ROBOTICS

Robotics is the engineering science and technology of robots, and their design, manufacture, application, and structural disposition. Robotics requires a working knowledge of electronics, mechanics, and software. A person working in this field is known as a roboticist. According to the Oxford English Dictionary, the word robotics was first used in print by saac Asimo!, in his science fiction short story "#iar$", published in %ay &'(& in Astounding )cience *iction. Asimo! was unaware that he was coining the term+ since the science and technology of electrical de!ices is electronics, he assumed robotics already referred to the science and technology of robots. ,owe!er, in some of Asimo!-s other works, he states that the first use of the word robotics was in his short story Runaround .Astounding )cience *iction, %arch &'(/0. 1he word robotics was deri!ed from the word robot, which was introduced to the public by 23ech writer 4arel 5apek in his play R.6.R. .Rossum-s 6ni!ersal Robots0, which premiered in &'/&. Although the appearance and capabilities of robots !ary !astly, all robots share the features of a mechanical, mo!able structure under some form of control. 1he structure of a robot is usually mostly mechanical and can be kinematic chain .its functionality became akin to the skeleton of a body0. 1he chain is formed of links .its bones0, actuators .its muscles0 and 7oints which can allow one or more degrees of freedom. %ost contemporary robots use open serial chains in which each link connects the one before to the one after it. 1hese robots are called serial robots and often resemble the human arm. )ome robots such as the )tewart platform use closed parallel kinematic chains. Other structures, such as those mimic the mechanical structure of humans, !arious animals and insects, are comparati!ely rare. ,owe!er, the de!elopment and use of such structures in robots is an acti!e area of research .e.g. biomechanics0. Robots used as manipulators ha!e an end effector mounted on the last link. 1his end effector can be anything from a welding de!ice to a mechanical hand used to manipulate en!ironment. 1he mechanical structure of a robot must be controlled to perform tasks. 1he control of a robot in!ol!es three distinct phases 8 perception, processing and action .robotic paradigms0. )ensors gi!e information about the en!ironment or the robot itself .e.g. the position of its 7oints or its end effector0. 6sing strategies from field of control theory, this information is processed to calculate the appropriate signals to the actuators .motors0 which mo!e the mechanical structure. 1he control of a robot in!ol!es !arious aspects such as path planning, pattern recognition, obstacle a!oidance, etc. %ore complex and adaptable control strategies can be referred to as artificial intelligence. Any task in!ol!es the motion of the robot. 1he study of motion can be di!ided into kinematics and dynamics. Direct kinematics refers to the calculation of end effector position, orientation, !elocity, and acceleration when the corresponding 7oint !alues are known. n!erse kinematics refers to

the opposite case in which required 7oint !alues are calculated for gi!en end effector !alues, as done in path planning. )ome special aspects of kinematics include handling of redundancy .different possibilities of performing the same mo!ement0, collision a!oidance and singularity a!oidance. Once all rele!ant positions, !elocities and accelerations ha!e been calculated using kinematics, methods from the field of dynamics are used to study the effect of forces upon these mo!ements. Direct dynamics refers to the calculation of accelerations in the robot. n!erse dynamics refers to the calculation of the actuator forces necessary to create a prescribed end effector acceleration. 1his information can be used to impro!e the control algorithms of a robot. n each area mentioned abo!e, researchers stri!e to de!elop new concepts and strategies, impro!e existing ones and impro!e the interaction between these areas. 1o do this, criteria for 9optimal: performance and ways to optimi3e design, structure, and control of robots must be de!eloped and implemented.

Three Laws of Robotics

HISTORY TIMELINE OF ROBOTICS

1921 ; 23echoslo!akian playwright 4arel 2pek introduces the word robot in the play R.6.R. 8 Rossum<s 6ni!ersal Robots. 1he word comes from the 23ech robota, which means tedious labor. 1938 ; 1he first programmable paint;spraying mechanism is designed by Americans =illard >ollard and ,arold Roselund for the De?ilbiss 2ompany.

1942 ; saac Asimo! publishes Runaround, in which he defines the 1hree #aws of Robotics. 1946 8 Emergence of the computer@ Aeorge De!ol patents a general purpose playback de!ice for controlling machines, using magnetic recording+ B. >resper Eckert and Bohn %auchly build the EC A2 at the 6ni!ersity of >ennsyl!ania 8 the first electronic computer+ At % 1, =hirlwind, the first digital general purpose computer, sol!es its first problem. 1948 8 Corbert =iener, a professor at %. .1., publishes 2ybernetics or 2ontrol and 2ommunication in the Animal, a book which describes the concept of communications and control in electronic, mechanicalm and biological systems. 1951 8 n *rance, Raymond Aoert3 designs the first teleoperated articulated arm for the Atomic Energy 2ommission. 1he design is based entirely on mechanical coupling between the master and sla!e arms .using steel cables and pulleys0. Deri!ati!es of this design are still seen in places where handling or small nuclear samples is required. 1his is generally regarded as the ma7or milestone in force feedback technology. 1954 8 Aeorge De!ol designs the first programmable robot and coins the term 6ni!ersal Automation, planting the seed for the name of his future company 8 6nimation. 1959 8 %ar!in %insky and Bohn %c2arthy establish the Artificial ntelligence #aboratory in % 1. 1960 8 6nimation is purchased by 2ondec 2orporation and de!elopment of 6nimate Robot )ystems begins. American %achine and *oundry, later known as A%* 2orporation, markets the first cylindrical robot, called the ?ersatran, designed by ,arry Bohnson and ?el7ko %ilenko!ic. 1962 8 Aeneral %otors purchases the first indi!idual robot from 6nimation and installs it on a production line. 1his manipulator is the first of many 6nimates to be deployed. 1963 8 Bohn %c2arthy heads up the new Artificial ntelligence #aboratory at )tanford 6ni!ersity. 1964 8 Artificial intelligence research laboratories are opened at % 1, )tanford Research nstitute .)R 0, )tanford 6ni!ersity, and the 6ni!ersity of Edinburgh. 1964 8 2DD Robotics founded. 1965 8 2arnegle %ellon 6ni!ersity establishes the Robotics nstitute.

1965 8 ,omogeneous transformations applied to robot kinematics 8 this remains the foundation of robotics theory today. 1967 8 Bapan imports the ?ersatran robot from A%* .the first robot imported into Bapan0. 1968 8 4awasaki licenses hydraulic robot design from 6nimation and starts production in Bapan. 1968 8 )R builds )hakey, a mobile robot with !ision capability, controlled by a computer the si3e of a room. 1970 8 >rofessor ?ictor )cheinman of )tanford 6ni!ersity designs the )tanford Arm. 1oday, its kinematic configuration remains known as the )tandard Arm. 1973 8 2incinnati %ilacron releases the 1E, the first commercially a!ailable minicomputer. 1974 8 >rofessor ?ictor )cheinman, de!eloper of the )tanford Arm, forms ?icarm nc., to market a !ersion of the arm for industrial applications. 1he new arm is controlled by a minicomputer. 1976 8 Robot arms are used in ?iking & and / space probes. ?icarm nc. incorporates a microcomputer into the ?icarm design. 1977 8 A)EA, a European robot company, offers two si3es of electric powered industrial robots. Foth robots use a microcomputer controller for programming and operation. 1977 8 6nimation purchases ?icarm nc. 1978 8 6sing technology from ?icarm, 6nimation de!elops the >6%A .>rogrammable 6ni!ersal %achine for Assembly0. 1he >6%A can still be found in many research labs today. 1978 8 Frooks Automation founded 1979 8 )ankyo and F% market the )2ARA . )electi!e 2ompliant Articulated Robot Arm0 de!eloped at Gamashi 6ni!ersity in Bapan. 1981 8 2ognex founded 1981; 2R) Robotics 2orp. founded. 1982 8 *anuc of Bapan and Aeneral %otors from 7oint !enture in A% *anuc to market robots in Corth America. 1983 8 Adept 1echnology founded.

1984 8 Boseph Engelberger starts 1ransition Robotics, later renamed ,elpmates, to de!elop ser!ice robots. 1986 8 =ith 6ninamation license terminated, 4awasaki de!elops and produces its own line of electric robots. 1988 8 )taubli Aroup purchases 6nimation from =estinghouse. 1989 8 2omputer %otion founded 1989 8 Farrett 1echnology founded 1993 8 )ensable 1echonologies founded 1994 8 2%6 Robotics nstitute<s Dante , a six;legged walking robot, explores the %t. )purr !olcano in Alaska to sample !olcanic gases. 1995 8 ntuiti!e )urgical formed by *red %oll, Rob Gounge, and Bohn *reud to design and market surgical robot systems. *ounding technology based on the work at )R , F% and % 1. 1997 8 CA)A<s %ars >athfinder mission captures the eyes and imagination of the world as >athfinder lands on %ars and the )o7ourner ro!er robot sends back images of its tra!el on the distant planet. 1997 8 ,onda showcases the >E, the Hth prototype in a humanoid design pro7ect started in &'HI. 2000 8 ,onda showcases Asimo, the next generation of its series of humanoid robots. 2000 8 )ony un!eils humanoid robots, dubbed )ony Dream Robots .)DR0, at Robodex. 2001 8 )ony releases second generation of its Alto robot dog 2001 8 Fuilt by %D Robotics of 2anada, the )pace )tation Remote %anipulator )ystem .))R%)0 is successfully launched into orbit and begins operations to complete assembly of nternational )pace )tation

THE BIRTH OF THE INDUSTRIAL ROBOTS After the technology explosion during =orld =ar , in &'JI, a historic meeting occurs between Aeorge 2. De!ol, a successful in!entor and entrepreneur, and engineer Boseph *. Engelberger, o!er cocktails the two discuss the writings of saac Asimo!. 1ogether they made a serious and commercially successful effort to de!elop a real, working robot. 1hey

persuaded Corman )chafler of 2ondec 2orporation in Danbury that they had the basis of a commercial success. Engelberger started a manufacturing company -6nimation- which stood for uni!ersal automation and so the first commercial company to make robots was formed. De!ol wrote the necessary patents. 1heir first robot nicknamed the -6nimate-. 1he first 6nimate was installed at a Aeneral %otors plant to work with heated die;casting machines. Also in &'I&, patent number /,''H,/EK was granted to De!olLthe first 6.). robot patent. Boseph Engelberger is widely considered the 9*ather of Robotics.: )ince &'KK, the Robotic ndustries Association has presented the annual Engelberger Robotics Awards to world leaders in both application and leadership in the field of robotics. INDUSTRIAL ROBOTS An industrial robot is officially defined by )O .)tandard HEKE@ &''(, %anipulating ndustrial Robots 8 ?ocabulary0 as an automatically controlled, reprogrammable, multipurpose manipulator programmable in three or more axes. 1he field of industrial robotics may be more practically defined as the study, design and use of robot systems for manufacturing . a top;le!el definition relying on the prior definition of robot0. An industrial robot is a general purpose reprogrammable machine possessing certain anthromorphic characteristics like mechanical arm, capability of responding to sensory inputs, to communicate with other machines, and make decision. ndustrial robots are being used more and more in many fields of industry where they are replacing human operators engaged in onerous, repetiti!e, or potentially ha3ardous 7obs. A particular point in fa!or of robots is that they can be taught to do 7obs which are not amenable to automation or mechani3ation n through con!entional approaches. Robot systems are certainly only one of the many possible means of automating and simplifying the production process. 1hey pa!e the way to a qualitati!ely new stage of automation, namely, the de!elopment of production systems which would require a minimum human attendance 1ypical applications of industrial robots include welding, painting, ironing, assembly, pick and place, palleti3ing, product inspection, and testing, all accomplished with high endurance, speed, and precision. %anufacturers of industrial robots include@ ntelligence Actuator, Adept, Epson Robots, Gaskawa;%otoman, AFF, E>)OC;)E 4O, igm Robotersysteme, 464A, 4awasaki and *AC62 Robotics. One of the main ad!antages of industrial robots is that they can be quickly reprogrammed to undertake tasks that differ in sequence and character of manipulation steps. Robots are therefore most cost;effecti!e in conditions of frequent product changes and in automation of 7obs requiring manual unskilled labor.

One of the main causes of the intensi!e de!elopment of robotics is the need to increase producti!ity in manufacturing. n comparison with con!entional automatic means, robots offer greater flexibility in sol!ing technical and organi3ational problems and shorten the time needed to complete and ad7ust automatic systems, and put them into ser!ice. Basic robot confi !rations" A robot manipulator is generally di!ided into two parts, first is body; and;arm assembly and other is wrist assembly, and it is also called end effector. 1his de!ice is either a gripper for holding a work part or a tool performing some process. Bo#$%an#%ar& confi !rations Cartesian or rectilinear configuration; >ositioning is done in the workspace with prismatic 7oints. 1his configuration is well used when a large workspace must be co!ered, or when consistent accuracy is expected from the robot.

*ig .a0 Cylindrical configuration; 1his robot configuration consist of a !ertical column, relati!e to which an arm assembly is mo!ed up and down. 1he robot has a re!olute motion about a base, a prismatic 7oint for height, and a prismatic 7oint for radius. 1his robot is well suited to round workspaces. 1he arm can be mo!ed in and out relati!e to the axis of the column.

*ig .b0

Polar or spherical configuration; 1wo re!olute 7oints and one prismatic 7oint allow the robot to point in many directions, and then reach out some radial distance.

*ig .c0 Articulated or oint spherical configuration; 1he robot uses E re!olute 7oints to position the robot. Aenerally the work !olume is spherical. 1his robot most resembles the human arm, with a waist, shoulder, elbow, wrist.

*ig .d0 !CA"A configuration; )2ARA is the acronym for )electi!e 2ompliance Robot Arm. 1his is similar to the 7oint arm or articulated arm configuration except that the shoulder and elbow axes are !ertical, which means that the arm is !ery rigid in !ertical direction, but compliant in the hori3ontal direction.

*ig .e0 1his permits the robot to perform insertion tasks for assembly in !ertical direction, where some side;to;side alignment may be needed to mate the two parts properly.

'rist confi !rations 1he robots wrist is generally for the orientation of the end effector. 1his usually consists of two or three degrees of freedom wrist assembly. 1he three 7oints are defined as RO##, > 12, and GA=. Roll is used to accomplish rotation about the robots arm axis, whereas pitch in!ol!es up and rotation and yaw ser!es the purpose of right and left rotation. Sensors in in#!stria( robots )ensors and actuators are used as control system components in industrial robots. 1here are two types; #nternal sensors; are those which are used for controlling position and !elocity of !arious 7oints. 1hese sensors form a feedback control loop with the control system. *or example; potentiometers, optical encoders, tachometers. $%ternal sensors; are used to coordinate operation of the robot with other components in the cell. 1hey are of following types; &actile sensors; used to determine whether contact is made between the sensor and the other ob7ect or not. Pro%i'ity sensors; these indicate the distance of the ob7ect from the sensor. t is also called Range sensor. (ptical sensor; used to detect presence or absence of ob7ects. Also used for proximity detection. *or example; photocells. )achine *ision; used for inspection and part identification. (ther sensors; these include measuring sensors of temperature, pressure, fluid flow, current, !oltage, etc. En# effectors 1hese enable the robot to perform specific tasks and are attached to the wrist of robot. 1here are two types; )ri**ers Are the end effectors used to grasp and manipulate ob7ects during work cycle. 1he ob7ects are usually work parts that are mo!ed from one place to another.

*ig .f0 1here are different types of grippers according to the shapes, si3es and weights of parts to be held. *or ex; mechanical grippers, !acuum grippers, magneti3ed grippers, dual, sensory multiple fingered grippers are used according to the application. Too(s 1hese are used in applications where the robot must perform some processing operation on the work part. 1herefore the robot manipulates the tool relati!e to stationery or slowly mo!ing ob7ects@ )pot welding gun, arc welding tool, spray painting gun, assembly tool, water 7et cutting tool, heating torch. n each case, the robot not only controls the relati!e position of tool with respect to work piece but also controls the operation of tool. n some applications, multiple tools are also used by the robots during the work cycle. *or ex; )e!eral si3es of drilling bits applied to the work part.

*ig .g0

Robot *ro ra&&in an# interfaces 1he setup or programming of motions and sequences for an industrial robot is typically taught by linking the robot controller to a laptop, desktop computer or .internal or nternet0 network. A robot and a collection of machines or peripherals is referred to as a workcell, or cell. A typical cell might contain a parts feeder, a molding machine and a robot. 1he !arious machines are -integrated- and controlled by a single computer or >#2. ,ow the robot interacts with other machines in the cell must be programmed, both with regard to their positions in the cell and synchroni3ing with them. !oft+are, 1he computer is installed with corresponding interface software. 1he use of a computer greatly simplifies the programming process. )peciali3ed robot software is run either in the robot controller or in the computer or both depending on the system design. 1here are two basic entities that need to be taught .or programmed0@ positional data and procedure. *or example in a task to mo!e a screw from a feeder to a hole the positions of the feeder and the hole must first be taught or programmed. )econdly the procedure to get the screw from the feeder to the hole must be programmed along with any MO in!ol!ed, for example a signal to indicate when the screw is in the feeder ready to be picked up. 1he purpose of the robot software is to facilitate both these programming tasks. 1eaching the robot positions may be achie!ed a number of ways@ Positional co''ands @ 1he robot can be directed to the required position using a A6 or text based commands in which the required N;G;O position may be specified and edited. &each pendant@ Robot positions can be taught !ia a teach pendant. 1his is a handheld control and programming unit. 1he common features of such units are the ability to manually send the robot to a desired position, or "inch" or "7og" to ad7ust a position. 1hey also ha!e a means to change the speed since a low speed is usually required for careful positioning, or while test;running through a new or modified routine. A large emergency stop button is usually included as well. 1ypically once the robot has been programmed there is no more use for the teach pendant. -ead./y.the.nose@ is a technique offered by many robot manufacturers. n this method, one user holds the robot-s manipulator, while another person enters a command which de;energi3es the robot causing it to go limp. 1he user then mo!es the robot by hand to the required positions andMor along a required path while the software logs these positions into memory. 1he

program can later run the robot to these positions or along the taught path. 1his technique is popular for tasks such as paint spraying. (ffline progra''ing @is where the entire cell, the robot and all the machines or instruments in the workspace are mapped graphically. 1he robot can then be mo!ed on screen and the process simulated. 1he technique has limited !alue because it relies on accurate measurement of the positions of the associated equipment and also relies on the positional accuracy the robot which may or may not conform to what is programmed Others n addition, machine operators often use user interface de!ices, typically touchscreen units, which ser!e as the operator control panel. 1he operator can switch from program to program, make ad7ustments within a program and also operate a host of peripheral de!ices that may be integrated within the same robotic system. 1hese include end effectors, feeders that supply components to the robot, con!eyor belts, emergency stop controls, machine !ision systems, safety interlock systems, bar code printers and an almost infinite array of other industrial de!ices which are accessed and controlled !ia the operator control panel. 1he teach pendant or >2 is usually disconnected after programming and the robot then runs on the program that has been installed in its controller. ,owe!er a computer is often used to -super!ise- the robot and any peripherals, or to pro!ide additional storage for access to numerous complex paths and routines.

A++LICATIONS OF INDUSTRIAL ROBOTS As this is the era of automation and computer aided manufacturing, industrial robots play a significant role in e!ery aspect of manufacturing. 1here are some properties of industrial robots which gi!e them the flexibility to work in any en!ironment such as accuracy, repeatability, multishift operation, easy assessment of infrequent changeo!ers in work cycle. )aterial handling. n this the robot mo!es material or parts from one place to other. 1o perform this transfer it is equipped with a gripper type end effector. 1he gripper design must be customi3ed as per the requirement. 1here are two cases included in this application namely material transfer and machine loading and unloading.

)aterial transfer.

1he basic application in this category is pick;and;place operation. #ow technology robots such as pneumatically powered are sufficient. *or ex; palleti3ing, stacking, insertion.

*ig .h0 )achine loading and0or unloading. n this the robot transfers parts into andMor from a production machine. 1he three possible cases are; machine loading in which robot loads parts of the production machine. )econd is machine unloading in which the raw materials are fed to the machine without using robot and it is used to unload finished products. =hile the third case is machine loading and unloading in this both loading of the raw product and unloading of finished work parts is done by robots ndustrial robot application of machine loading andMor unloading included in following processes@ 1ie casting; the robot unloads parts from the die casting machine. )ame is done in plastic moulding. )etal 'achining operations; here robots are used to load raw work piece material to the machine tool and to unload the finished product. )imilar kind of work is done by robots in forging, press working, heat treating and many more. Asse'/ly and inspection; 1his application may in!ol!e either the handling of material or the manipulation of tool. 1hese two are the labor intensi!e acti!ities traditionally and also boring and highly repetiti!e. Due to these reasons, abo!e acti!ities are logical candidates for robotic applications. Asse'/ly. t in!ol!es the addition of two or more parts to form a new entity. 1his is made secure by fastening two or more parts by different mechanical fastening techniques. Fecause of the economic importance of assembly,

automated methods are often applied. 1he most appealing application for industrial robots for assembly is where a mixture of similar products are produced in the same work cell or assembly line.

*ig .i0 #nspection. 1here is often need in automated plants and assembly systems to inspect the work that is supposed to do. 1his performs following functions; %aking sure that the gi!en process is complete. Ensuring that the parts ha!e been added in assembly line as specified. dentifying defects in raw material and the end products. 1he robot performs loading and unloading tasks to support an inspection or testing machine. 1he robot also manipulates an inspection de!ice, as mechanical probe, to test the product. 1o perform testing, the part must be presented at the workstation in the correct position and orientation, and the robot automatically manipulates the de!ice as required. Processing applications. 1his application includes the use of industrial robots in different categories like different types of welding techniques, spray painting, !arious machining and other rotating spindle processes. 2elding. >erhaps the most popular applications of robots is in industrial welding. 1he repeatability, uniformity quality, and speed of robotic welding is unmatched. 1he two basic types of welding are spot welding and arc welding, although laser welding is done. )ome en!ironmental requirements should be considered for a successful operation.

*ig .70 Robots used in spot welding are usually large, with sufficient payload capacity to hold the hea!y welding gun. *i!e or six axes robots are preferred. ndustrial robots are also used to automate the continuous arc welding. !pray Coating. t makes use of spray gun directed at the ob7ect. 1he robots used for this application must be capable of continuous path control. Bointed arm robots most commonly used. 1he spray painting applications seems to epitomi3e the proper applications of robotics, relie!ing the human operator from a ha3ardous, albeit skillful 7ob, while at the same time increasing work quality, uniformity, and cutting costs. (ther processing applications. Drilling, Routing and other machining processes, grinding, wire brushing, water 7et cutting, laser cutting, ri!eting are some other processing applications which are performed by the industrial robots in manufacturing plants. Other than these applications there are many purposes ser!ed by the industrial robots.

LE))ED ROBOTS
=alking, running, 7umping, and skipping are some of the most sophisticated mo!ements that occur in nature, because the feet are quiet small and the balance at all times has to be dynamic+ e!en standing still requires sophisticated control. f one falls asleep on ones feet he falls o!er. 1he human stabili3es the mo!ement by integrating signals from@ P ?ision, which includes ground position and estimates of the fi rmness of the ground and the coefficient of friction. P >roprioception, that is, knowledge of the positions of all the interacting muscles, the forces on them and the rate of mo!ement of the 7oints.

P 1he !esicular apparatus, the semicircular canals used for orientation and balance. A !ery large number of muscles are used in a coordinated way to swing legs and the muscle in an engine consisting of a power source in series with an elastic connection. ?arious walking machines ha!e been de!eloped to imitate human legs, but none is as efficient as those of humans. E!en the walking of four;legged animals is also highly complex and quite difficult to reproduce. 1he history of interest in walking machines is quite old. Fut until recently, they could not be de!eloped extensi!ely, because the high computational speed required by these systems was not a!ailable earlier. %oreo!er, the motors and power storage system required for these systems are highly expensi!e. Ce!ertheless, the high usefulness of these machines can discount on some of the cost factors and technical difficulty associated with the making of these systems. =alking machines allow locomotion in terrain inaccessible to other type of !ehicles, since they do not need a continuous support surface, but the requirements for leg coordination and control impose diffi culties beyond those encountered in wheeled robots. )ome instances are in hauling loads o!er soft or irregular ground often with obstacles, agricultural operations, for mo!ements in situations designed for human legs, such as climbing stairs or ladders. 1hese aspects deser!e great interest and, hence, !arious walking machines ha!e been de!eloped and se!eral aspects of these machines are being studied theoretically. n order to study them, different approaches may be adopted. One possibility is to design and build a walking robot and to de!elop study based on the prototype. An alternati!e perspecti!e consists of the de!elopment of walking machine simulation models that ser!e as the basis for the research. 1his last approach has se!eral ad!antages, namely lower de!elopment costs and a smaller time for implementing the modifi cations. Due to these reasons, se!eral different simulation models were de!eloped, and are used, for the study, design, optimi3ation, and gait analysis and testing of control algorithms for artificial locomotion systems. BALANCE OF LE))ED ROBOTS 1he greatest challenge in building a legged robot is its balance. 1here are two ways to balance a robot body, namely static balance and dynamic balance. !tatic 3alance )ethods 1raditionally, stability in legged locomotion is taken to refer to static stability. 1he necessity for static stability in arthropods has been used as one of, if not the most important, reason why insects ha!e at least six legs and use two sets of alternating tripods of support during locomotion. Cumerous in!estigators ha!e discussed the stepping patterns that insects require to maintain static stability during locomotion. Get, few ha!e attempted to quantify static stability as a function of gait or !ariation in body form. Research on

legged walking machines pro!ided an approach to quantify static stability. 1he minimum requirement to attain static stability is a tripod of support, as in a stool. f an animal<s center of mass falls outside the triangle of support formed by its three feet on the ground, it is statically unstable and will fall. n the quasi;static gait of a robot or animal, the center of mass mo!es with respect to the legs, and the likelihood of falling increases the closer the center of mass comes to the edge of the triangle of support. n *igure K.& static balance is compared between six;legged and four;legged robotic platforms. 1he problem of maintaining a stable platform is considerably more complex with four legs than it is with fi!e, six, or more, since to maintain a statically stable platform there must always be at least three legs on the ground at any gi!en time. ,ence, with only four legs a shift in the center of mass is required to take a step. A six;legged robot, on the other hand, can always ha!e a stable triangleLone that strictly contains the center of mass. n *igure K.& two successi!e postures or steps are shown for a four; and six;legged robot. n *igure K.& .a0 the triangle for the fi rst posture is stable because it contains the center of mass, but for the second posture the center of mass must be shifted in order for the triangle to be stable. n contrast, for the six;legged robot in *igure K.& .b0 the center of mass can remain the same for successi!e postures.

1yna'ic 3alance )ethods Dynamic stability analysis is required for all but the slowest mo!ements. t was disco!ered that the degree of static stability decreased as insects ran faster, until at the highest speeds they became statically unstable during certain parts of each stride, e!en when a support tripod was present. )ix; and eight;legged animals are best modeled as dynamic, spring;load, in!erted pendulums in the same way as two; and four;legged runners. At the highest speeds, ghost crabs, cockroaches and ants exhibit aerial phases. n the hori3ontal plane, insects and other legged runners are best modeled by a dynamic, lateral leg spring, bounce centering the animal from side to side. 1hese models, and force and !elocity measurements on animals, suggest that running at a constant a!erage speed, while clearly a dynamical process, is essentially periodic in time. =e define locomotor stability as the ability of characteristic measurements .i.e., state !ariables such as !elocities, angles, and positions0 to return to a steady state, periodic gait after a perturbation. Robots can prosper from the aspects of animal dynamics in se!eral ways@ P Robot bodies can be designed to take ad!antage of potential kinetic energy transformations, and especially forward inertia.

P Robot legs can be designed to absorb, store, and then rerelease the energy of foot impact. P Robot legs can be arranged, like in the roach, to take ad!antage of self; stabili3ing forces, which in turn can lessen the complexity of controllers and impro!e the o!erall stability of the de!ices in dynamic situations. RESEARCH ON LE))ED MACHINES 1he scientific study of legged locomotion began 7ust o!er a century ago when #eland )tanford, then go!ernor of 2alifornia, commissioned Eadweard %uybridge to find out whether or not a trotting horse left the ground with all four feet at the same time. )tanford had wagered that it ne!er did. After %uybridge pro!ed him wrong with a set of stopmotion photographs that appeared in )cientific American in &HKH, %uybridge went on to document the walking and rurming beha!ior of o!er (Q mammals, including humans. ,is photographic data are still of considerable !alue and sur!i!e as a landmark in locomotion research. 1he study of machines that walk also had its origin in %uybridge<s time. An early walking model appeared in about &HKQ. t used a linkage to mo!e the body along a straight hori3ontal path while the feet mo!ed up and down to exchange support during stepping.1he linkage was originally designed by the famous Russian mathematician 2hebyshe! some years earlier. During the HQ or 'Q years that followed, workers !iewed the task of building walking machines as the task of designing linkages that would generate suitable stepping motions when dri!en by a source of power. %any designs were proposed, but the performance of such machines was limited by their fixed patterns of motion, since they could not ad7ust to !ariations in the terrain by placing the feet on the best footholds. Fy the late &'JQs, it had become clear that linkages pro!iding fixed motion would not suffice and that useful walking machines would need control.

%ilestones in the De!elopment of #egged Robots &HJQ 2hebyshe! Designs linkage used in early walking mechanism. 6ses stop;motion photography document running animals. to

&H'E Rygg

>atents human;powered mechanical horse. >atents hopping tank with reaction wheals that pro!ide stability. Eight;legged kinematic walks in outdoor terrain. and 2ontrol system double, and pendulums. machine

&'I& )pace Aeneral

&'IE 2annon, )chaefer

,igdon

balances single, limber in!erted

&'IH *rank and %cAhee

)imple digital logic controls walking of >hony >ony. Ruadruped truck climbs railroad ties under control of human dri!er. Fig %uskie, a &J,QQQ;ton walking dragline, is used for strip mining. ft mo!es in soft terrain at a speed of 'QQ ft.Mh.. Digital computer coordinates leg motions of hexapod walking machine. ,ybrid computer controls hexapod walker in 6))R. ,uman runners set new speed records on tuned track at ,ar!ard. ts compliance is ad7usted to mechanics of human Ruadruped machine climbs stairs and climbs o!er obstacles using simple sensors. 1he leg mechanism simplifies control. ,ydraulic biped quasidynamic gait. walks with

&'IH %osher AE

&'I' Fucyrus;Erie 2o.

&'KK %cAhee

&'KK Aurfinkel

&'KK %c%ahon and Areene leg.

&'HQ ,irose and 6metani

&'HQ 4ato

&'HQ %atsuoka

%echanism balances in the plane while hopping on one leg. =alking biped balances acti!ely in three;dimensional space. ,exapod carries human rider, 2omputer, hydraulics, and human share computing task. )elf;contained hexapod lifts and mo!es back end of pickup truck.

&'H& %iura and )himoyama

&'HE Aumerland

&'HE Odetics

-in4age used in early +al4ing 'achines =hen input crank AF rotates, the output point % mo!es along a straight path during one part of the cycle and an arched path during the other part. 1wo identical linkages are arranged to operate out of phase so at least one pro!ides a straight motion at all times. 1he body is always supported by feet connected to the straight;mo!ing linkage. #inkages of this sort, consisting of pi!ots and rigid members, are a simple means of generating patterned motion.

)echanical 5orse 1his was patented by #ewis A.Rigg. 1he stirrups double as pedals so the rider can power the stepping motions. 1he reins mo!e the head and forelegs from side to side for steering. Apparently this machine was ne!er built.

2al4ing &ruc4 1his machine was de!eloped by Ralph %osher at Aeneral Electric in about &'IH. 1he human dri!er controlled the machine with four handles and pedals that were hydraulically connected to the four legs.

6uadruped )achine

1esign for &hree.1i'ensional 5opping )achine

Basic ter&s !se# in Robotics

Accuracy ; 1his is determined by the resolution of the workspace. f the robot is commanded to tra!el to a point in space, it will often be off by some amount, the maximum distance should be considered the accuracy. 1his is an effect of a control system that is not necessarily continuous. Control "esolution . 1his is the smallest change that can be measured by the feedback sensors, or caused by the actuators, whiche!er is larger. f a rotary 7oint has an encoder that measures e!ery Q.Q& degree of rotation, and a direct dri!e ser!o motor is used to dri!e the 7oint, with a resolution of Q.J degrees, then the control resolution is about Q.J degrees .the worst case can be Q.JSQ.Q&0. Coordinates ; 1he robot can mo!e, therefore it is necessary to define positions. Cote that coordinates are a combination of both the position of the origin and orientation of the axes.

1egree of 7reedo' ;Each 7oint on the robot introduces a degree of freedom. Each degree of freedom can be a slider, rotary, or other type of actuator. Robots typically ha!e J or I degrees of freedom. E of the degrees of freedom allow positioning in ED space, while the other /or E are used for orientation of the end effector. I degrees of freedom are enough to allow the robot to reach all positions and orientations in ED space. J degrees of freedom requires a restriction to /D space, or else it limits orientations. J degrees of freedom robots are commonly used for handling tools such as arc welders. t is usually the same as the number of axes. 8u'/ers of a%es 8 two axes are required to reach any point in a plane+ three axes are required to reach any point in space. 1o fully control the orientation of the end of the arm .i.e. the wrist0 three more axes are required. )ome designs .e.g. the )2ARA robot0 trade limitations in motion possibilities for cost, speed, and accuracy. -in4s and 9oints ; #inks are the solid structural members of a robot, and 7oints are the mo!able couplings between them.

*ig.&0 1here are fi!e types of 7oints &0 linear 7oint Ttype #U /0 E0 (0 J0 orthogonal 7oints Ttype OU rotational 7oint Ttype RU twisting 7oint Ttype 1U re!ol!ing 7oint Ttype ?U

(rientation A%es ; Fasically, if the tool is held at a fixed position, the orientation determines which direction it can be pointed in. Roll, pitch and yaw are the common orientation axes used. #ooking at the figure below it will be ob!ious that the tool can be positioned at any orientation in space.

*ig ./0 Position A%es ; 1he tool, regardless of orientation, can be mo!ed to a number of positions in space. ?arious robot geometries are suited to different work geometries. Payload ; 1he payload indicates the maximum mass the robot can lift before either failure of the robots, or dramatic loss of accuracy. t is possible to exceed the maximum payload, and still ha!e the robot operate, but this is not ad!ised. =hen the robot is accelerating fast, the payload should be less than the maximum mass. 1his is affected by the ability to firmly grip the part, as well as the robot structure, and the actuators. 1he end of arm tooling should be considered part of the payload.

*ig .(0 "epeata/ility; 1he robot mechanism will ha!e some natural !ariance in it. 1his means that when the robot is repeatedly instructed to return to the same point, it will not always stop at the same position. Repeatability is considered to be SM;E times the standard de!iation of the position, or where ''.JV of all repeatability measurements fall. 1his figure will !ary o!er the workspace, especially near the boundaries of the workspace, but manufacturers will gi!e a single !alue in specifications. !ettling &i'e ; During a mo!ement, the robot mo!es fast, but as the robot approaches the final position is slows down, and slowly approaches. 1he settling time is the time required for the robot to be within a gi!en distance from the final position. !peed ; refers either to the maximum !elocity that is achie!able by the 12>, or by indi!idual 7oints. 1his number is not accurate in most robots, and will !ary o!er the workspace as the geometry of the robot changes .and hence the dynamic effects0. 1he number will often reflect the maximum safest speed

possible. )ome robots allow the maximum rated speed .&QQV0 to be passed, but it should be done with great care. 2or4 en*elope02or4space ; 1he robot tends to ha!e a fixed and limited geometry. 1he work en!elope is the boundary of positions in space that the robot can reach. *or a 2artesian robot .like an o!erhead crane0 the workspace might be a square, for more sophisticated robots the workspace might be a shape that looks like a Wclump of intersecting bubbles. &ool Centre Point :&CP; ; 1he tool centre point is located either on the robot, or the tool. 1ypically the 12> is used when referring to the robots position, as well as the focal point of the tool. .E.g. the 12> could be at the tip of a welding torch0 1he 12> can be specified in 2artesian, cylindrical, spherical, etc. coordinates depending on the robot. As tools are changed we will often reprogram the robot for the 12>. *ig .(0

*ig .E0