Abstract— intention detection. Finally, somedesign approaches, guidelines and
Abstract— this chapter outlines important factors for the designprocess of robots. First, the challenges are discussed and possibleuser groups are detailed and categories of devices given. Themajor advanced design methods for the field of robotics areillustrated. This is due to linking between the machines and theuser and challenges of user intention detection. Finally, somedesign approaches, guidelines and best practices for thedevelopment of wearable devices are devices are discussed.I. INTRODUCTIONDefinition: Robot is an electromechanical device that isReprogrammable, multifunctional, sensible for environment.Robotics is a multi-disciplinary field, best robotics researchersand engineers will touch upon all disciplines:1. Mechanical Engineering: It’s concerned primarilywith manipulator/mobile robot design, kinematics,dynamics, compliance and actuation.2. Electrical Engineering: It’s concerned primarily withrobot actuation, electronic interfacing to computersand sensors, and control algorithms.3. Computer Science: It’s concerned primarily withrobot programming, planning and intelligentbehavior.Robotics mainly works on dirty tasks, repetitive tasks,dangerous tasks, impossible tasks and robot assisting thehandicapped.Robotics Timeline:1922 Czech author Karel Capek wrote a story called Rossum’sUniversal robots and introduced the word “Robota”(meaningworker).1954 George Devol developed the first programmable robot.1955 Denavit and Hatenberg developed the homogenoustransformation matrices.1962 Unimation was formed, first industrial robots appeared.1973 Cincinnati Milacron introduced the T3 model robot,which become very popular in industry.Robotic Terminologies:1. Robot: Mechanical device that performs human taskseither by automatically or remote control.2. Robotics: Study and applications of robot technology.3. Telerobotics: Robot that is operated remotely.Types of Robots:Mobile Robots: Robots move around on legs, tracks orwheels.Educational Robots: Robotics kits are extensively used ineducations e.g. robolab, lego etc.Domestic Robots: These are designed to performhousehold tasks and modern toys. Which are programmedto do things like talking, dancing.Two Famous Robots:1. PUMA (Programmable Universal Machine forAssembly) introduced in 1978.2. SCARA (Selective Complaint Articulated Robot forAssembly) introduced by Japan and the US(by AdeptTechnologies) in 1979.Laws of Robotics:According to Asimov proposed three laws of robotics.Law1: A robot may not injured human being or throughinaction, or allow the human being to come to harm.Law2: A robot must obey order given to it by human beings,expect where such orders would conflict with first law.Law3: A robot must protect its own existence as long as suchprotection does not conflict with first law.Classification of Robots:According to Robotics Institution of America, Robots areclassified in to four types.1. Variable Sequence Robot: A device that performs thesuccessive stages of a task according to apredetermined method easy to modify.2. Playback Robot: A human operator performs the taskmanually by leading the robot.3. Numerical Control Robot: The operator supplies themovement program rather than teaching the taskmanually.4. Intelligent Robot: A robot with the means tounderstand its environment and the ability tosuccessfully complete a task.II. ROBOT COMPONENTSRobot components can be listed below:1. Manipulator or Rover: Main body of the robot suchas links, joints, other structural elements of therobot).2. End Effectors: It is the part that is connected to thelast joint hand of a manipulator and end effectors arethe device at the end of the robot arms, designed tointeract with the environment.End effectors may consist of a grippers or tool. Thegrippers can be two fingers, three fingers or even fivefingers.3. Actuators: An actuators is a component of a machinethat is responsible for moving or controlling amechanism or system, for example for actuating avalve; in simple terms, it is a “Mover”.An actuator requires a control signal and a source ofenergy. The control signal is relatively low energyand may electrical voltage or current, pneumatic orhydraulic pressure or even human power.4. Sensors: To collect information about the internalstate of the robot or to communicate with outsideenvironment.Sensors provide awareness of the environment bysensing things and sensors are core of the robot. It isthe system that alerts the robot.Sensing can be in different forms likes – light, sound,heat, chemicals, magnetic and electric fields,resistance.5. Controllers: It is similar to cerebellum and itscontrols and coordinates the motion of the actuators.6. Processors: The brain of the robot and it’s calculatesthe motions and the velocity of the robot’s joints etc.7. Software: It’s includes Operating system, roboticsoftware and the collection of routines.8. Electric Motors: The vast majority of robots useelectric motors, often brushed and brushless DCmotors in portable robots or AC motors in industrialrobots and CNC Machines. These motors are oftenpreferred in systems with lighter loads, and where thepredominant form of motion is rotational.9. Air Muscles: Pneumatic muscles are also known asair muscles and air muscles are special tubes thatexpand up to 40% by using air. They are used insome robot applications.III. ROBOT PROGRAMMINGRobots are typically performed one of the following:1. Teach Pendant.2. Lead Through Programming.3. Robot Programming Languages.Use of Teach Pendant:In this method we are controlled the motion of the robots byusing the switches. Every moment or motion of the robots isrecorded in memory.Every robot actions executed by using playback mechanismand it is determined by a robot controller. This method issuitable for point to point applications.It is easy to use and no special programming languages arerequired for implementation.Lead through programming:It leads the robot physically through the required sequence ofmotions. Here trajectory and endpoints are recorded inmemory.In this method large amount of memory space is required andwhen a played back results in a smooth continuous motion.Programming languages:Programming languages need to interface the robot controlsystem to external sensors and to provide the real timechanges based sensory equipment.Robot computation is based on geometry of environment andit is ability to interface with CAD / CAM (Computer aideddesign/Computer aided manufacturing) systems.Large number of robot languages available such as AML ,VAL , AL, RAIL, Robot studio etc.Each robot manufacturer has their own robot programminglanguages and no standards exist. Programming languages canperform the meaningful task description and portability ofprograms virtually non-existent.A Manufacturing Language (AML):AML is a high level programming language and it isimplemented to manage RS/1Assembly robot.The RS/1Assembly robot incorporates with a mini – computers(IBM Series/1), 192 KB Memory, Matrix Printers, diskdrivers, display terminals, and key board.AML was developedby IBM Corporation for robot programming.Variable Assembly Language (VAL):VAL is a Robot programming language is adopted mainly forUnimation robots. It is designed with simple syntax andcapable of illustrating the robot functions easily.Its Includes two major tasks such as:1. Program instructions are used to provide VALPrograms in order to manage the robot functions.2. Monitor commands are used to execute the userwritten programs.RAIL:RAIL will be the best language for controlling two major taskssuch as the manipulation and vision systems. It is high levelrobot language based on Pascal and it will implement aMotorola-68000 central processor, tech pendent and terminal.This Language designed by automatic for arc welding andinspection purposes.RAIL comes with three different kinds of systems namely:1. Hitachi Process Robot- Arc Welding.2. Cartesian Arm- Assembly Functions.3. Vision System without arm.Artificial Language (AL):AL robot language was developed in Artificial IntelligenceLab at Stanford University. It is the Second generationLanguage based on simultaneous Pascal and the programswritten and executed on PDP-10.If program is developed with high level code, then it should bewritten in Stanford Artificial Intelligence Language (SAIL).And the AL systems include a big mainframe computer, and itgenerally runs on PDP 11/45 environment.The PDP 11/45 implement with one terminal, 128KB RAMmemory and floating point processors and this language hasgot the capability to control two Stanford Scheinman and TwoPUMA 600 arms simultaneously.RPL:RPL is a robot programming Language and its makes theimprovement, checking and correction of control algorithmsvery easy. It can be done even by an unskilled programmerlike machine operator, production engineer etc.The programs are typically written in BLISS-11 and run inRT-11. The DECPDP-10 cross complied into the LSI-11 orPDP-11.This robot language was designed by SRI International(Standford Research Institute).IV. CAD/CAM SYSTEMSIn computer integrated manufacturing (CIM) technology,CAD/CAM play an effective role and it ideally suited for thedesigning and manufacturing mechanical components of freefrom complex two dimensional and three dimensionalshapes.CAD/ CAM play an important role in functioning of robots. InCAD/CAM systems the robot work data is prepared fromCAD data from first design process.Computer aided design (CAD), computer aided manufacturing(CAM) can be defined as the digital computer to assist thedesigner in the creation, development, modification, analysis,optimization of a design and manufacturing activity.Interactive computer graphics (ICG) is an important part ofCAD systems and user oriented systems using computers tocreate, transform and display data in the form of pictures orsymbols.In design and analysis phase, CAD systems would be a bestsuited for drawing officers where frequent modifications arerequired in drawing and several parts repeat. And it mustremember that it very easy with computer to makemodifications and very fast to draw part profile ones its detailsare feed in computer.Below the figure shows design of a robot using CAD/CAMsystem.Benefits of CAD/CAM:1. CAD has enable creation of assemblies and parts incomputer, there analysis, optimization, stimulatingthe functionality, aesthetic requirements etc.2. It has resulted led time in the design office.3. Easy referencing and material of earlier design, dataand information.4. Dependency on design subcontractors is reduced.Limitation of CAD/CAM:1. There are two primary limitations to CAD/CAMrestorations. (Like Cerec and E4D) it is not yetpossible to do multiple bridges and esthetics islimited.2. The esthetic was improved dramatically from theearly days as the quality of materials has improved.Multi shade material blocks can duplicate dentin andenamel shades.3. CAD and CAM is not suited for highly estheticsituation.V. ARTIFICIAL INTELLIGENCEAccording to the father of artificial intelligence, JohnMcCarthy, it is “The science and engineering of makingintelligence machines, especially intelligent computerprograms”.Artificial intelligence is a way of making a computer, acomputer controlled robot, or the software thinkintelligently, and in the similar manner the intelligent humansthink.AI is accomplished by studying the human brain thinks andhow humans learn, decide, and work while trying to solve aproblem, and then using the outcomes of the study as a basisof developing intelligent software and systems.Goals of AI:1. To create expert systems- The systems which exhibitintelligent behavior, learn, demonstrate, explain andadvice it users.2. To implement Human intelligence in machinescreating systems that understand, think, learn andbehave like a humans.Artificial intelligence is a science and technology based ondisciplines such as Computer Science, Biology, Psychology,Linguistics, Mathematics and Engineering. A major thrust ofAI is in the development of computer functions associatedwith human intelligence such as reasoning, learning, andproblem solving.AI Technique:In the real world, the knowledge has some unwelcomedproperties:1. Its volume is huge, next to unimaginable.2. It is not well-organized or well-formatted.3. Its keeps changing constantly.AI technique is a manner to organize and use the knowledgeefficiently in such a way that-1. It should be perceivable by the people who provideit.2. It should be easily modifiable to correct errors.3. It should be useful in many situations though it isincomplete or inaccurate.AI techniques elevate the speed of execution of a complexprogram it is equipped with.Applications of AI:AI has been dominant in various fields such as-1. Gaming – AI plays crucial role in strategic gamessuch as chess, poker, tic-tac-toe, etc., where amachine can think a large number of possiblepositions based on heuristic knowledge.2. Natural language processing – It is possible tointeract with the computer that understands naturallanguage spoken by humans.3. Expert systems – There are some applicationsintegrate machines, software, and special informationto impart reasoning and advising. They provideexplanations and advice to the users.4. Vision systems – Though systems understand,interpret, and comprehend visual input on thecomputer. For example,a. A spying aero planes takes photographs, whichare used to figure out of spatial information ormap of the areas.b. Doctors use clinical expert system to diagnosethe patient.c. Police use computer software that can recognizethe face of criminal with the stored portrait madeby forensic artist.5. Speech recognition – Some intelligent systems arecapable of hearing and comprehending the languagein terms of sentences and their meanings while ahuman talk to it. It can handle different accents, slangwords, noise in the background, change in thehuman’s noise due to cold, etc.6. Handwriting recognition – The handwritingrecognition software reads that written on paper by apen or on screen by a stylus. It can recognize theshapes of letters and convert it into editable text.7. Intelligent robots – Robots are able to perform thetask given by a human. They have sensors to detectphysical data from real world such as light, heat,temperature, moment, sound, bump, and pressure.They have efficient processors, multiple sensors andhuge memory, to exhibit intelligence. In addition,they are capable of learning from their mistakes andthey can adapt to the new environment.VI. COMPUTER VISIONThis is a technology of AI with which the robots can see. Thecomputer vision plays vital role in the domains of safety,security, health, access and entertainment.Computer vision automatically extracts, analyzes, andcomprehends useful information from a single image or anarray of images. This process involves development ofalgorithms to accomplish automatic visual comprehension.Hardware of Computer Vision Systems:This involves-1. Power supply.2. Image acquisition device such as camera.3. A Processor.4. Software.5. A display devise for monitoring the system.6. Accessories such as camera stands, cables, andconnectors.Task of computer vision:1. OCR – In the domain of computers, optical characterreader, software to convert scanned documents intoeditable text, which accompanies a scanner.2. Face detection – Many state-of-the-art camerascome with this feature which enables to read the faceand take a picture of that perfect expression. It is usedto let a user access the software on correct match.3. Objects recognition – They are installed in supermarkets, cameras, high-end cars such as BMW, GM,and Volvo.4. Estimating position – It is estimating position of anobject with respect to camera as in position of tumorin human’s body.Application domains of Computer Vision:1. Agriculture.2. Autonomous vehicles.3. Biometrics.4. Character recognition.5. Forensics, security, surveillance.6. Industrial quality inspection.7. Face recognition.8. Gesture analysis.9. Geosciences.10. Medical imagery.11. Pollution monitoring.12. Process control.13. Remote sensing.14. Robotics.15. Transport.VII. MACHINE LEARNINGMachine learning is a branch of science that deals withprogramming the systems in such way that they automaticallylearn and improve with experience. Here, learning meansrecognizing and understanding the input data and making wisedecisions based on the supplied data.It is very difficult to cater to all the decisions based on allpossible inputs. To tackle this problem, algorithms aredeveloped. These algorithms build knowledge from specificdata and past experience with the principles of statistics,probability theory, logic, combinational optimization, search,reinforcement learning, and control theory.The developed algorithms form the basis of variousapplications such as:1. Vision processing.2. Language processing.3. Forecasting (e.g., stock market trends).4. Pattern recognition.5. Games.6. Data mining.7. Expert systems.8. Robotics.Machine learning is a vast area and it is quite beyond thescope of this chapter to cover all its features. There are severalways to implement machine learning techniques, however themost commonly used ones are Supervised, UnsupervisedLearning and Reinforcement Learning.Supervised Learning:Supervised learning deals with learning a function fromavailable training data. A supervised learning algorithmanalyzes the training data and produces an inferred function,which can be used for mapping new examples. Commonexamples of supervised learning include:a. Classifying e-mails as spam.b. Labeling WebPages based on their content, andc. Voice recognition.There are many supervised learning algorithm such as neuralnetworks, Support Vectors Machines (SVMs), and NaïveBayes classifier. Mahout implements Naïve Bayes classifier.Unsupervised Learning:Unsupervised learning makes sense of unlabeled data withouthaving any predefined dataset for its training.Unsupervised learning is extremely powerful tool foranalyzing available data and look for patterns and trends. It ismost commonly used for clustering similar input into logicalgroups. Common approaches to unsupervised learninginclude:a. K-means.b. Self-organizing maps, andc. Hierarchical clustering.Reinforcement Learning:This strategy built on observation (a decision by observing itsenvironment). If the observation is negative, the networkadjusts its weights to be able to make a different requireddecision the next time.VIII. STEP-BY-STEP PROCEDURE FOR DESGINING AROBOTStep 1: Defining the Problem.In this step involvesa. Identifying the purpose of construction.b. Identifying the specific requirements.You are confronted with a situation. Here are the twoexamples:A community wants to construct a robot zoo in which the’animals’ moves their heads open their mouth and makeappropriate sounds when they sense that someone is comingtowards them. Design and built a prototype device whichcould satisfy this need.A local pet shop wishes to sell a range of devices whichautomatically feed small cage pets (such as rabbits, gerbils,mice, etc.) when their owners are away for the weekends.Design and built a prototype device which could satisfy thisneed.You need to determine what problem you are trying to solvebefore you attempt to design and build a robot to solve aproblem. Take a time to study a number of different situationsand once you have decided what the situation is and youunderstand exactly what the problem is then write a designbrief in a log book (this will be your working document asyou work on your robot. This log book can be a paper the notebook or an electronic document.)This is a short statementwhich explains the problem that is to be solved.Step 2: Researching and Designing.In this step involvesa. Gathering information.b. Identifying specific details of the design which mustbe satisfied.c. Identifying possible and alternative design solution.d. Planning and designing a appropriate structure whichincludes drawings.Step 3: Creating a PrototypeIn this step involves -a. Testing the design.b. Troubleshooting the design.You should ideally think of at least three different ways tosolve the problem before you concentrate on any one inparticular. Sketches and notes are required at this stage. Youcan also create prototypes using lego for this step. Once youhave created a lego prototype, take a digital picture of it. Printout the picture and jot your notes below the picture in your logbook. Once you have settled on one situation, go back over thelist of specifications you have made. Make sure that eachspecification is satisfied.Now, it’s the time to produce some working drawings. Theseare the drawings that will assist you as you begin constructingthe prototype of your structure. (Here again, lego and digitalcamera might be your be best friend.) You may choose to doyour drawings by hand or you might want to use a drawprogram on the computer to assist you.Determine a working schedule for yourself. Draw up atimetable showing how much time you expect spend on eachpart of the design process. Your planning should also ensurethat you have all the necessary materials and equipment thatyou need to complete your project.Step 4: Building your Robot.Construction work can now begin with help of the art ofLEGO design by Fred Martin – an excellent resource forbuilding very strong structures.Step 5: Programming and Testing your Robot.Now it is time to program your robot. This can be achieved inmany different ways. Use can achieve rudimentaryintelligence in your robot by using only relay, potentiometers,bump switches and some discrete components. You canincrease complexity in intelligence in your robot by addingmore sensors and continuing in the same vein of usinghardwired logic. By introducing a more sophisticated controlelement, the microprocessor, you introduce a significant newtool in solving the robot control problem. For our robots weused the RCX Brick that was first developed by Fred Martin atMIT as the Programmable Brick.Step 6: Evaluating your Robot.In this step involves -a. Evaluate the design.b. Evaluate the planning process.As building and programming work progresses, and the designbegins to take shape, you will automatically carry out tests onthe design. You will also need to complete systems tests atvarious stages of the construction. If any of the tests show thatyou have failure in a joint, or that part of your structure is notmeeting specifications, then you will have to makemodifications in your plan.When building and programming is complete, the entireproject must be tested to see if it does the job for which it wasdesigned an evaluation needs to then be written.This should be a statement outlining the strengths andweakness in your design. It should describe where you havesucceeded and where you have failed to achieve the aim setout in the specifications.Here is a list of the questions which will help you to preparethis statement.a. How well does the design function?b. Does the design look good?c. Is the product safe to use?d. Did I plan my work adequately?e. Did I find the construction straight forward ordifficult?f. Were the most suitable material used?g. Did it cost more or less expected?h. How could I have improved my design?IX. CONCLUSIONRobotics is an interdisciplinary branch of engineering andscience that includes mechanical engineering, electricalengineering, computer science and others. Robotics deals withthe design, construction, operation, and use of robots, as wellas computer systems for their control sensory feedback, andinformation processing.Today we find most robots working for people in industries,factories, warehouses, and laboratories. Robots are useful inmany ways. For instance, it boosts economy because businessneeds to be efficient to keep up with the industry competition.Therefore, having robots helps business owners to becompetitive, because can do job better and faster than humanscan, e.g. robot can built, assemble a car. Yet robots cannotperform every job; today robot roles include assisting researchand industry. Finally, as technology improves, there will benew ways to use robots which will bring new hopes and newpotentials.REFERENCES1. John J.Craig , “Introduction to Robotics”, Pearson,2009.2. Nocks, Lisa (2007). The robot : the life story of atechnology. Westport, CT: Greenwood PublishingGroup.3. Computer Integrated design and Manufacturing, DavidD Bedworth, McGraw Hill International, 19914. Artificial Intelligence, 2nd ed., Rich, Tata McGraw Hill.5. Genetic Algorithms in Search and Optimization, andMachine Learning, D. E. Goldberg, AddisonWesley,1989.6. Jones, A.C. (2002) ‘Minds, matter and machines’, T209Resource CD-ROM, Cyborg, part 1 – Robotics.