A recurrent neural network for sound-source motion tracking and prediction

Murray, J. C.; Erwin, H.; Wermter, S.

Abstract

Recurrent neural networks (RNN) have been used in many applications for both pattern detection and prediction. This paper shows the use of RNN's as a speed classifier and predictor for a robotic sound source tracking system. The system requires extensive training to classify all possible speeds to enable dynamic tracking of the most prominent sound within the environment.

Citation

Murray, J. C., Erwin, H., & Wermter, S. (2005). A recurrent neural network for sound-source motion tracking and prediction. In Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005 (2232-2236). https://doi.org/10.1109/IJCNN.2005.1556248

• Conference Name: Proceedings of the International Joint Conference on Neural Networks
• Conference Location: Montreal, Que., Canada
• Start Date: Jul 31, 2004
• End Date: Aug 4, 2005
• Online Publication Date: Dec 27, 2005
• Publication Date: Dec 1, 2005
• Deposit Date: Oct 26, 2018
• Volume: 4
• Pages: 2232-2236
• Series ISSN: 2161-4393
• Book Title: Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005.
• ISBN: 0780390482
• DOI: https://doi.org/10.1109/IJCNN.2005.1556248
• Keywords: Recurrent neural networks; Tracking; Robot sensing systems; Signal to noise ratio; Azimuth; Microphones; Hybrid intelligent systems; Manufacturing; Human robot interaction; Navigation
• Public URL: https://hull-repository.worktribe.com/output/799639
• Publisher URL: https://ieeexplore.ieee.org/document/1556248