Download - Robotics and expert systems
ROBOTICS AND EXPERT SYSTEMS.
WHAT IS ROBOTIC? Is the field of computer science and
engineering conscience with creating robot
is a branch of AI, which is composed of Electrical Engineering, Mechanical Engineering, and Computer Science for designing, construction, and application of robots.
PARTS OF ROBOTS. Sensors Control system manipulator . Power suppler. Software.
CHARACTERISTIC OF ROBOTS. Movement : move around its
environment by roller, wheels or legs. Energy: to power itself solar , battery or
electricity. Intelligences: smartness and is done by
programmer. Sensors: to senses its surrounding.
WHAT IS EXPERT SYSTEMS? Is a computer application that
performance task that would otherwise be performed by human expert
PARTS OF EXPERT SYSTEM. User interface. Knowledge based. Inference engine.
HOW EXPERT SYSTEM WORKS. USER INTERFACE; Is the system that allows a none expert
user to quarry all question to the expert system and to receive advice.
HOW EXPERT SYSTEM WORKS CONT: KNOWLEDGE BASED. It is a collection of facts and rules. It is created from the information
provide by human expert.
HOW EXPERT SYSTEM WORKS CONT: INFERENCE ENGINE. It act as search engine which examine the
knowledge based for information that match the user quarry.
None expert user quarry the expert system by asking question or answering question asked by expert system
The inference engine uses the quarry to search the knowledge based and then provides answer or advice to the user.
EXPERT SYSTEM
use
r in
terfa
ce Inference engine
Know
ledg
e ba
sed
Knowledge from expert
None expert gives quarryquarr
y
Advice
Expert system
COMPONENT OF KNOWLEDGE BASED It is a store for both :- factual knowledge based. heuristic knowledge based rule based knowledge based
FACTUAL KNOWLEDGE BASED Is the information widely acquainted by
the knowledge engineer and scholars in the task domain.
HEURISTIC KNOWLEDGE BASED. Is about practice accurate judgment
once a ability of evaluation and gauzing.
KNOWLEDGE REPRESENTATION . Is the method used to organized and
formulizing knowledge in the knowledge based it is in the form of IF-THEN-S RULES
KNOWLEDGE ACQUISITION . The success of any expert system
mainly depend in the quality, completeness and accuracy of the information stored in the knowledge based.
The knowledge based is formed by reading from different expert, scholar and knowledge engineers.
WHO IS KNOWLEDGE ENGINEER? Is the person with the quality of empathy ,
quick learning and cause analyzing skills. He acquires information from subject expert
by recording, interviewing and observation. He then categories and organize information
in a meaningful way in the form of IF-THEN –S RULES to be used by inference engine.
He also monitor the development of expert system.
INFERENCE ENGINE . It acquires and manipulate knowledge
from knowledge based to arrived to a particular solution.
IN CASE OF RULE BASED EXPERT SYSTEM. It applies rules repeatedly to the facts
which are obtain from earlier rule application.
It adds new knowledge to the knowledge based if required.
It resolves rule conflict when multiples rules are applicable to a particular case.
STRATEGIES USED BY INFERENCE ENGINE TO RECOMMEND SOLUTION ARE?
Foreword chaining. Back word chaining.
FOREWORD CHAINING It is a strategies of expert system to
answer the question what can happen next.
The inference engine follows the chain of conditions and directions and finally deduced/come up with the out come.
It consider all the fact and rules and sort them before concluding to a solution as shown on next slide.
FOREWORD CHAINING.
fact1
fact2
fact3
fact4
and
or
Decision 1
Decision 2
Decision 3
and
BACK WORD CHAINING. With this strategies expert system finds
out the answer to the question why this happen.
On the basic of what has already happened the inference engine tries to find out which condition could have happened in the past for the result.
This strategies is followed finding out cause or reason. As shown no next slide.
BACK WORD CHAINING
fact2
fact1
fact3
fact4
and
or
decision1
decision2
and
decision3
USER INTERFACE. It provides the interaction between the
user of the expert system and the expert system itself.
It is generally natural natural language processing so as to be used by the user who is well vast in the task domain.
It explain how the expert system has arrived to a particular outcome.
USER INTERFACE CONT: The explanations may appear in the
following formsa) Natural language displayed on screen.b) Verbal narration in natural language.c) Listing rule number displayed on the
screen.
REQUIREMENT FOR EFFICIENT EXPERT SYSTEM USER INTERFACE.
It should help user to accomplish their goals in shortest possible way.
It should be design to work for user exciting or desire work practiced.
Technology should be adoptable to user requirement, not the other way a round.
It should make efficient use of user input.
LIMITATION OF EXPERT SYSTEM. Are difficult to maintain. Difficult in knowledge acquisition. High development cost. Limitation of technology Require significant development time
and computer resources.
BENEFITS OF EXPERT SYSTEMS Availability :- they are easily available due to mass
production. Less production cost:- cost is reasonable and affordable. Speed:- offer great speed hence reduce amount of work. Less error rate:- error rate is low as compaired to human
error. Reduce risk:- can work in dangers environment to
human. Steady response:- work steadily without getting
emotional, tenses and fairtiged .
APPLICATION OF EXPERT SYSTEM. Medical domain:-are used in diagnostic
system to deduced cost of disease from observation data.
Mortaring system :- it is used for comparing data continues with observed system or with prescribe behavior e.g. mortaring leakage along petroleum pipeline.
Process control system
EXPERT SYSTEM TECHNOLOGY Expert system development
environment Tools Shell.
EXPERT SYSTEM DEVELOPMENT ENVIRONMENT
Includes:- hard wares and tools they are working stations
High level symbolic programming language such as LISP program and PROLOG.
Large data bases.
TOOLS. Includes:-powerful editors and multiple
windows. They provides rapid prototyping. They have end bit definition of model
knowledge representation and inference.
SHELLS Is an expert system without knowledge based. It provide the developer with knowledge acquiring,
inference engine, user interface and explanation facilities
Example of shells are:- JAVA expert system shell(JESS) which provide a fully developed java API(application programming interface) for creating an expert system.
Vidwan this is a shell developed is developed at national centre for software technology in Mumbai in 1993 it enable knowledge encoding in the form of IF THEN- S RULES
STEPS IN THE DEVELOPMENT OF EXPERT SYSTEM.
Identify the problem domain:- the problem must be suitable for an expert system to solve it. fine the expert in task domain for the expert system project. Establish cost effectiveness of the system.
Design the systems:- identify the expert system technology. Know and establish the degree of integration with other system and data bases. Realize how the concept can represent the domain knowledge best.
STEPS IN THE DEVELOPMENT OF EXPERT SYSTEM CONT. Develop the prototype :- the knowledge engineer uses
sample cause to test the prototype for any defenses in the performance. End user also test the prototype of the expert system.
Develop and complete expert system:-test and ensure the interaction of the expert system with all elements of its environment including the end user data bases and other information system. Document the expert system well. Train the user to use the expert system.
Maintained the expert system:-keep the knowledge based up to date by regular review and up dates. Carter for new interface with other information system as those system evolves .
ASPECTS OF ROBOTICS. The robots has mechanical construction
form or shape design to accomplish a particular task.
They have electrical components which power and control the machinery.
They contained some level of computer program that determine what when and how a robot does somethings.
DIFFERENT BETWEEN ROBOTS AND ARTIFICIAL INTELLIGENT . ARTIFICIAL INTELLIGENT ROBOTS
They usual operates in computer simulated world.
They operate in real physical world.
The input to an AI program is in symbols and rules
Input to robot is analogs signal in the form of speech waves form or images.
They need general purpose computers to operate on
They need special hardware with sensor and effectors .
ROBOTS LOCOMOTION. Locomotion is the mechanism that
make the robot capable of moving in its environment.
They are various types of locomotion which include:-legged
wheeled combined legged and
wheeled
LEGGED LOCOMOTION. These type of locomotion consumes more
power while demonstrating walking It requires more number of motors to a
accomplish a movement. It is suited for rough as well as smooth
surface makes it consumes more power for a wheel locomotion.
It is little difficult to implement due to stability issues.
LEGGED CONT: The total number of possible gaits a
robot can travel depends upon the number of its leg.
If a robot has k legs then the number of possible events is
N=(2K-1)! K=number of leg! =factious.
CALCULATION OF EVENTS In case of a two legged robot (k-2) the
number of possible events is lifting left leg. N=(2K-1)! Release left leg. =(2*2-1)! Lifting right leg. =(4-1)! Release right leg =3! Lifting both legs
togeth. =3*2*1 release both legs. =6 ans.
WHEELED LOCOMOTION Requires fewer number of motors to a
accomplish a movement It is little easy to implement as there
are less stability issues in case of more number of wheels.
It is power efficient as to legged locomotion.
WHEELED LOCOMOTION CAN BE IMPLEMENTED IN THE FOLLOWING FORM
Standard wheel It rotate around the wheel axis and around the
contact. Caster wheelIt rotate around the wheel axis and the off set
staring joint. Swidish 45 degree and 90 degree wheelThey are owni wheel and rotate around the
contact point around the wheel axis and around the roles.
WHEELED LOCOMOTION CAN BE IMPLEMENTED IN THE FOLLOWING FORM CONT:
Boll or spiral wheel.The are owni directional wheel and are
technical difficult to impliment
TRACKED SLIP/SKID LOCOMOTION In this type of locomotion the vechcal
use tracks as in a trunk. The robot is stirred by moving the trunk
with different speed in same or opposite direction
It offer stability due to large contract area and the ground.
COMPONENTS OF A ROBOT Robots are constructed with the following:-a) Power supply the robots are powered by batteries, solar power,
hydraulic or pneumatic power sourcesb) Electric motors(AC/DC) they are required for rotational
movement.c) Actuators they converts energy into movement.d) Pneumatic air muscles they contract almost 40%when air is
sacked in them.e) Muscle wires they contract by 5% when electric current is passed
through them.f) Sensors they provide knowledge of real time information on the
task environment . Robots are equipped with vision sensors and a tactile sensor which imitates the mechanical properties of touch of human fingertips
COMPUTER VISION. Is the technology with which the robots can
see. The computer vision plays a vital role in the domains of safety, security, health, access and entertainment.
A computer vision automatically extracts, analysis and comprehends useful information from a single image or an array of images.
This process involves development of algorithms to accomplish automatic vision comprehension.
THE HARDWARE OF COMPUTER VISION SYSTEM.
This involves:-i. Image acquisition device eg cameraii. A processoriii. A softwareiv. A display device for monitoring the systemv. Accessories such as camera stands,
cables and connectors.
USES/TASKS OF COMPUTER VISION Face detection:-many state of the art cameras come with
this feature which enables the computer to read the face and take the picture of that perfect expression. it is used to let a user access the software on a correct match
Object recognition:-are installed in supermarkets, cameras and high-end cars such as BMW, GM and VOLVO.
Estimating position:-it is used in estimating the position of an object with respect to camera i.e the position of tumor in human’s body.
Optical character reader:-is a software that converts scanned documents into editable texts which accompanies scanner
ARTIFICIAL NEURAL NETWORKS. This is a computing system made up of
a number of simple highly interconnected processing elements which process information by their dynamic state exchange to external inputs.
STRUCTURE OF ARTIFICIAL NEURAL NETWORKS (ANN)
The idea of artificial neural networks is based on the belief that, the working of the human brain by making the right connections can be imitated using silicon and wires as living neurons and dendrites .
The human brain is composed of 100 billion nerve cells called neurons.
They are connected to other 1000 cells by Axons.
STRUCTURE OF ARTIFICIAL NEURAL NETWORKS CONT;
Stimuli from the external environment or inputs from sensory organs are accepted by dendrites. This inputs create electric impulses which quickly travel through the neural network.
A neuron can then sent message to other neuron to handle the issue.
Artificial neuron network are composed of multiple neurons which imitates biological neurons of human brain.
The neurons are connected by links and they interact with each other. The nodes can take input data and perform simple operations on the data.
The results of this operations is passed to other neurons the output at each node is called its activation or node value.
Each link is associated with weight and A.N.N are capable of learning which takes place by altering weight values.
THE FOLLOWING ILLUSTRATION SHOWS A SIMPLE ARTIFICIAL.
Input hidden output
TYPES OF ARTIFICIAL NEURAL NETWORK They are two types of artificial neural
network topologiesi. Feedforword artificial neural network.ii. Feedback artificial neural network.
FEED FORWARD ARTIFICIAL NEURAL NETWORK.
The information flow is uni-directional. A unit sends information to other unit from which it does not receive any information.
There are no feedback loops. They are used in pattern
generation/recognition/classification. Have fixed inputs and outputs.
FEED FORWORD ARTIFICIAL NEURAL NETWORK
FEEDBACK ARTIFICIAL NEURAL NETWORK Here feedback loops are allowed. They are used in
content addressable memories
The diagrams shown above each arrow represents a connection between two
neurons and indicates the pathway for the flow of information
CONT: Each connection has a weight i.e an
integer number that controls the signal between the neurons
If the network generates a good or desired output, then there is no need to adjust the weight however if the network generates a poor on a desired output or error, then the system alters the weights in order to improve subsequent results.
MACHINE LEARNING IN ARTIFICIAL NEURAL NETWORKS.
Artificial neuron network are capable of learning and they to be trained.
TYPES OF LEARNING.1. Supervised learning.2. Unsupervised learning.3. Reinforcement learning
SUPERVISED LEARNING It involves a teacher that is a scholar than the
artificial neuron network itself.eg teacher feeds some example data about which the teacher already knows the answer.
This type of learning is used in partner recognition and artificial neuron network comes up with guess while recognizing then the teacher provide the artificial neuron network with answer, artificial neuron network then compare its guess with the teacher correct answer and make adjustment according to error.
UNSUPERVISED LEARNING. It is required when there is no example
data set with known answer.eg searching a hidden partner
In this type of learning clustering is applied ie dividing a set of element into group according some unknown partner based on the existing data set present.
REINFORCEMENT LEARNING This type of learning is built on
observation. The artificial neuron network make a
decision by observing it environment. If the observation is negative the
artificial neuron network adjust its weight to make required decision.
APPLICATION OF ARTIFICIAL NEURON NETWORKS.
Military :- they are used for weapons Electronics :-they are used in cording sequence
prediction Financial :-loan a devisor . Industrial:- used in manufacturing process
control. Transportation:- for routing system. Signal processing ;can be trained to process an
audio signal and filter a propriety . Time service prediction .