is removed. By utilizing this feature of bending pneumatic muscles, a flexible robotic gripper will be developed that can expand the working area and grasp objects of different shapes and sizes without damaging fragile and deformable objects. In this study, an automatic and manually controlled flexible robotic gripper based on a bending pneumatic artificial muscle, with 4 fingers, with a gripping range that can be changed in x, y and z axis, with a width and length between 50 mm and 200 mm, with a height between 12 mm and 330 mm, capable of grasping rectangular and square prism objects with a mass of 600 g or less, capable of positioning to the pre-grip position according to different objects with image processing, was designed and manufactured. Material selection was made for the fingers to be used in the flexible robotic gripper. Five types of RTV-2 molding silicones, 20 Shore VRM-520, 20 Shore VRM-620, 20 Shore Aksil-G20, 40 Shore VRM-540, 40 Shore Aksil-G40, were selected due to their availability in the market, low cost and biocompatibility. The selected materials were supplied and casting processes were carried out in the molds produced in order to be subjected to tensile tests. The resulting silicone plates were inserted into the cutting mold to produce a tensile test specimen similar to a dog bone. Tensile tests were performed on 4 specimens produced from each material group. Tensile strength and elongation data obtained from 4 specimens of each material group were averaged for each material group. When the results were compared, 20 Shore VRM-520 material, which is the most ideal material in terms of both flexibility and durability, was selected for finger production. For the selected material, 4 different force-extension graphs obtained by tensile test were converted into numerical data with PilotDigitizer program. The numerical data of 4 samples were averaged and converted into a single table. The stress-strain graph of 20 Shore VRM-520 material was created using the table created. This graph can be used in computer-aided simulations in future studies. After determining the material, the production of the fingers was started. Since there are many parameters in finger design, wall thicknesses were determined as variables. It was decided to produce 3 different types of fingers with 2 mm, 3 mm and 4 mm wall thickness. The dimensions other than the total height of the fingers remain constant. For this reason, as the wall thickness increased, the material consumption increased and the internal chamber volume decreased. In addition, the production was carried out in 4 stages, in parts, so that the fingers could be easily demolded. The design of the finger consists of 1 upper body, 1 inextensible base and 1 connector connection. In order to produce all these parts, plastic molds were designed and produced using a 3D printer. The molds consist of 4 parts: inner chamber mold, outer surface mold, bottom base mold and connector mold, one for each wall thickness. After the silicone raw material and catalyst are mixed and vacuumed, they are ready for use. Liquid silicone was first poured into the outer surface molds. Inner chamber molds were placed on the outer surface molds and liquid silicone was added to the empty parts. The upper body parts were left to cure in this way. Then the glass fiber strip was placed in the bottom base molds and liquid silicone was poured into the molds. The lower base parts were left to cure in this way. After curing, the upper body and lower base parts were removed from the molds and glued together with liquid silicone and left to cure. When the bonding process was completed, epoxy adhesive was applied to the air inlet of the finger, a pneumatic tube was attached, liquid silicone was poured on it using a connector mold and left to cure. When curing was completed, the fingers were ready for use. Computer-aided finite element analyses were conducted to examine the bending performance of three types of fingers with completed designs. Analyses were conducted using a monolithic silicone body and a 0.15 mm thick glass fiber geometry that fully contacted the lower surface of the body. The stress-strain values of the silicone material, previously determined through tensile tests, were defined in the simulation. To ensure inextensibility, the mechanical properties of the glass fiber were defined in the simulation. Because the silicone material is flexible, the analysis was performed using the Yeoh model, a hyperelastic material model requiring only uniaxial tensile testing. A nonlinear mesh was assigned to the geometries. Mesh elements were designated as hexahedrons. Tetrahedron and pyramid elements were allowed in the transition regions if necessary. After the necessary boundary conditions for the silicone body and glass fiber to move together were met, the fingers were fixed at the air inlet surface. Gravitational acceleration was defined perpendicular to the inlet surface and directed toward the fingertip. The internal chambers of the finger were pressurized starting from 0 bar pressure and increasing in 0.1 bar increments until the finger reached its maximum bending angle. All analyses were conducted for fingers with 2 mm, 3 mm, and 4 mm wall thicknesses. As a result of the analyses, the finger with a 2 mm wall thickness reached a maximum bending angle of 295° at a pressure of 0.5 bar. The finger with a 3 mm wall thickness reached a bending angle of 295° at a pressure of 0.8 bar. The finger with a 4 mm wall thickness reached a bending angle of 295° at a pressure of 1.3 mm. It was observed that the bending angle decreased as the wall thickness increased. In order to select the finger with the most efficient wall thickness, an experimental setup was built and the compressive strengths, holding forces and bending performances of the fingers were investigated. The experimental setup is controlled by a control interface through a computer. Pressure was applied to the finger types with gradual increments of 0.1 bar. It was observed that the finger with 2 mm wall thickness exploded at 1.1 bar pressure, the finger with 3 mm wall thickness exploded at 1.3 bar pressure, and the finger with 4 mm wall thickness exploded at 1.4 bar pressure. It was determined that the compressive strength increased with increasing wall thickness. In terms of compressive strength, the finger with 4 mm wall thickness was the best performing finger, while the finger with 2 mm wall thickness was the least performing finger. The bending performances of the fingers were monitored by pressurizing the fingers with gradual increases of 0.1 bar from the burst pressure of the fingers to 0.1 bar lower pressure. The bending angle was determined as the angle of the finger end surface with the ground and measurements were made. The finger with 2 mm wall thickness reached the maximum rotation angle at 0.5 bar by reaching the position where it could not rotate further. The finger with 3 mm wall thickness reached an angle of 295° at 0.8 bar. The finger with 4 mm wall thickness reached 295° angle at 1.3 bar. When the data were evaluated, it was determined that the bending angle decreased with increasing wall thickness. In terms of bending angle, the finger with 2 mm wall thickness was the most performing finger, while the finger with 4 mm wall thickness was the least performing finger. In the fingertip force experiments, the fingers were pressurized up to 0.1 bar lower than the burst pressure of the fingers with gradual increases of 0.1 bar. When the values measured with the force meter were analyzed, it was observed that the finger with 2 mm wall thickness produced a force of 4.08 N at 1 bar, the finger with 3 mm wall thickness produced a force of 3.31 N at 1.2 bar, and the finger with 4 mm wall thickness produced a force of 1.27 N at 1.3 bar. According to the collected data, it was observed that the fingertip force decreased as the wall thickness increased. In terms of fingertip force, the best performing finger was the finger with 2 mm wall thickness, while the least performing finger was the finger with 4 mm wall thickness. Considering all the experimental data obtained, it was decided to use the 3 mm wall thickness finger, which showed a balanced performance. A gripper was developed to control the fingers. The gripper mechanism is based on moving the fingers positioned on ball screws with stepper motors. Horizontal movement is provided by 1 stepper motor. 2 fingers connected on 2 motors move opposite and equal to each other and change the gripping range. One motor allows the fingers to move vertically. Thanks to the camera integrated into the gripper, the external dimensions of the object are detected and the fingers are brought to the pre-grip position according to the object. After the pre-grip position, the object is grasped and the contact with the ground is interrupted. An experimental setup was developed to measure the grasping ability. The manipulability of objects of different sizes was examined. As a result of the studies, it was observed that objects in the shape of a rectangular or square prism with a width and length ranging from 50 mm to 200 mm and a height ranging from 12 mm to 330 mm could be grasped. It was understood that as the height of the object decreases, the pressure value used should also decrease. However, the decrease in pressure also reduces the gripping force. In order to determine the load capacity of the gripper, the maximum mass that it can lift at 0.2 bar, 0.4 bar, 0.6 bar and 0.8 bar pressure was determined. The mass values that the gripper can lift efficiently were determined as 50 g at 0.2 bar, 200 g at 0.4 bar, 400 g at 0.6 bar and 600 g at 0.8 bar. Pressure values of 1 bar and 1.2 bar were not tested because the stepper motor providing vertical movement was insufficient after 600 g load.
Benzer Tezler
- Fog computing architecture for e-textile applications
E-tekstil uygulamaları için sis bilişim mimarisi
KADİR ÖZLEM
Doktora
İngilizce
2024
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrolİstanbul Teknik ÜniversitesiBilgisayar Mühendisliği Ana Bilim Dalı
DOÇ. DR. GÖKHAN İNCE
DOÇ. DR. ÖZGÜR ATALAY
- Pnömatik yapay kas tabanlı üst ekstremite manipülatöründe agonist-antagonist kontrol
Agonist-antagonist control in a pneumatic artificial muscle-based upper limb manipulator
FİLİZ ATAK METİNER
Yüksek Lisans
Türkçe
2024
Elektrik ve Elektronik MühendisliğiSakarya ÜniversitesiElektrik-Elektronik Mühendisliği Ana Bilim Dalı
PROF. DR. UĞUR ARİFOĞLU
- Doğadan esinli interaktif bina kabuğu tasarımı için örüntüye dayalı bir model
A model for nature inspired interactive building envelope design based on pattern
HÜLYA ORAL
Yüksek Lisans
Türkçe
2015
Mimarlıkİstanbul Teknik ÜniversitesiBilişim Ana Bilim Dalı
PROF. DR. GÜLEN ÇAĞDAŞ
- Yapay kaslar kullanarak hava araçlarında konum kontrolü
Location control in air vehicles using artificial muscles
MUSTAFA BAKIR
Yüksek Lisans
Türkçe
2019
Havacılık MühendisliğiErciyes ÜniversitesiSivil Havacılık Ana Bilim Dalı
DR. ÖĞR. ÜYESİ MUSTAFA SOYLAK
- Eklemeli imalatla üretilmiş pnömatik yapay kas ile el rehabilitasyon ortezi mekatronik sistem tasarımı
Mechatronic system design of a hand rehabilitation orthosis with additive manufactured pneumatic artificial muscle
EMRE TUĞBERK GÜLNERGİZ
Yüksek Lisans
Türkçe
2020
Mekatronik Mühendisliğiİstanbul Gedik ÜniversitesiMekatronik Mühendisliği Ana Bilim Dalı
DOÇ. DR. SAVAŞ DİLİBAL