An Adaptive Neural Mechanism for Acoustic Motion Perception with Varying Sparsity.

Area of Science:

• Neuroscience
• Robotics
• Acoustics

Background:

• Auditory motion perception is less understood than visual motion perception.
• Biological systems excel at tracking moving stimuli, a skill crucial for survival.
• Previous work introduced a bio-inspired neural mechanism for unoccluded auditory motion perception.

Purpose of the Study:

• To investigate auditory motion perception for both unoccluded and occluded sound stimuli.
• To extend a bio-inspired neural learning mechanism to handle environmental occlusions.
• To compare the performance of the mechanism under different sound stimulus velocities.

Main Methods:

• Developed a bio-inspired neural learning mechanism modeling the peripheral auditory system.
• Utilized directional information from sound stimuli and specific motor behavior.
• Tested the mechanism with unoccluded and occluded 2.2 kHz tonal sound stimuli at varying velocities (0.5°, 1.0°, 1.5°/time step).
• Implemented and evaluated the mechanism on a wheeled mobile robot for practical validation.

Main Results:

• The neural mechanism successfully tracked unoccluded and occluded sound stimuli.
• Performance was evaluated across different stimulus velocities and occlusion scenarios.
• The mobile robot implementation demonstrated practical auditory tracking capabilities.

Conclusions:

• The extended bio-inspired neural mechanism shows promise for robust auditory motion perception, even with occlusions.
• This research contributes to understanding and replicating biological auditory tracking abilities in artificial systems.
• The findings have implications for robotics and autonomous systems requiring sound-based navigation and tracking.