The objective of this book is to deliver exactly what is on the cover – Artificial Intelligence for Robotics. The emphasis is on machine learning techniques applied to ground mobile robots. The book starts with professional robot design principles that have been scaled down for smaller robot projects. The AI section begins with convolutional neural networks for object recognition and continues with reinforcement learning and genetic algorithms. The robot gets a voice and learns to tell jokes using AI-based voice recognition that can discern user intent. The book introduces a novel way to navigate without a map using a literal divide and conquer program that uses the upper part of the room to remember paths, and the lower part of avoid obstacles. The book demonstrates how path planning, decision trees, object classification, and navigation are all part of the same problem set. We finish by giving the robot an artificial personality. The final chapter concludes with thoughts on the future of robots and gives advice on robotics as a career.
The entire book is built around a single fun example task, which is to design and build a robot that can pick up toys in an indoor, unstructured environment. As you will learn, this project is anything but easy.
