Development of voice controlled drive interface for Car Like Mobile Platform

Abstract

Intelligent Transportation Systems (ITS) involve projects that aim to integrate modern communication and information technology into existing transportation management systems. In this project, one approach to ITS for a prototype elementary vehicle is presented. One of the most important research topics in this field is the design and development of locomotion methods and intelligent communication systems for ITS. Voice is one of the convenient methods for interaction between humans and vehicles. Therefore, we explore the design and development of a prototype Car-Like Mobile Platform (CLMP) that mimics a real car. The design is rear wheels drive because it creates stable motion for calculation. The drive system for this mobile platform was developed using bipolar stepper motors for accurate lateral and horizontal movements. For the voice control, we developed an application that is able to learn and recognize voice commands. Freely available libraries on voice recognition using Hidden Markov Models (HMM) were integrated into the application. From the tested voice commands, 8 words were chosen to represent all possible types of motion for the CLMP. Communication between the interface and the drive system was achieved through serial communication. In the testing phase, it was found that the CLMP responded well to all given voice commands and conformed to standard non-holonomic behavior. The idea behind this application is to create an alternative form of interaction between car and user. Such an interaction can be very useful in the framework of an ITS designed for the handicapped.