Book Image

Artificial Vision and Language Processing for Robotics

By : Álvaro Morena Alberola, Gonzalo Molina Gallego, Unai Garay Maestre
Book Image

Artificial Vision and Language Processing for Robotics

By: Álvaro Morena Alberola, Gonzalo Molina Gallego, Unai Garay Maestre

Overview of this book

Artificial Vision and Language Processing for Robotics begins by discussing the theory behind robots. You'll compare different methods used to work with robots and explore computer vision, its algorithms, and limits. You'll then learn how to control the robot with natural language processing commands. You'll study Word2Vec and GloVe embedding techniques, non-numeric data, recurrent neural network (RNNs), and their advanced models. You'll create a simple Word2Vec model with Keras, as well as build a convolutional neural network (CNN) and improve it with data augmentation and transfer learning. You'll study the ROS and build a conversational agent to manage your robot. You'll also integrate your agent with the ROS and convert an image to text and text to speech. You'll learn to build an object recognition system using a video. By the end of this book, you'll have the skills you need to build a functional application that can integrate with a ROS to extract useful information about your environment.
Table of Contents (12 chapters)
Artificial Vision and Language Processing for Robotics
Preface

Recurrent Neural Networks


In this section, we are going to review Recurrent Neural Networks (RNNs). This topic will first look at the theory of RNNs. It will review many architectures within this model and help you to work out which model to use to solve a certain problem, and it will also look at several types of RNN and their pros and cons. Also, we will look at how to create a simple RNN, train it, and make predictions.

Introduction to Recurrent Neural Networks (RNN)

Human behavior shows a variety of serially ordered action sequences. A human is capable of learning dynamic paths based on a set of previous actions or sequences. This means that people do not start learning from scratch; we have some previous knowledge, which helps us. For example you could not understand a word if you did not understand the previous word in a sentence!

Traditionally, neural networks cannot solve these types of problem because they cannot learn previous information. But what happens with problems that cannot...