A bio-mimetic miniature drone for real-time audio based short-range tracking

Area of Science:

• Bio-inspired robotics
• Sensorimotor behavior
• Acoustic signal processing

Background:

• Flying animals exhibit complex sensorimotor behaviors, including sound-based tracking of conspecifics, prey, or mates.
• Challenges include sound detection, binaural matching, real-time localization, and rapid pursuit of moving targets.

Purpose of the Study:

• To mimic the sound-based tracking ability of flying animals using a bio-mimetic autonomous drone.
• To develop and test a real-time sound localization system for autonomous aerial pursuit.

Main Methods:

• Equipped a commercial drone with an in-house 2D sound localization electronic circuit using two microphones.
• Implemented real-time sound signal processing to steer the drone in the horizontal plane.
• Conducted experiments to evaluate the drone's ability to track and pursue sound-emitting targets.

Main Results:

• The autonomous drone successfully tracked the position of a moving sound-emitting target in real-time.
• The system demonstrated the capability for real-time pursuit based on acoustic cues.
• The study highlights the practical challenges and solutions in replicating biological sensorimotor capabilities.

Conclusions:

• The bio-mimetic drone effectively replicates key aspects of sound-based aerial tracking.
• This technology has potential applications beyond bat echolocation, adaptable to various acoustic signals.
• Robotic implementation provides insights into the real-world difficulties faced by biological trackers.