# I.

Overview t is believed that a mechanism must be designed with a cosistent structure and invariant functions during operation. However, as the knowledge for mechanism design was surprisingly promoted in the past century and new design ideas are continuously evolutionary and born, people started to realize that a mechanism may be elegantly structured so that its configuration can be manipulated. The manipulation should be with multiple operation modes in order to collectively fulfil multiple tasks based on a sole mechanism. Because of the problem faced, people start to search to another alternatives, and the alternatives lead to creation of robot which can be reconfigured to adjust itself to suit to the environment. Such mechanism, which can change its configuration during operation, is normally so-called the reconfigurable mechanism.One of the objectives in the field of mobile robot is so that it can replace man to achieve a task whenever man are not capable of doing it. The building systems of a robot are created to operate in natural environment which are dramatic, unstructured and hazardous, which is very difficult for human. Nowadays, the reconfigurable mechanism have caught so many attention among people. It has also found excellent applications in various industries [1][2][3].

The study on reconfigurable mechanisms especially in structural design and kinematics has not been noticed even until the middle 1990s. From then on, some pioneering researches were started to surface.

After that, the attentions onto this topic gradually increase. Until today, the results of its relevant work have show progressing succesful work around this new mechanism concept.

Several new studies such as kinematotropic linkage, metamorphic mechanisms, mechanism with variable chains, and etc are leading the reconfigurable mechanism studies into a new level.

According to Merriam-Webster Dictionary, "configuration" is defined as "relative arrangement of parts or elements." Basically, the configuration looks at the relative arrangement. The relative arrangement that we discuss may be relative distance, relative angle, relative position, etc. The "parts or elements" are the links and joints of the mechanism and the "relative arrangement" are the relative positions, or further the relative motions, among the links and joints. Since the movement of the mechanism is depend not only by the arrangement of the joints but also their kinematic types, the configuration of the mechanism is described together by geometric arrangement and the kinematic types of its all kinematic joints [1-5].


# a) Reconfigurable to robots

In our times today, we can see a lot of handful of commercially available mobile robots, wether used in industry factory, house, or for research in the laboratory. Most of all cases, the robot get around on wheels, or perhaps tracks. For example, the remote-controlled bomb disposal robot. It cannot be denied that the wheels are definitely the most sensible choice for the environments that these robots run in. Wheels are very easy to control since it act predictably given a flat surface to run on. Other than cheap in cost, the power systems behind the locomotion are very well developed and understood. Besides, the wheel's speed can be easily configured just by controlling the amount of torque applied to the shaft connected to the wheels. Thus, it can make the robot to run faster [3][4][5][6].

However, wheels may not fully function in some cases. There are environments in which wheels become useless such as getting up to stairs. Although today's large buildings that have uses for robot have uses lift to accomodate the robot, it is not practical to implement it to the outside world. Some estimates suggest that half of the earth's surface is presently inaccessible to wheeled vehicles [4][5][6].

Thus, engineers take the idea of using the walking leg as another alternatives to overcome this problem. Plus, the legs can be reconfigured to adjust itself to the environment. It is quiet difficult for a wheeled robot to suddenly change the tyres to a different direction compared to walking leg robot [5].


# b) History

Starting from 1960's, most of legged machines have used a degree of computer control. Since its advent, this system technique has become greatly compacted and also be applied to the small machine. The idea of building systems by homogeneous components comes from von Neumann's "Theory of self reproducing cellular automata". His model assumes complete homogeneity of initial modules. This is an analogy to biological systems. In the system, the cell division from a single cell after fertilization produce the differentiated cells. However, the biological system has it's own weakness as it need a large number of cells in order to form their bodies [4][5][6][7]. (N. Inou etal, 2003) In 1988, a robot called CEBOT (Fukuda etal, 1990) was created and this is the first self-reconfigurable robot designed based on heterogenous components. It is composed of several different module. The module include transportation, rotational joint, telescopic arm, and grasping modules. Because of these modules combination, the CEBOT was able to perform a lot of different tasks [3] .

Following the heterogeneous module, the study of reconfigurable robot with homogeneous module takes place. In the homogeneous system, all the modules are identical. A set of common rules should sufficiently describe the differentiation and behaviour of each module. Any module can be replaced with another module as soon as this is realized. This property makes the self-repair and self-reproduction schemes become easier.


# c) Problem statement

People nowadays normally use a lot of help from the robot to do something that they cannot do by themselves. A lot of robot that we see in the market today have their own programme to do a task, but ufortunately every robot have their own limitation. As we can see today, the normal robot may be able to move into a certain direction only. Most of them cannot adapt to sudden change in environment. Whenever the surrounding change, such as from a floor to the staircase, the robot will become completely useless. In the end, people need to replace that robot with another robot that is built only for that task. This makes the time taken to complete the task become very slow. It such a waste of timeand a bit tiring.


# d) Objectives

This paper approach four central challenges-

To develop a robot that will be able to reconfigure if it stuck in a changed environment. To design the mechanical system for the robot. To design the control system so that the robot can move efficiently without any problem. To test the performance of the robot.


# II.


# Introduction

Reconfigurable robot is a robot where it can modify their shape and size in order to accomplish difficult mission. This capability is very desireble in many tasks. Such reconfigurable robot is very useful to overcome obstacles during doing the task. Reconfigurable robots offer a new approach in robotics.

The applications can be applied in many areas such as urban search and rescue, military reconnaisance and civil exploration [5][6][7].

This late few years, considerable progress has been made in the field of reconfigurable robotic systems. Normally, it would compromise with several segments that are connected with joint. The kinematic modes that we commonly see are multiple legs, wheeled and chain-track vehicle. According to study, the robot with multiple legs kinematic is less adapted compared to wheeled and chain-track vehicle robot. The multiple legs was said to be too complex since the structure has so many degree of freedom. Robots with wheeled and chain-track vehicle are usually portable since the adaptability to unstructured environment is high [6].

Usually for reconfigurable robot with multiple legs, the robots are facbricated based on modularity. Modularity consists of any number of identical interconnected units or modules. This system have advantages in term of manufacturing and robustness because of their homogeneity and redundancy. However in this type of kinematics, there are many degree of freedom (D.O.F) which make it complex and difficult to build. For example, a 14 degree of freedom (D.O.F) reconfigurable manipulator robot has been developed as a part of the Dockwelder EU project to perform in ship manufacturing industry.     This paper emphasizes the reconfigurable robot with four legs. It would not be complex enough since it will have several degree of freedom only. Each of the leg contain two link which will be connected through a servo motor. The principal of the movement of this robot is simple enough to understand. The servo motor will act as a joint and the movement of the legs will be based on the servo motor rotation or translation move.


# Figure 2.7 : Servo motor a) A chronological development of self reconfigurable robot

CEBOT is the first modular self-reconfigurable robot that have been ever created in history. The modules were considered cellular in structure [7].

ii. Fracta (1994) This robot (Murata etal,1994) is capable of reconfiguration, transportation in 2 dimensions, and selfrepair [8,9].    This robot (Hosokawa etal,1998)has climbing stairs-like structures and have a capability to reconfigure against gravity. This robot also has potential to be used as to build a bridge structure for transporting cargo across a gap. There were four prototype modules that have been built [11][12][13].

As we can see in the figure below, these type of robots (Unsal etal,1999)consist of active link and passive cube. The cubes can be rotated, translated simultaneously in two direction, and also act as pivot joint for a moving link. These type of robots are very ptential to move over obstacles, climb stairs, traverse through tunnels and pipes, manipulate objects, formbridges and towers, and be utilized for space applications [12][13][14]. Crystalline robots (Rus etal, 2000)are homogeneous square modules. Similar to muscles, this type of robot perform locomotion using expansion and contraction movements. Using a key and lock (channel) mechanism, the connection between the modules can be performed [14].


# ix. Polybot (2000)

This robot, polybot (Yim etal, 2000) was built from a chain structure consisting of segment and node modules. The structure make the robot capable of several types of locomotions, including rolling, earthworm motion, and spider-like motion. The modules are connected by using a pin/hole mechanism [13][14][15]. xi. Pneumatic (2002) The pneumatic (Inou etal, 2002) was inspired by animals such as worms and caterpillars with hydrostatic skeleton. The pneumatic modules are homgeneous, cubic-shaped structures with pneumatic actuators consisting of flexible bellows [15][16][17][18][19][20].

xii. Telecubes (2002) This type of robot (Vassilvitskii, 2002) perform motion by expanding and contracting the sides of the cube. Permanent switching magnet is used for the connection mechanism [14][15][16][17].

xiii. Chobie (2003) This robot, Chobie (Koseki etal, 2004) can reconfigure by means of slide motion and successive cooperative movements. This robot is potential in cooperative transportation, collection, and construction [15][16][17][18].

xiv. Deformatron ( 2006) It is a homogeneous 3D modular robot (Stoy, 2006)with modules. The modules can form in two structures which are rigid lattice and flexible chain. This robot has limitations which are no communication between modules and the modules do not contain power source. Therefore, it need to be controlled externally [15,16].


# xv. Odin (2007)

It is a hierarchical lattice-based robot (Stoy etal, 2007). It is built with two different types of modules which are cylinder-shaped and sphere-shaped. The link modules are able to communicate with neighbours, performing computation, and power sharing among modules [16,17].

xvi. Morpho (2008) Morpho robot is based on deformation and transegrity model of cellular structure. The cells exert expansion and contraction forces on the entire structure. There are four types of modules where this robot can be assembled. They are active link, passive link, surface membrane, and interfacing cubes [18]. xvii. Anatomy-based catomns (2008) The claytronics project (Goldstein etal, 2004) is an inspiration for this anatomy-based catomns (Christensen D.J., 2008) robot being built. Truthfully, this type of robot has not been physically created yet. However, several promising simulations have been done using Open Dynamics Engine. This type of robot is expected to be able to performing computation, sensing points of contact with neighbours, sensing the gravity direction and local actuation. This robot can assemble a muscle-actuated arm, and grasping objects by using whisker feedback [17][18][19][20]. xviii. Swarmorph (2009) The SWARMORPH (O'Grady etal, 2009) project involves the development of a distributed scheme forgenerating morphologies using autonomous selfassembling mobile robots [18,19].

This technique is applied to the swarmrobot (sbot) platform. However, there is a limitation of this selfassembling mobile robots. There is little control over the structure of the formed robot assembly. This work proposes a mechanism for control of the growth of morphological structures based on self-assembly. One by one, individual robots connect to the forming assembly.


# III.

Drawing of the Robot The main thing that should be consider when doing this project is about how the motion of the robot would be. In this project, after consulting with the supervisor, we agreed on building a robot with walking leg for this type of reconfigurable robot. A servomotor is a crucial part to determine the direction of the robot to move, wether forward or backward. The servo motor can be placed into the robot once the code have been made and have been transferred through arduino board. For the robot to turn the leg, we also need a motor to turn the bevel gear. The rotation motion from the bevel gear will be tranferred to a worm gear. Since the leg is attached to the worm gear, it will automatically turn the direction of the leg once the worm gear start rotating.

The robot also consists of inertial sensor that will detect when the robot fall or flip, so that the robot will be able to reconfigure and walk again.


# b) Force division between the legs

The whole body of the robot will stand on two legs while it is taking forward motion. Thus, the whole weight will be divided between these two legs equally so that the robot will stand balanced. 


# Conclusion

The design of the robot for this project is not achieved just by thinking once. The redesign and the development of this reconfigurable robot is always challenging. In the process of making this project, some fundamental design principles which are applicable to all components of the system became more clear. The very first important step that we should be aware of before beginning the implementation of a robot systemis to go through a thorough design, analysis and also the specification. It is crucial in order to verify wether the implementation will work or not.

Upon completing this project also, I also learned that the simplicity of the design should not be overlooked. As an engineer, it is our job to make something which is easy for humankind to handle but at the same time give a maximum output. In designing, creativity and experimentation are two of the most valuable design tools available until today. From the research that have been made, it can be concluded that the reconfigurable robot is one of the most desirable technology that people need today. 
1![Figure 1.1 : shows self-assembly and self-repair.(N.Inou etal, 2003)    ](image-2.png "Figure 1")
21![Figure 2.1: Picture of modular reconfigurable robot.](image-3.png "Figure 2 . 1 :")
23![Figure 2.3 : Picture of wheel and chain-track vehicle Nowadays, we can see a lot of robots designed by engineers to make it walk with four legs and even six legs.](image-4.png "Figure 2 . 3 :")
24![Figure 2.4 : Multi legs Robot](image-5.png "Figure 2 . 4 :")
25![Figure 2.5 : Multi legs RobotHowever, there are not much on focusing the reconfigurable legs. The objectives of this project is to improve the capability of the robots by making the legs of the robot to be adjustable according to the environment. The motor plays important role to realize this objective.](image-6.png "Figure 2 . 5 :")
26![Figure 2.6 : four legs robot](image-7.png "Figure 2 . 6 :")
28![Figure 2.8 : Fracta modules iii. Metamorphic (1996) Metamorphic robots (Pamecha etal, 1996)are homogenous, 2 dimensional, and lattice-based reconfigurable robots. It has square or hexagon-shaped modules. Figure below illustrated the metamorphic robot.](image-8.png "Figure 2 . 8 :")
29![Figure 2.9 : metamorphic robot with hexagon-shaped](image-9.png "Figure 2 . 9 :")
211![Figure 2.11 : 3D-Unit module v. Molecule (1998) The Molecule (McGray etal, 1998) is a robot based on 3D lattice. The modules comprised of two atoms connected by a bond while the modules are connected by electromagnets. Figure below shows the picture of molecule robot that was created[10,11].](image-10.png "Figure 2 . 11 :")
2![Figure 2.12 : molecule robot vi. Vertical (1998)This robot(Hosokawa etal,1998)has climbing stairs-like structures and have a capability to reconfigure against gravity. This robot also has potential to be used as to build a bridge structure for transporting cargo across a gap. There were four prototype modules that have been built[11][12][13].](image-11.png "Figure 2 .")
2![Figure 2.13 : i-cube link structure(Murata etal, 1998) viii. Crystalline (2000)Crystalline robots(Rus etal, 2000)are homogeneous square modules. Similar to muscles, this type of robot perform locomotion using expansion and contraction movements. Using a key and lock (channel) mechanism, the connection between the modules can be performed[14].](image-12.png "Figure 2 .")
214![Figure 2.14 : polybot modules in a snake-like structure](image-13.png "Figure 2 . 14 :")
3![Figure 3.1 : (a) and (b) show thethe 3D robot design in solidworks](image-14.png "Figure 3")
3333536![Figure 3.2 and 3.3 show the forces that are equally divided between the two sides of the frame. By using the formula : F1 + F2 = W F1 = F2 = W/2 c) Bill of materials](image-15.png "Figure 3 .Figure 3 . 3 FFigure 3 . 5 :Figure 3 . 6 :")
![1. Castano, A., Chokkalingam R., Will, P. Autonomous and Self-Sufficient CONRO Modules for Reconfigurable Robots, In Proceedings of the 5th International Symposium on Distributed Autonomous Robotic Systems, pp. 155-164, 2000. 2. Christensen, D. J., Campbell, J. Anatomy-Based Organization of Modular Robots, In Proceedings of the IEEE International Conference on Robotics &](image-16.png "")
1Research issue concerning self reconfigurablerobotThe part and theSelf-similar structure, boundary ofwhole diversityasystem,flexibility,multifunctionality; adaptable tochange according to environment.MorphogenesisLogic of growth from homogeneityto heterogeneity; computationalRobustnesscomplexitySelf-repair, graceful degradation,Self-reproductionscalabilityevolutionDriving force of evolution, novelmethod of production co-evolutionbetweenmorphology and motion;acceleration of evolutionSystemCentralized/decentralized,architechturehomogeneous/heterogenous,local/global communication
			© 2016 Global Journals Inc. (US)
			© 2016 Global Journals Inc. (US) vii. I-Cubes(1999)    x.Conro (2000)   
		
		
			

				


* 
	
		
	
	
		References Références Referencias Automation
		
			
			2008
		
	




* 
	
		Cellular Robotic System (CEBOT) as One of the Realization of Self-Organizing Intelligent Universal Manipulator
		
			TFukuda
		
		
			YKawauchi
		
	
	
		Proceedings of the IEEE International Conference on Robotics & Automation
				the IEEE International Conference on Robotics & Automation
		
			1990
			
		
	




* 
	
		Claytronics: A Scalable Basis for Future Robots (extended abstract)
		
			SCGoldstein
		
		
			TMowry
		
		
			2004
			In Robosphere
		
	




* 
	
		Self-Organizing Collective Robots with Morphogenesis in a Vertical Plane
		
			KHosokawa
		
		
			TTsujimori
		
		
			TFujii
		
		
			HKaetsu
		
		
			HAsama
		
		
			YKuroda
		
		
			IEndo
		
	
	
		Proceedings of the IEEE International Conference on Robotics & Automation
				the IEEE International Conference on Robotics & AutomationLeuven, Belgium
		
			1998
			23
			
		
	




* 
	
		
		
			HouxiangZhang
		
		
	




* 
	
		Zhicheng Den
		
			WeiWang
		
		
	




* 
	
		A Novel Reconfigurable Robot for Urban Search and Rescue
		
			GuanghuaZong
		
		
			& JianweiZhang
		
		
			2013
		
	




* 
	
		Cellular Robots Forming a Mechanical Structure(Evaluation of Structural Formation and Hardware Design of "CHOBIE II
		
			MKoseki
		
		
			KMinami
		
		
			NInou
		
	
	
		Proceedings of the International Symposium on Distributed Autonomous Robotic Systems
				the International Symposium on Distributed Autonomous Robotic Systems
		
			2004
			
		
	




* 
	
		Self-Assembling Machine
		
			SMurata
		
		
			HKurokawa
		
		
			SKokaji
		
	
	
		Proceedings of the IEEE, International Conference on Robotics & Automation
				the IEEE, International Conference on Robotics & Automation
		
			1994
			
		
	




* 
	
		
			SMurata
		
		
			HKurokawa
		
		
			EYoshida
		
		
			KTomita
		
		
			SKokaji
		
		
			Self-Reconfigurable Structure
		
		Proceedings of the IEEE International Conference on Robotics & Automation
				the IEEE International Conference on Robotics & AutomationLeuven, Belgium
		
			1998
			
		
	




* 
	
		Self-Reconfigurable Molecule Robots as 3D Metamorphic Robots
		
			CMcgray
		
		
			DRus
		
	
	
		Proceedings of the International Conference on Intelligent Robots and Systems
				the International Conference on Intelligent Robots and Systems
		
			1998
		
	




* 
	
		Group robots forming amechanical structure-development of slide motion mechanism and estimation of energy consumption of the structural formation
		
			NInou
		
		
			KMinami
		
		
			MKoseki
		
	
	
		inProc.IEEE Int. Symp. Computational Intelligence in Robotics and Automation
		
			2
			
			2003
		
	




* 
	
		Multirobot Morphogenesis Using Directional Self-Assembly
		
			RO'grady
		
		
			ALChristensen
		
		
			MDorigo
		
		
			Swarmorph
		
	
	
		IEEE Transactions on Robotics
		
			25
			3
			
			2009
		
	




* 
	
		Design and Implementation of Metamorphic Robots
		
			APamecha
		
		
			CChiang
		
		
			DStein
		
		
			GChirikjian
		
	
	
		Proceedings of the ASME Design Engineering Technical Conference and Computers in Engineering Conference
				the ASME Design Engineering Technical Conference and Computers in Engineering ConferenceIrvine, CA, USA
		
			1996
		
	




* 
	
		A Physical Implementation of the Self-Reconfiguring Crystalline Robot
		
			DRus
		
		
			MVona
		
	
	
		Proceedings of the IEEE International Conference on Robotics & Automation
				the IEEE International Conference on Robotics & Automation
		
			2000
			
		
	




* 
	
		The Deformatron Robot: a Biologically Inspired Homogeneous Modular Robot
		
			KStoy
		
	
	
		Proceedings of the IEEE International Conference on Robotics & Automation
				the IEEE International Conference on Robotics & AutomationOrlando, FL
		
			2006
			
		
	




* 
	
		
			KStoy
		
		
			ALyder
		
		
			RF MGarcia
		
		
			DChristensen
		
		
			Robots
		
		IROS Workshop on Self-Reconfigurable Robots, Systems, and Applications
				
			2007
		
	




* 
	
		I(CES)-cubes: A Modular Self-Reconfigurable Bipartite Robotic System
		
			CUnsal
		
		
			HKiliccote
		
		
			PKKhosla
		
	
	
		Sensor Fusion and DecentralizedControl in Robotic Systems II
				Boston, MA, USA
		
			1999
			3839
			
		
	
	Proceedings of SPIE




* 
	
		On the General Reconfiguration Problem for Expanding Cube Style Modular Robots
		
			SVassilvitskii
		
		
			JKubica
		
		
			ERieffel
		
		
			JSuh
		
		
			MYim
		
	
	
		Proceedings of the IEEE International Conference on Robotics & Automation
				the IEEE International Conference on Robotics & AutomationWashington DC, USA
		
			2002
		
	




* 
	
		PolyBot: A Modular Reconfigurable Robot
		
			MYim
		
		
			DGDuff
		
		
			KDRoufas
		
	
	
		Proceedings of the IEEE International Conference on Robotics & Automation
				the IEEE International Conference on Robotics & AutomationSan Francisco,CA, USA
		
			2000
			
		
	




* 
	
		Development of Pneumatic Cellular Robots Forming a Mechanical Structure
		
			NInou
		
		
			HKobayashi
		
		
			MKoseki
		
	
	
		Seventh International Conference on Control, Automation, Robotics, and Vision (ICARCV '02)
				Singapore
		
			December 2002