Abstract
An audio-based control modelling to perform robotic tasks
In robotics, the use of aural perception has received recently a growing interest but still remains marginal in comparison to vision. Yet audio sensing is a valid alternative or complement to vision in robotics, for instance in homing tasks. Most existing works are based on the relative localization of a defined system with respect to a sound source. Robotics is generally considered as a field of application of the localization process, consequently the control scheme of the robot is performed separately to the localization system.
By contrast, our approach explore a different line of work by considering the hearing sense as a direct and real-time input of closed loop control scheme for a robotic task. Built upon sensor-based control framework, we propose a modelling of sound features suitable for robotic tasks without requiring any explicit sound source localization. This modelling of auditory cues is based on the measurement of the time difference of arrival (TDOA) between two microphones. The robotic task is implicitly performed with respect to angular measurements: it can thus be related to beaconing or bearing-only homing techniques.