Classification of plant species from CTFM ultrasonic range data using a neural network

This paper describes the application of an artificial neural network (ANN) to the recognition of plants using continuous tone frequency modulation (CTFM) ultrasonic range data. Ultrasonic mobility aids have been used successfully by unsighted people for some time. One of these systems uses CTFM ultrasonics and allows perception of the environment by transmitting at ultrasonic frequencies and converting the echo into audible tones. After sufficient practice, unsighted users can navigate independently and confidently because they can recognise objects by the tonal patterns that they produce. This audio signal can also be captured and analysed by a computer. This paper discusses research that is being done in the Intelligent Robotics Laboratory at the University of Wollongong that aims to isolate the information in the signal that facilitates recognition.