Virtual environment using cellular automata for experimental studies of sound propagation and echo classification with artificial neural networks

This work presents a Java platform capable to emulate sound propagation in a controlled 2D environment (obstacles and sound sources selected by the user) based on a cellular automata model. The platform is expandable and so far includes a feature preprocessor for the echo waves and a neural classifier. The proposed virtual environment allows performing various virtual experiments with relevance in robotics and Simultaneous Localization and Mapping (SLAM). Performing such virtual experiments allows important reduction in development cost and the fine tuning of a solution for a given problem before testing it on a real ultrasound workbench.