System Engineering Management

ROBOT PAINTING USING ARTIFICIAL INTELLIGENCE

SUBMITTED BY
Hari priya Kapileswarapu
1504606
ADVISOR
Steven Maher; Geoffrey Rodman
OKLAHOMA CHRISTIAN UNIVERSITY

ABTRACT:

This paper describes about one the functionality of the robot and how robot paints an object using artificial intelligence. And objects like iron wall, wooden box, motor bike etc. Suppose if the robot has to paint a motor bike it has to identify the parts of the motor bike and has to paint the color to those parts. For example, if the rims of the wheels had to be painted in black color then the robot has to identify the rims and paint the black color. In the same way, if the chassis had to be painted with black color then the robot has to identify the chassis and paint it accordingly. In this way, the whole motor bike would be painted by the robot based on the parts. For painting the parts of the motor bike there are some methods like generating the optimum path and the trajectory methods. In this way, robot can paint the respective object based on the models with different colors.
Index Terms— Artificial Intelligence, Dec

ACKNOWLEDGEMENT: I feel it a great pleasure and honor to express our immense gratitude towards our esteemed guide, Geoffrey Rodman for standing by our side all through the implementation of the project. His able technical guidance and expertise have Contribute to the success of the technical seminar. And this made better understandability of the project. I also express out deep felt thanks and a sincere gratitude to out Steven Maher, for his kind suggestions and encouragement during the preparation of this project work.

INTRODUCTION: Robots come under two classifications, those are autonomous or manual handled technical machines, which can be both either static or mobile. They were initially invented to help humans in hazardous environments and repetitive tasks, but now-a-days robots are also starting to replace human labour. Robots are controlled by computer programs as well as electronic circuitry. A robot is any machine or mechanical device, which operates automatically or semi-automatically. Hence we general consider a robot to be limited to something which is powered by electricity.

The particular advantage or usage of modern robots is in the workplace on product lines. They are used exclusively to make production more efficient and productive for the company, cheaper for the customers and safer for the employees. Robots are widely used in such industries as automobile manufacture to perform simple repetitive tasks, and in industries where work must be performed in environments hazardous to humans. Many aspects of robotics involve artificial intelligence. Robots may be equipped with the equivalent of human senses such as vision, touch, and the ability to sense temperature. Some are even capable of simple decision making, and current robotics research is geared toward devising robots with a degree of self-sufficiency that will permit mobility and decision-making in an unstructured environment. Today’s industrial robots do not resemble human beings; a robot in human form is called an Andriod.

The process of painting is based on the shape of the object and robot does not know what parts of the object it is painting. In several motor bikes making industries to reduce the work load on the workers or to prevent the hazardous paints the industries use their own spraying robots to paint the motor bike. The spray painting production is widely used in industry and becomes every day more competitive. Current research deals mainly with automation of the process, optimization and quality of the paint. In order to specify a path for a robotic manipulator a manual tool planning that is a teaching method is usually performed. An operator teaches the robot, painting path by moving the end effector around the parts to be painted and in the meantime, positions and orientations are recorded by the robot controller. In this way the robots paint the parts of the motor bike separately.

This paper illustrates about the robots which can paint the parts of the motor bike based on the instructions given by the instructor or user. If the instructor gives some instruction to paint the doors of the motor bike then the robot must find the doors by itself, by scanning the each and every part of the motor bike. The robot will be able to identify the parts of the motor bike based upon the information given to it which can be programmed in it using some programming languages. The painting of the parts can be done using some optimum path and trajectory generated systems. The decisions which the robot is taking is totally dependent on the information given to it. While painting the parts of the motor bike based upon the instructions it should contain all the information regarding the parts of the motor bike.

The information can be a 3D representation model of the motor bike. We can say that the robot is having some artificial intelligence. This is because when we give an instruction to a human to paint the front wheel of a motor bike he directly goes to the side of front wheel and paints the paint, in the same way this robot can also paint the parts of the motor bike based upon the instructions given by the instructor. There are several robots like Aaron which paints some drawings with the information whatever it is having. Now a day’s research is being carried out on robots to make them artificially intelligent, so that work load on human beings can be reduced and to save the human beings from working in the hazardous environment. For example, the spray painting used for painting the motor bikes.

II. STRUCTURAL ANALYSIS OF PAINTER ROBOT:

We are trying to build a robot arm that performs some actions based upon the instructions given by the instructor. To know more accurately about the spray robot, initially we have to understand the physical structure of this steel structure (robot). The main parts of the robot that we have to consider would be manipulator or rover (main body of the robot, link joints and other structural element of robot), end effector (the part that is connected to thelast joint hand of a manipulator), actuators (muscles of the manipulator like servo meter, stepper motor and hydraulic cylinder) , the sensors, controller( controls and co-ordinates the motion of actuators), processor (calculates the motions and the velocity of the robot joint) and software(operating system, robotic software and collection of routines) . For any robot the sensors would play a very vital role because these sensors are helpful in maintaining quality control in work paint inspection, data collection of objects in the work cell. The sensors which we can use for the robot are the position sensors which are very helpful in sensing and controlling the arm position of the robot. The velocity sensors are also very useful in providing the velocity feedback. The device which does this is the tachometer which is also called the rpm gauge and revolution counter. A tachometer is employed in a motor to calculate the rotational speed of a shaft. There are several other components which are to be added to the robotic arm to make it intelligent so that it can paint the parts of the motor bike by scanning those parts and deciding whether these parts are to be painted which are mentioned in the instruction given by the instructor. Initially a scanner as to be attached to the robotic arm so that it can scan all the parts of the motor bike which are to be painted. Second of all a medium as to be selected through which the instructor can give the instructions to the robot it can be a voice recognition system or a command recognition system. If the robot cannot understand the instructions given by the system then it should indicate through a message on the control system saying that the instruction was not received properly. The scanner would be attached to the front of the robotic arm so that it can scan the parts of the motor bike when a certain instruction is given. The motor bike or the object can be placed on a platform which can rotate so that the motor bike can reach the other parts of the motor bike. When a certain instruction is given by the instructor for example paint the gas tanker of the motor bike, then the robot scans the each part of the motor bike and finds the gas tanker to be painted, if the doors are on the other side of the robot then the robot asks the instructor to rotate the platform on which the gas tanker is placed so that it can paint the other side of the gas tanker. The scanner which we are using is the Leica Scan Station P510 which is very useful in scanning the whole parts of the gas tanker. This scanner rotates about the vertical axis of rotation which scans all the parts of the motor bike in a three dimension way. Regarding scanner, it’s provided with a camera with the help of which it can store the information regarding the scanned parts of the object. The medium through which the instructor interacts with the robot is with the help of a microphone. There are several methods which are mentioned in several papers like [3, 4] using which the interaction can be achieved. There are other sensors being used which can help to stop the collisions involving the spray gun of the robotic arm with the parts of the gas motor bike.

Coming to the image storage information, all information that is given to the robot is stored is some libraries so that it can identify the parts based on the information given in these libraries. The programming of the robot is done in such way that whenever the instructor gives an instruction to the robot then it tries to find that word in the library which is already stored in it. For example if the instructor is trying to give the instruction to paint back wheel of the gas tanker then it tries to look for these words in its library, now the word back wheel is the key word so the robot then looks for the part called back wheel. The robot finds out this part by scanning the parts of the motor bike and it compares with the image of the back wheel which is already stored in the library. Here the robot must make a decision about what parts are matching with the images which are already stored in its library so that it can spray the part which is instructed by the instructor. So for the decision making purposes there are some decision making algorithms which are to be included in the programming of the robot. If the robot cannot find the parts of the motor bike then there as to be an indication of sound alarm saying that the parts that are specified in the instruction cannot be found. This should also be mentioned in the programming part of the robot. For a robot to identify an object is a difficult task to achieve but when the hardware parts that are being used are arranged properly and the software for the robot is written properly this can be achieved. The problems which arise while executing this model would be because of the software but not because of the hardware. This is because any robot can perform its tasks efficiently only when the software for the robot is written properly.

III WORKING ENVIRONMENT OF THE ROBOT:

Coming to performance issue of the robot, it has to perform the tasks it requires a model for its task. This model includes the object models which the robot handles, the task model which the robot performs, the work space models where the robot is performing its tasks.
The object models may include geometric models for its operation, voice recognition and visual identification. The task model describes the knowledge of the task which the robot can perform. The workspace model includes the work range of the robot and the restriction of the movement in the area. The object model for this robot would be the objects like motor bikes and the rotating platform. The task model would be recognizing the parts of the motor bike, spraying the parts of the motor bike and if the parts are not found then telling the instructor with the help of sound alarm that the parts are not found. The working model for the robot would be the working space in which it is working. The robot after completing the recognition of the parts as to do some path planning to spray the parts which it as recognized. As mentioned in the starting there are lot of methods through which the optimum path planning and the trajectories can be generated. While doing the programming for the software, we have to mention these things in the software so that the robot can generate an optimum path planning which makes it to take a path that is more efficient. We are using lot of sensors to measure the environment so that the robot can work efficiently. This robot should pick the necessary information from huge volumes of information. To know about this the robot needs a well-designed perceptual system. For perceiving the environment there are some features that are being used like image understanding , voice recognition, knowing the shapes of the parts in the images etc. For any robot to be intelligent it as to be provided with the information regarding the environment in which it is working. The environment for this robot would the object whose parts as to be identified, spray gun and paint. The actuators are, a camera to record the information while scanning, the Leica scanner to scan the parts of the object, the spray gun for spraying the parts. The different sensors are used for maintaining the threshold distance between the spray gun and the parts of the object while spraying the parts. As mentioned in the starting there are other sensors which would be needed to maintain the velocity needed for the movement of the robotic arm.
In this way the perceptual system as some impact on the actions taken by the robot. Thus in this way the perceptual system for robot must be designed in such a way that we can get the optimum output. There is a cycle mentioned in called perceiving-acting cycle model with the help of which the robot design, perception and action sequences can be perfectly set for the robot. Thus by making the robot to work under the perception-acting cycle we can make the robot to take the actions properly.

The robot performs its tasks based upon the instructions given by the instructor. To understand some words that are being used in the instruction the instructor as to interact with the robot many times. The higher the number of times the instructor interacts with the robot then the chances of understanding the instruction by the robot will be high. We have seen that in several Chabot’s the higher the number of users interact with the Chabot the more it will understand the words texted or talked by the user. There are several methods using which the instructions can be given. In this paper we are trying to build a robot which is artificially intelligent. As we know that for a machine to be artificially intelligent there are lot of things which help the machine to be intelligent. These can be called as the agents. We can use the agent theory mentioned in the book [6]. In the book the author as explained how the agents work together for any robot to achieve the task of the robot.

For any robot to work properly there are things which are important, they are the software, environment should be properly defined. The robot performs the tasks properly only when the software is designed properly and in the software we mention what all the things have to be considered by the robot to take a decision for performing the tasks. The environment is also important it helps the robot in performing the tasks. The sensors are also the important parts of the robot because of the sensors only the robot will make a decision for preventing the collision between the spray gun and the motor bike the robot is provided with a sensor.

In the same way the scanner is also the important part with the help of which the robot is able to scan the parts of the motor bike. A camera is used to record the pictures of the motor bike. Based on these pictures the robot will take a decision to whether these are matching with the pictures of the corresponding parts that are already stored in its library. These are the things which are really important for making the robot artificially intelligent.

FLASHBACK ARTIFICIAL INTELLIGENCE: * The 1956 Dartmouth conference was the moment that AI gained its name, and is widley considered the birth of AI. * In 1950, Alan Turing published a landmark paper in which he speculated aabout the possibility of creatin machines with true intelligence. * The Turing Test was the first proposal in the philosophy of artificial intelligence.

TURING MACHINE:

A Turing machine is the original idealized model of a computer, and it’s manipulates symbols on a strip of tape according to a table of rules. Turing machines are equivalent to modern electronic computers at a certain theoretical level, but differ in many details. A Turing machine consists of a line of cells known as the "tape", together with a single active cell, known as the "head". The cells on the tape can have a certain set of possible colors, and the head can be in a certain set of possible states.
The Turing machine was invented in 1936 by Alan Turing, who called it an "a-machine" (automatic machine). The Turing machine is not intended as practical computing technology, but rather as a hypothetical device representing a computing machine. Turing machines help computer scientists understand the limits of mechanical computation.
An unlimited memory capacity obtained in the form of an infinite tape marked out into squares, on each of which a symbol could be printed. At any moment there is one symbol in the machine; it is called the scanned symbol. The machine can alter the scanned symbol and its behavior is in part determined by that symbol, but the symbols on the tape elsewhere do not affect the behavior of the machine. However, the tape can be moved back and forth through the machine, this being one of the elementary operations of the machine. Any symbol on the tape may therefore eventually have an innings. A generalization of the Turing machine in which the head is allowed to move steps in either direction was considered by Macura (2005).

A template for specifying a 3-state, 2-color Turing machine is shown above using a form of notation due to Wolfram (2002). In this figure, a state is represented by a square containing a pointer indicating any of three possible directions; the property of 'color' is represented by the color of the square; and an instruction is represented by two squares in a column, with the top one representing a possible color and state of the active cell and the bottom one giving the new state and color of the active cell together with the direction the tape should be moved. The special state 0 (with no pointer) indicates a state at which the Turing machine should halt, i.e., cease computation.
The number of -state, -color Turing machines (disallowing machines with halting states) is given by (Wolfram 2002, p. 888). A Turing machine consists of:

1. A tape divided into cells, one next to the other. Each cell contains a symbol from some finite alphabet. The alphabet contains a special blank symbol (here written as '0') and one or more other symbols. The tape is assumed to be arbitrarily extendable to the left and to the right, i.e., the Turing machine is always supplied with as much tape as it needs for its computation. Cells that have not been written before are assumed to be filled with the blank symbol. In some models the tape has a left end marked with a special symbol; the tape extends or is indefinitely extensible to the right. 2. A head that can read and write symbols on the tape and move the tape left and right one (and only one) cell at a time. In some models the head moves and the tape is stationary. 3. A state register that stores the state of the Turing machine, one of finitely many. Among these is the special start state with which the state register is initialized. These states, writes Turing, replace the "state of mind" a person performing computations would ordinarily be in. 4. A finite table (occasionally called an action table or transition function) of instructions (usually quintuples [5-tuples] : qiaj→qi1aj1dk, but sometimes quadruples [4-tuples]) that, given the state(qi) the machine is currently in and the symbol(aj) it is reading on the tape (symbol currently under the head) tells the machine to do the following in sequence (for the 5-tuple models): * Either erase or write a symbol (replacing aj with aj1), and then * Move the head (which is described by dk and can have values: 'L' for one step left or 'R' for one step right or 'N' for staying in the same place), and then * Assume the same or a new state as prescribed (go to state qi1).
In the 4-tuple models, erasing or writing a symbol (aj1) and moving the head left or right (dk) are specified as separate instructions. Specifically, the table tells the machine to (ia) erase or write a symbol or (ib) move the head left or right, and then (ii) assume the same or a new state as prescribed, but not both actions (ia) and (ib) in the same instruction. In some models, if there is no entry in the table for the current combination of symbol and state then the machine will halt; other models require all entries to be filled.

Initially all tape cells are marked with 0:

State table for 3 state, 2 symbol busy beaver | Tape symbol | Current state A | Current state B | Current state C | | Write symbol | Move tape | Next state | Write symbol | Move tape | Next state | Write symbol | Move tape | Next state | 0 | 1 | R | B | 1 | L | A | 1 | L | B | 1 | 1 | L | C | 1 | R | B | 1 | R | HALT |

For Example:

Example: state table for the 3-state 2-symbol busy beaver reduced to 5-tuples | Current state | Scanned symbol | | Print symbol | Move tape | Final (i.e. next) state | 5-tuples | A | 0 | | 1 | R | B | (A, 0, 1, R, B) | A | 1 | | 1 | L | C | (A, 1, 1, L, C) | B | 0 | | 1 | L | A | (B, 0, 1, L, A) | B | 1 | | 1 | R | B | (B, 1, 1, R, B) | C | 0 | | 1 | L | B | (C, 0, 1, L, B) | C | 1 | | 1 | N | H | (C, 1, 1, N, H) |

THE STATE IN TURING MACHINE: The word "state" used in context of Turing machines can be a source of confusion, as it can mean two things. Most commentators after Turing have used "state" to mean the name/designator of the current instruction to be performed—i.e. the contents of the state register.
But Turing (1936) made a strong distinction between a record of what he called the machine's "m-configuration", (its internal state) and the machine's (or person's) "state of progress" through the computation - the current state of the total system. What Turing called "the state formula" includes both the current instruction and all the symbols on the tape: Thus the state of progress of the computation at any stage is completely determined by the note of instructions and the symbols on the tape. That is, the state of the system may be described by a single expression (sequence of symbols) consisting of the symbols on the tape followed by Δ (which we suppose not to appear elsewhere) and then by the note of instructions. This expression is called the 'state formula'.

Turing machine "state" diagrams:

The table for the 3-state busy beaver ("P" = print/write a "1") | Tape symbol | Current state A | Current state B | Current state C | | Write symbol | Move tape | Next state | Write symbol | Move tape | Next state | Write symbol | Move tape | Next state | 0 | P | R | B | P | L | A | P | L | B | 1 | P | L | C | P | R | B | P | R | HALT |

The above table is expressed as a "state transition" diagram:

UNIVERSAL TURING MACHINES:

It is possible to invent a single machine which can be used to compute any computable sequence. If this machine U is supplied with the tape on the beginning of which is written the string of quintuples separated by semicolons of some computing machine M, then U will compute the same sequence as M.

This finding is now taken for granted, but at the time (1936) it was considered astonishing. The model of computation that Turing called his "universal machine"—"U" for short—is considered by some (cf Davis (2000)) to have been the fundamental theoretical breakthrough that led to the notion of the stored-program computer.

GOALS OF ARTIFICIAL INTELLIGENCE:

* To build system that exhibit intelligent behaviour. * To understand intelligence in order to model it. * To increase our understanding of perceptual, reasoning and learning. * To design and construction of useful new tools in science, industry, and culture.

BRANCHES OF ARTIFICIAL INTELLIGENCE:

* Perceptive System : A system that approximates the way a human sees, hears, and feels objects. * Vision System : Capture, store, and manipulate visual images and pictures.

* Expert System : Mechanical and computer devices that performs tedious task with high precision.

* Learning System : Computer changes how it function or reacts to situation based on feedback.

* Natural Language Processing :

Computer understand and react to statements and commands made in a "natural" language, such as English.
Neural Network:

Computer system that can act like or stimulate the functioning of the human.

CATEGORIES OF ARTIFICIAL INTELLIGENCE:

* Conventional AI : Conventional AI mostly involves methods now classified analysis.This is also known as symbolic AI, logical AI, neat AI and Good Old Fashioned Artificial Intelligence (GOFAI).

Methods Include:

* Experts systems: apply reasoning capabilities to reach a conclusions.An exert system can rocess large amounts of known information and rovide conclusions based on them. * Case based reasoning. * Bayesian networks. * Behaviour based AI: a modular method of building AI Systems by hand.

* Comutational Intelligence (CI): Comutational Intelligence involves learning.Learning is based on empirical data and is associated with non-symbolic AI, Scruffy AI and soft comuting.

Methods Includes:

* Neural networks: systems with very strong pattern recognition capabalities. * Fuzzy systems: techniques for reasoning under uncertainity, has been widely used in modern industrial and consumer product control systems.

* Evolutionary computation: applies biologically inspried concepts such as populations, mutation and survival of the fittest to generate increasingly better solutions to the problem. These methods notably divide into evolutionary algorithms (e.g.genetics algorithms) and swarm intelligence (e.g.ant algorithms).

TYES OF ROBOTS:

Robots can be separated into different types. Most of them share the same basics like components but their purpose can be very different. While some robots are built to improve the productivity, others are built to help us in our daily life or to entertain us.

* Mobile robots :
In contrast to industrial one arm robots, mobile robots have the ability to move around, they are not fixed to one location. An example for mobile robots are automated guided vehicles (AGV). They use markers or wires in the floor to navigate, some also use vision or lasers. Most of them are used in warehouses to move materials.
Mobile robots are also found in military and security. There are also vacuum cleaners designed as mobile robots.
The current research has a big focus on mobile robots, so every university has a lab focusing on mobile robot research. * Medical :

Intuitive Surgical, Inc. da Vinci Surgical System
The use of robots in health care can offer significant benefits to both patients and staff in a hospital.
Surgical robots have multiple arms where different tools are attached. The surgeon is able to ensure precise actions with these robotic arms. The movement control of these arms is controlled via a computer station. Using such a surgical robot offer significant advantages for the patient and the surgeon compared to previous techniques.
Some advantages include: * More accurate surgery * Lower risk of bleeding * Smaller scars * Risk of infections is reduced * Faster healing process due to lesser strain on the body

Subtypes of Medical robots * Surgical robots;
Their most important feature is their high precision. * Rehabilitation robots:
This group facilitates and supports the lives of infirm, elderly people, or those with dysfunction of body parts effecting movement. These robots are also used for rehabilitation and related procedures, such as training and therapy. * Bio robots:

A group of robots designed to imitate the cognition of humans and animals. * Telepresence robots:
Allow off-site medical professionals to move, look around, communicate, and participate from remote locations. * Elderly care / Support robots:
A support robot helps disabled or old people to move, hear and see. They come in a various forms. These robots can be used as a stairlift. It is also possible to use them as hearing aids, so that even deaf people can hear again.

ADVANTAGES: * Physical labor will and can be turned over to robots. Examples for this are: * Picking fruit * Driving a car * Fire fighting * Simple tasks in household (doing dishes, vacuuming, etc.) * Robots will take over work, which humans don’t like to do. This includes most dangerous tasks like working in a mine or disgusting tasks (e.g. cleaning the toilet). * Robots can discover places (e.g. deep sea, space exploration, deep earth) which are impossible to discover for humans under normal conditions. * Robots mostly work in the industry because they are more efficient, precise and accurate than humans, for example the car industry. Or they often help humans with their work in the industry. * Robot designing and maintenance create new jobs. * Less workless caused by illness. * They are able to work without exhaustion. * Robots could become a big part in the medical field. They could assist doctors in surgeries or maybe do routine surgeries. In that case humans would only be needed to observe the robot and its actions. * Robots keep the proprieties of the source code. * Jobs which are in extremely harsh conditions like underwater, in the space, where air is needed. Robotics can work without problems. * Robots don’t need a rest they can work 24 hours. * They could help the elderly so retirement homes would be useless. * Robots could take over the part of assisting, help the pensioners and do their household or maybe even manage their medications. * Robots need many different mechanical components. * Cameras, Cables, Processors etc. all need to be build and bought from somewhere so the market of those chips and devices would be influenced positively but this could be rated as a disadvantage too. * Cost-efficiency * High precision -> robots can work with nanotechnology

DISADVANTAGES: * Many people are likely to lose their job as robots can easily replace many jobs because on the long term they are cheaper than men. Some of them are: * Cashiers * Cab drivers * Waitresses * Factory workers * Jobs and productions of devices needed for the building of robots could not be done by humans because robots could take over in this parts of the industry, too. * Expensive to build & maintain * Limited to only the work/task they're build and programmed for. * If used wrong robots can become really dangerous for humans. * Humans will depend too much on their robots and will unlearn basic skills the robots are doing for them. * Robots are often developed and used by the military. * Wrong programmed factory robots can destroy a whole production which costs a lot of money for the company. * Robots need a constant supply of power. * Employees will need a training program to interact with robots which takes time and costs money. * They can’t respond to danger like humans can if not programmed so. * Robots are very expensive to buy. * The worker will require a training program to interact with the new robotic equipment. It takes time to learn these things and financial output. * Robots are very good at doing their jobs but they can’t handle unexpected situations as well as people.

CONCLUSION: In this paper we have discussed how a robot does identify the object using artificial intelligence. Generally the robots that work in the automobile industry work based upon automation and the software is written in such a way that they just have the idea how to paint that object but they don’t know what object they are painting. But in this paper we have discussed about a robot that identifies the object based upon the object instruction given by the instructor. Firstly the robot scans the object and loads the scanned images of the object and then it identifies that object by matching those images with the pictures of the corresponding parts stored in its library. It takes instructions from the instructor through a microphone. The perceptual system for the robot should also be defined properly so that the actions taken depending on this should be in such a way so that they can complete the task.

REFERANCES: http://www.appropedia.org/Robotics http://www.ijeat.org/attachments/File/v2i4/D1443042413.pdf http://www.slideshare.net/shivbindal/artificial-intelligence-submitted-by-shiv http://en.wikibooks.org/wiki/Robotics_and_the_World