Robotics,+Artificial+Intelligence+and+Expert+Systems

= 3.11 Robotics, artificial intelligence and expert systems = = =

Introduction
The increasing capability of IT systems has allowed developers to implement systems that attempt to understand and imitate human behaviour. These systems have already had profound effects on society, although their effectiveness is largely determined by the accuracy of the algorithms that underpin them. The increasing use of robotics, artificial intelligence (AI) and expert systems raises a range of ethical issues. For example, at which point should humans hand over key decision-making to a computer? Should robots have the same rights as humans? What social impacts might arise with the replacement of human workers or the creation of smart weapons?

Possible scenario
A hospital administrator is considering using a computer-controlled robotic device to assist with knee surgery. The robotic system models the patient’s knee area prior to surgery. During the procedure the robot is controlled by the surgeon using a joystick. The system eliminates the effects of tremors in the surgeon’s hands and limits the range of movement of the cutter to areas of the knee that have been predetermined by the model. Although there are claims that the surgery is less invasive and patients have a shorter recovery time, questions may be raised about the reliability and the cost of the system.

**Social and/or Ethical impacts:** · Reliability o Any defects in the system may be harmful to the patient(s). · Digital divide o The cost of the device will be costly. Therefore not all hospitals may be able to afford it. It’s most probable that the more developed countries will have them before others. · People and machines o People and machines are definitely impacts because the doctors are actually using the robots. They are controlled using joysticks.

Robotics
**Input Devices** **Camera** Many robots with complex programming use cameras along with high-speed computers and software to determine the locations and shapes of objects. This is called “machine vision.” In manufacturing, robots are able to identify pieces. Machine vision can be used in quality checks, sorting, and material handling. Many MV (Motion vision) software packages use various digital image processing techniques, allowing the robot to recognize what it is looking at.

**Sensors** Sensors are devices that receive analogue signals and convert those signals into digital data that can be read or displayed. Many different sensors are used with robots. A camera is an example of a sensor. It can be used as a motion sensor (infrared motion detectors). Other sensors that are possibly used: · Temperature sensor · Pressure sensor · Touch sensor · Heat sensor - it measures temperature. · Proximity sensor - it detects nearby objects without phisical contact, emitting eletromagnetic radiation. · Magnetism sensor · Light sensor - a device that senses the absence of light. · Humidity sensor - a device consisting of a special plastic material whose electrical characteristics change according to humidity levels. · pH sensor - a device that measures how acidic or basic a substance is.

**Microphones** The use of microphones allows robots to collect, process and record sounds. A microphone acts as the robot’s ears, allowing it to “hear” what goes on in the surrounding environment.

**Claws ** Robots are sometimes equipped with claws. Robots with claws are programmed for carrying objects. These robots can be used in factories where heavy parts need to be carried and relocated to a different station. They are also used in junkyards to lift large unwanted objects.
 * Output Devices **

media type="youtube" key="BKnXWMPbCyc" height="349" width="425" align="center"

**Wheels ** Many robots are programmed to travel certain distances under different circumstances. These robots use wheels to get around.



**Motors ** Motors, also known as engines, are machines that **convert other forms of energy into mechanical energy**. This results in motion. There are different types of motors. Examples of motors are: · Pneumatic motors · Hydraulic engines · Heat engines · Steam engines · Electric motors **Relays ** **Relays are electrically operated switches. Many relays are made up of electromagnets and contacts. Current flows through the coil of the relay, which creates a magnetic field. the field then attracts a lever and changes the switch contact. The point of the relay is to use a small amount of power in the electromagnet. **

**Speakers ** <span style="font-family: Calibri,sans-serif; font-size: 11pt; margin-left: 49.65pt;">Speakers are devices that convert electrical signals into sounds that can be heard <span style="font-family: Calibri,sans-serif; font-size: 11pt; margin-left: 14.2pt;">A robot is an automatically guided machine that can perform tasks on its own, or with specific guidance. Robots are used in many occasions such as: · <span style="font-family: Calibri,sans-serif; font-size: 11pt;">Car production · <span style="font-family: Calibri,sans-serif; font-size: 11pt;">Packaging <span style="font-family: Calibri,sans-serif; font-size: 11pt; margin-left: 50.2pt; text-indent: -18pt;"> Electronics · <span style="font-family: Calibri,sans-serif; font-size: 11pt;">Automated guided vehicles (AGVs) · <span style="font-family: Calibri,sans-serif; font-size: 11pt;">Telerobots · <span style="font-family: Calibri,sans-serif; font-size: 11pt;">At home
 * <span style="font-family: Calibri,sans-serif; font-size: 11pt;">Robot **


 * <span style="font-family: Calibri,sans-serif; font-size: 11pt;">Android **

<span style="font-family: Calibri,sans-serif; font-size: 11pt; margin-left: 14.2pt;">An android is a robot that has been designed to have human characteristics. They usually have realistic skin and sensors. The sensors enable the androids to react to touch. The androids are able to make movements that humans do unconsciously, such as breathing and blinking. <span style="font-family: Calibri,sans-serif; font-size: 11pt; margin-left: 14.2pt;">There are many projects involving the creation and development of androids. An example is HRP-4C. This is a female humanoid robot that was created by a Japanese research group called the National Institute of Advanced Industrial Science and Technology. HRP-4C uses Vocaloid to be able to sing. media type="youtube" key="xcZJqiUrbnI" height="314" width="504" align="center"

<span style="font-family: Calibri,sans-serif; font-size: 11pt;">media type="custom" key="9801201"[|Download this podcast]
 * <span style="font-family: Calibri,sans-serif; font-size: 11pt;">Cyborg **

<span style="font-family: Calibri,sans-serif; font-size: 11pt;">

<span style="font-family: Calibri,sans-serif; font-size: 11pt; margin-left: 14.2pt;">A cyborg is a being with both biological and artificial parts.

<span style="font-family: Calibri,sans-serif; font-size: 11pt; margin-left: 14.2pt;">In **art**, the idea of cyborgs is often associated with science fiction. Many artists attempt to raise the awareness of cybernetics. A man named Neil Harbisson claims to be a cyborg. He has a condition that only lets him see in black and white. Because of this, Harbisson started on the eyeborg project. The eyeborg works with a camera that is mounted on his head. It picks up colours directly in front of him and converts them in real-time into sound waves.

In **medicine** there are two types of cyborgs: the **restoratives** and the **enhanced**. Restorative technology works to restore lost functions of organs and limbs. They don't enhance their abilities. The enhanced cyborg intends to exceed normal processes or even gain new functions that were not originally present.

In the **military**, organizations utilise **cyborg animals** for the purpose of a supposed tactial advantage. [|Military Cyborg Menagerie] [|16 Geniune Cyborg Technologies]

Artificial intelligence
**Artificial Intelligence vs. Computational Intelligence** Artificial intelligence (AI) is the science and engineering of creating intelligent machines. AI researchers developed algorithms that mimic the reasoning that humans use to solve problems. In contrast to AI, Computational Intelligence uses numerical, knowledge-representation and knowledge-processing. Some techniques of CI are **Fuzzy Logic**, **Artificial Neural Networks** and **Evolutionary Algorithms (EA)**. Fuzzy logic deals with reasoning that is approximate, like a partial truth. Neural Networks are based on the construction of animal brains. They are a form of a multiprocessor computer system. These networks attempt to recreate the thinking process of the brain. Evolutionary algorithms is a term used to describe problem solving systems that uses processes inspired by biological evolution.

**Man or Machine** **Turing Test** This is a test of a machine’s ability to show intelligent behaviour. A conversation is held between a human judge, a machine, and a human. If the judge cannot distinguish between machine and the human, the machine is said to have passed the test.

**CAPTCHA** A CAPTCHA is a challenge-response test. It attempts to ensure that the response is not produced by a machine. It involves a computer asking to complete a simple test. Unlike a turing test, CAPTCHAs are administered by a machine and aimed at a human.



**Capabilities and Limitations** **Emotion Recognition** Research about emotional recognition has been conducted. Aiming to detect the emotional state of a speaker from different speech samples, scientists perform **facial expression analysis** or **speech analysis**. Research in the field of emotion recognition shows, that emotional reactions are strongly related to **pitch** and **energy** of the spoken message. The pitch of happiness or anger is always higher than that of sadness or fear. Intuitive understanding was another component of Intelligence that has been neglected. Unlike logical reasoning, understanding is subconscious, intuition-based and virtually instant recognition of objects or concepts. The ability to understand something immediately, without the need for conscious reasoning, current AI systems aren't able to understand these things. Artificial Intuition (AN) is a new, but unproven, method of approaching large problems that require artificial intelligence. AN is used to recognize sensory data and understand concepts at low levels. [|A New Possible Path To Artificial Intelligence by Monica Anderson]
 * Intuition**

**AI Techniques** **Searching** Search is a universal problem-solving mechanism in Artificial Intelligence. Problems addressed by AI search algorithms are divided into three classes: · Single-agent path-finding problems The task is to find a sequence of operations that map an initial state to a goal state. (e.g. Rubik’s Cube, vehicle navigation). · Two-player games Chess, Othello and checkers. · Constraint satisfaction problems Planning and scheduling applications are some examples.

**Pattern Recognition** Pattern recognition aims to organize data based on either prior knowledge or on information extracted from the patterns.



**Heuristics** This refers to experience-based techniques for solving a problem, discovery and learning for oneself. These methods are used to speed up the process of finding a good enough solution. A couple examples are an intuitive guess or common sense.

<span style="font-family: arial,sans-serif; font-size: small; line-height: normal;">The ability of a computer to evolve and improve its performance based on previous results from sensory data or databases.
 * Machine Learning**

Cleverbot, created by Rollo Carpenter, is an AI application that learns to mimic human conversations by communicating with humans. However, when I asked Cleverbot, "Who is Rollo Carpenter?" the clever robot did not know. Hmmm...

Fuzzy logic deals with reasoning that is fluid or approximate rather than fixed and exact. FL can be implemented in hardware, software or a combination of both. FL provides a method of arriving to a conclusion based on vague or imprecise information.
 * Fuzzy Logic (FL)**

**Can machines become independent?** **<span style="font-family: arial,sans-serif;"> Are you secretly planning on taking over the world and killing all the humans?? **

**Natural Language Communication** Progress on building computer systems that process natural language in any meaningful sense. A goal of AI work in natural language is to enable communication between people and computers without memorizing complex commands and procedures. media type="youtube" key="M3tcSlWLS_g" height="313" width="381" align="center"


 * Watson, IBM's Super Intelligent Robot**

It took IBM four years to create Watson. It is said to have the ability to compete against a human's ability to answer questions that are spoken in natural language.

Watson, whose size actually takes up a whole room, was represented using a monitor as the output device. Watson competed on stage with former //Jeopardy!// champions Ken Jennings and Brad Rutter.

media type="youtube" key="PHhDLUVAtqU" height="459" width="767" align="center"

**Neural Networks: Similarity to Biological Systems** Traditionally used to refer to a network of biological neurons, The term neural network is used to refer to artificial neural networks. Processing is structured in terms of an interconnected group of artificial neurons. **Pattern Recognition** It is the electronic <span style="font-family: arial,sans-serif; font-size: small; line-height: normal;">translation of images of handwritten, typewritten or printed text (usually captured by a scanner) into machine-editable text. This is the extraction of meaningful information from images. Facial recognition is an example.
 * Biological neural networks consist of real biological neurons that are connected to the central nervous system.
 * Artificial neural networks consist of interconnecting artificial neurons (they mimic the properties of biological neurons). These networks may be used to study biological neural networks, or used for solving AI problems without creating a model of a real biological system.
 * OCR (Optical Character Recognition)**
 * Image Analysis **

This is a machine or software that is capable of recognizing spoken language. The spoken language is taken and translated into text. The user can also instruct the machine to do certain tasks by speaking his/her commands. This is the artificial production of human speech. An example of speech synthesizing software is Vocaloid. it is a singing synthesizer developed by the Pompeu Fabra University in Spain and Yamaha Corpration. ** Expert Systems ** An expert system is a computer program that acts as a human expert. **Collection, Creation and Maintenance of Knowledge Base** The expert system must first get its "knowledge." Human experts contribute their information on a particular subject, which is then programmed into the system. The data should be kept up to date accordingly. This will maintain the reliability of the expert system **Creation of the Inference Engine** The inference engine consists of a collection of executable rules. They are used to support reasoning. **Inference Rule** This is a function which takes place and gives a conclusion. An example is the rule of inference modus ponens. it takes to premises. One in the form 'if x then y" and another in the form x and returns the conclusion y. **Chaining** Two ways of reasoning with the inference rules are backward chaining and forward chaining. Forward chaining starts with the data available and uses the inference rules to extract more data until a desired goal is reached. Backward chaining starts with a list of goals and works backwards to see if there is data which will allow it to conclude any of these goals.
 * Speech Recognition **
 * Speech Synthesizers **

**Applications of Robotics, Artificial Intelligence and Expert Systems** iPhone applications allow users to control the Quadricopter Drone, developed by Parrot
 * Embedded Systems**
 * Cell Phones**

media type="youtube" key="4lHOHYhp6b4" height="458" width="767" Using the AI searching technique. People are able to find their way across town and reach their destinations.
 * GPS**

Hitachi washing machines use fuzzy controllers to load-weight, fabric-mix, and dirt sensors and automatically set the wash cycle for the best use of power, water, and detergent. This washing machine uses fuzzy logic.
 * Washing Machines**

An application of AI is programming a computer to play chess. Chess presents challenges that aren't in other games, because of the complexity and the number of possible moves. A common strategy for chess programs is to develop a tree data structure with all the possible moves from a given starting point up to a certain depth. The tree is then evaluated to find a good move.
 * Chess**

This is a key enabler of **human-machine interface**. Ford Motor Company has developed an advanced voice technology system that allows users to communicate with thier cars.
 * Voice Recognition**


 * Predictive Text**
 * <span style="color: #333333; font-family: 'Lucida Grande',Arial,Helvetica,sans-serif; font-size: 12px; font-weight: normal; line-height: 18px;">Swiftey uses AI to predict user's text. It’s the first keyboard of its kind. Not only does Swiftkey analyze the way words are put together, but it can scan the user's text message history to “learn” how he writes in order to custom-tailor his experience. **



Without visiting a real doctor, users are able to input various symptoms into the expert system and receive a response, This can be a diagnosis and/or suggested medicine to be prescribed by a doctor. The same can be done with fault diagnosis. For example in mechanics, a person would not have to take his car to a garage to find a fault in his vehicle. He can answer questions on an expert system that will eventually inform him what the problem is and suggest a solution for it.
 * Expert Systems **
 * Medical Diagnosis **
 * Fault Diagnosis **

Today many robots are use in factories and assembly lines. With the use of these robots, production has **increased** and has become more **efficient**. Robots perform the major assembling of parts while the humans sometimes inspect the minor details. Robots are used during operations. This is a positive impact for surgeons as they are able to control the robots and see from a larger screen. The work done can be more accurate when robots are used. However there could be a psychological issue where it's a machine doing the job and not a human. Patients may not want machines to make incisions in their bodies.
 * Robots **
 * In the Industry**
 * Health**

In many cases, these places are dangerous for humans to visit. Using robots for exploring the ocean and space protects the lives of humans. They can operate the robots from the control rooms on land.
 * Space and Underwater Exploration**