Robotics

➔ Robotics is a branch of computer science that incorporates the design, construction and
operation of robots
➔ Examples: factory equipment, drones, and domestic robots

Characteristics of a robot
↳ electronic components, such as actuators, sensors, and microprocessors
↳ mechanical structure or framework
↳ programmable

➔ There are two types of physical robots (software robots are not physical, working bots):
↳ Independent robots:
- have no direct human control (are autonomous)
- can replace human activity totally (no human intervention required)
 ↳ Dependent robot
- has human control
- can temporarily replace human activity

- Robots in Industries
➔ Have sensors and end-effectors
➔ They're programmed to do a specific task
➔ They either have a built-in microprocessor or are connected to a computer system

Advantages Disadvantages

Are capable of working in conditions that may Can find it difficult to do ‘non-standard’ tasks
be hazardous to humans

Are less expensive in the long run (since there Can lead to higher rates of unemployment
will be fewer salaries to pay)

Can work 24/7 Risk of deskilling

Are more productive than humans Factories can now be moved to anywhere in the
world where operation costs are lower (leading,
again, to unemployment in some countries)

Are more consistent and almost accurate Are expensive to buy and set up in the first
place

Are best used for repetitive tasks and hence Needs high maintenance
make less mistakes

The cost of heating and lighting will decrease —

- Aeroplanes

Advantages Disadvantages

Reduced running costs Vulnerable to cyber attacks

Improves safety since human error is removed High capital costs and operational costs

Improvement in passenger comfort Passengers may be reluctant to use this new

technology

Improved aerodynamics (no cockpit required) Security risks are high

— Glitches in the software or hardware can be

disastrous

(Most of the advantages and disadvantages are the same, with the exception of a few)

- Robots in Agriculture
➔ In agriculture, robots are mainly used for:
↳ harvesting of crops/picking vegetables and fruit
↳ phenotyping (the process of observing physical characteristics of a plant in order to assess its
health and growth)
↳ seed-planting and fertiliser distribution
↳ grass mowers/cutters
↳ weeding, pruning and harvesting

➔ All of these devices use sensors and cameras to go around obstacles; they can even be
programmed to ‘go to sleep’ if the weather turns bad

➔ Use of sensors, microprocessors and actuators helps the process to be done effectively and
accurately. Drones, cameras and end-effectors are used, too.
Artificial Intelligence
➔ Artificial intelligence (AI) is a branch of computer science that deals with the simulation of
intelligent behaviours by a computer

➔ Characteristics of AI:
↳ Collecting data
↳ Stores rules for using the data
↳ The ability to reason
↳ The ability to learn
↳ The ability to adapt
↳ The ability to change its own rules
↳ The ability to change its own data

➔ The three categories of AI:

↳ narrow AI: better at doing one specific task compared to a human
↳ general AI: similar at doing one specific task compared to a human
↳ strong AI: much better at doing many different tasks compared to human
➔ The AI system is capable of learning and adapting to its surroundings
➔ Can make predictions on the collected new data
➔ Examples of AI:
↳ smart home devices such as Alexa and Siri
↳ chatbots
↳ autonomous cars
↳ facial features recognition
➔ There are two types of AI (only these are to be studied):
↳ Expert System
↳ Machine Learning

- Expert System
➔ A type of AI that has been developed to mimic human knowledge and experiences
➔ Has a knowledge base, and inference engine, and interface, and a rules base
➔ Uses knowledge and inference to solve problems or answer questions that would normally
require a human expert

➔ Applications of expert systems:

↳ oil and mineral prospecting
↳ diagnosis of a patient’s illness
↳ tax and financial calculations
↳ strategy games, such as chess
↳ logistics
↳ identification of plants, animals and chemical/biological compounds

Advantages Disadvantages

Can have many areas of expertise Need considerable amount of training

High accuracy Set up and maintenance costs are very high

Results are consistent Can give very cold responses, that may not be
suitable in certain medical situations

Have the ability to store vast amounts of ideas They are only as good as the information/facts
and facts entered into the system

Make traceable logical solutions and —

diagnostics

Have very fast response times —

Provide unbiased reporting and analysis of the —

facts
- User interface
Note: Diagram to be added.

➔ Interacts with the user

➔ Interaction can be done through dialogue boxes, command prompts or other input methods
➔ Yes/No questions can be asked, and are based on previous answers

- Inference Engine
➔ The main processing element
➔ Acts like a search engine examining the knowledge base for information or data that matches
with the user's answers
➔ Is the problem-solving part as it makes use of inference rules present in the rules base

- Knowledge base
➔ Repository of facts
➔ Stores all the available authentic knowledge and information about an area/areas of expertise
➔ Collection of objects and their attributes

- Rules base
➔ Contains the inference rules
➔ rules are used by the inference engine to draw conclusions
➔ Follows logical thinking, which often includes "IF" statements

- Setting up an expert system

➔ Information about a particular area of expertise is collected and checked by experts; this can be
from websites on the internet, books, research papers, etc.
➔ All this information makes up the knowledge base
➔ A rules base is made by writing all the inference rules that may be needed
➔ The inference engine is made and tested extensively, ensuring that it works as needed and there
are no errors
➔ The user interface is made by choosing an appropriate method of interaction between the user
and the system, and then incorporating the hardware required for it into it
➔ When all the components are completed, the expert system is tested
Machine Learning
➔ It is a type of AI which has the ability to automatically learn and adapt to its own processes and/or
data
➔ Possible for the system to make predictions or even take decisions based on previous scenarios
➔ Offers fast and accurate outcomes due to its very powerful processing capability
➔ Has the ability to manage and analyse considerable volumes of complex data
➔ An example:

A robot needs to find its way through different puzzles. Each puzzle has a series of paths that the
robot needs to follow to find its way to the end of the puzzle. The puzzle contains dead ends and
obstacles, so the robot needs to decide which way to go. The robot’s program will use artificial
intelligence (AI).

The robot will use machine learning. It will adapt to deal with data provided to it. Data would be
about its location so that it does not follow the same route that may lead to a dead end. Data
would be about common and repeated obstacles and would allow the robot to take appropriate
reaction(s). It would store successful actions as well, as it would allow it to understand what’s
more likely to work at a certain condition.

Differences between Expert Systems and Machine Learning

Expert systems Machine learning

Represents a simulated intelligence in The practice of getting the machines to make

machines decisions about new situations, processes
etc. without being programmed to do so

They are machines that are capable of The aim is to make machines that learn from
thinking exactly like humans data acquisition, and solve new problems
from past experience