Vision and Voice Recognition for 6 DOF Industrial Robot Model Based on a Multi-layered Artificial Intelligence Network

Abstract

Speech and vision recognition solutions that support intelligent control of a 6 DOF robotic arm using a multi-layered artificial intelligence network presented in this chapter. The research objective is to design a robot control system concerning three deep learning models of voice recognition and vision recognition to provide input signals to solve the inverse kinematics problem of the 6 DOF robotic arm. The first deep learning model is used to process voice commands to obtain information containing the robot's behavior and properties of a requested object. The second deep learning model performs image processing to recognize the requested object among the observed objects. The third deep learning model is designed to train and evaluate the accuracy of object recognition. Performance of the vision and speech recognition solutions is tested on a 6 DOF industrial robot model.