The Power of a Thinking Machine
An Introduction to Artificial Intelligence and Robotics
Exactly 3 years ago, I was sitting across my lecturer in his office discussing my interest for my final year project. Even with limited exposure and experience, I had always been fascinated by ‘Automation’. Automation is the use of computer-controlled devices to assume control of processes; it is the technique of making a system or process operate automatically. The term, ‘automation’, was first used in the automobile industry to describe the increased use of devices and controls that could operate automatically in the production lines. This greatly increased the efficiency and reliability of systems during manufacturing. As a Mechanical Engineering student of Obafemi Awolowo University interested in robotics, mechatronics and computers, I chose my final year project to be the ‘Design and Development of a Moving and Sorting Robot’, and set to learning all I could about automation and robotics.
My project work was greatly inspired by how the world around us has changed over time. Many things that are possible today were almost unimaginable some years ago. Our lives are now greatly characterized by the use of digital devices like smartphones, smartwatches and laptop computers which aid us with interaction in a virtual and digital world. Things like communicating with our friends and families on social media, placing orders and paying for products and services online, using maps to find our way around a new city, and general access to the plethora of information on the internet, are all available at our convenience through these digital devices. Some people are yet to realize this, but automation and artificial intelligence are already playing a huge part in our lives. Just like most technologies are built to simulate a naturally occurring system or organism (e.g a car was built to simulate an animal, a submarine; whale, a camera; an eye, an airplane; bird, etc.), artificial intelligence is the simulation of the human brain and intelligence processes by machines. These processes include;
- Acquisition and usage of information (learning)
- Reaching approximate or definite conclusions through predefined rules (reasoning)
- Ability to change a decision based on the results being achieved (self — correction).
Like the human baby, an AI system does not know anything when it is first created. Babies use their senses to perceive the world and their bodies to interact with it and then learn from the consequences of their actions. A few years ago, my little brother would regularly put sand from the floor into his mouth; after being caught in the act and stopped a couple of times, he soon realized that the act was inappropriate, or that some things tasted better than sand. In the same manner, an Artificial Intelligence system is given a lot of labeled and unlabelled data to process and determine certain patterns. It then uses these patterns to make certain decisions like recommending a video on Netflix or YouTube based on your interests and watch history or determining whether to unlock your phone when your face or fingerprint is recognized. Other familiar use cases include self — driving cars, Google Maps, Alexa, and Robotics — the area I chose to specialize in for my final year project.
My final year project involved the design, modelling, and building of a robotic arm that could recognize the colour of a product and make a decision on where that product should go based on the colour. The project basically uses robots to automate a process that would normally be handled by a human in a manufacturing plant, and this is where a lot of people worry that Artificial Intelligence and Automation might make a lot of people jobless. This is only as true as the last industrial revolution, where the jobs of a lot of people were taken by those who knew how to use computers. Humans have always evolved and we must keep evolving to see Artificial Intelligence & Automation as Augmented Intelligence; that will make us more efficient and effective in doing our jobs. Augmented Intelligence is an alternative conceptualization of AI that focuses on AI’s assistive role, emphasizing the fact that cognitive technology is designed to enhance intelligence rather than replace it. A job is more likely to be replaced by AI & Automation if it is made up of menial & repetitive tasks rather than a role with responsibilities and if it does not require taking initiative, empathy, relationships or creativity.
Behind this thinking machine, is the speed and processing power of modern computers available to us today. The real revolution in modern days is not actually Artificial Intelligence, this has been available to us for over 5 decades now. The real revolution is that we now have computers that can take in a lot more data and process these data in less than a second. The speed of a computer is linked to the number of transistors it has to do operations. Every 2 years since 1965, engineers have doubled the number of transistors that can fit into the same amount of space, so computers have gotten much more faster (this is called Moore’s Law). For context, see the table below that shows the number of processes that can be handled by different devices at different points in time;
This goes to show the power of the devices we have at our fingertips. Our future will be powered by these powerful computers with the ability to make certain decisions, learn from the decisions made and correct themselves. This is the power of the thinking machine. We will be able to use this power to solve a lot of problems today and make life a lot easier for ourselves. Self — Driving Cars, Computer Vision, Chatbots, Image Processing, Speech, Object & Pattern Recognition, Robotics & Automation, Virtual Assistants and AutoCorrect are several ways Artificial Intelligence has been and will continue to disrupt the way we live our lives affecting finance, communication, education, manufacturing, healthcare, law, security, business and almost every other industry in the world today.
Examples of AI Technology include;
- Automation: Involves making a system or process function automatically, performing repetitive, high — volume tasks. An example is my final — year project.
- Machine Learning: This involves getting a computer to learn and act without programming the computer. An example is playing a game against a computer program that has played the game a thousand times and learned from the actions.
- Machine Vision: This involves capturing and analyzing visual information using a camera, analog-to-digital conversion, and digital signal processing. An example is image recognition and self — driving cars.
- Natural Language Processing (NLP): The processing of human language by a computer program. An example is text translation, speech recognition or spam detection.
- Robotics: Involves the design and manufacturing of robots that can perform tasks that are difficult for humans to perform or perform consistently.
To get started with AI, you might want to start off with learning programming languages like Python, Java, C, Lisp or R. If you’re new to programming, I will suggest starting with Python. Python is considered the best choice for AI development languages due to simplicity. Python syntaxes can be easily learned because of its readability. Python supports object-oriented, functional as well as procedure-oriented styles of programming. There are plenty of helpful libraries and frameworks in Python, which makes it really easy to get started with complex algorithms and solutions on Python; examples are Matplotlib, Numpy and a host of others. Some free resources to get started are available on www.codecademy.com, www.datacamp.com, and www.udemy.com.
First Published on Unboxed; the edu360 magazine — for young people.