Abstract
Non-speech sound processing is of great interest for indoor mobile robot localization. This paper presents a technique based on feature extraction from continuous wavelet transform (CWT) and a dynamic feed-forward neural network that will approximate the position of the robot in the spatial domain. The link between the function approximation stage (ANN) and the feature extraction stage (CWT) is performed by feature comparative analysis.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have