The Foundation of Robotics and Artificial Intelligence

In this book, I invite you to go on a journey with me to discover how to add Artificial Intelligence (AI) to a mobile robot. The basic difference between what I will call an AI robot and a more regular robot is the ability of the robot and its software to make decisions and to learn and adapt to its environment based on data from its sensors. To be a bit more specific, we are leaving the world of pre-coded robot design behind. Instead of programming all of the robot’s behaviors in advance, the robot, or more correctly, the robot software, will learn from examples we provide, or from interacting with the outside world. The robot software will not control its behavior as much as the data that we use to train the AI system will.

The AI robot will use its learning process to make predictions about the environment and how to achieve goals, and then use those predictions to create behavior. We will be trying out several forms of AI on our journey, including supervised and unsupervised learning, reinforcement learning, neural networks, and genetic algorithms. We will create a digital robot assistant that can talk and understand commands (and tell jokes), and we will create an Artificial Personality (AP) for our robot. We will learn how to teach our robot to navigate without a map, grasp objects by trial and error, and see in three dimensions.

In this chapter, we will cover the following key topics:

• The basic principles of robotics and AI
• What is AI and autonomy (and what is it not)?
• Are recent developments in AI anything new?
• What is a robot?
• Introducing our sample problem
• When do you need AI for your robot?
• Introducing the robot and our development environment