Feed-forward selection of cerebellar models for calibration of robot sound source localization

Baxendale, M. D.; Nibouche, M.; Secco, E. L.; Pipe, A. G.; Pearson, M. J.

Abstract

© 2019, Springer Nature Switzerland AG. We present a responsibility predictor, based on the adaptive filter model of the cerebellum, to provide feed-forward selection of cerebellar calibration models for robot Sound Source Localization (SSL), based on audio features extracted from the received audio stream. In previous work we described a system that selects the models based on sensory feedback, however, a drawback of that system is that it is only able to select a set of calibrators a-posteriori, after action (e.g. orienting a camera toward the sound source after a position estimate is made). The responsibility predictor improved the system performance compared to that without responsibility prediction. We show that a trained responsibility predictor is able to use contextual signals in the absence of ground truth to successfully select models with a performance approaching that of a system with full access to the ground truth through sensory feedback.

• Presentation Conference Type: Conference Paper (published)
• Conference Name: Living Machines: Conference on Biomimetic and Biohybrid Systems
• Start Date: Jul 9, 2019
• End Date: Jul 12, 2019
• Acceptance Date: Jul 6, 2019
• Online Publication Date: Jul 6, 2019
• Publication Date: Jul 6, 2019
• Deposit Date: Apr 21, 2020
• Publisher: Springer Verlag
• Pages: 3-14
• Series Title: Lecture Notes in Computer Science
• Series Number: 11556
• Series ISSN: 0302-9743
• ISBN: 9783030247409
• DOI: https://doi.org/10.1007/978-3-030-24741-6_1
• Public URL: https://uwe-repository.worktribe.com/output/5914610
• Additional Information: First Online: 6 July 2019; Conference Acronym: Living Machines; Conference Name: Conference on Biomimetic and Biohybrid Systems; Conference City: Nara; Conference Country: Japan; Conference Year: 2019; Conference Start Date: 9 July 2019; Conference End Date: 12 July 2019; Conference Number: 8; Conference ID: lm0; Conference URL: http://livingmachinesconference.eu/2019/