Avian-inspired embodied perception in biohybrid flapping-wing robotics

Area of Science:

• Robotics
• Bio-inspired Engineering
• Materials Science

Background:

• Avian feathers offer intricate adaptability for maneuverable aircraft.
• Mimicking avian somatic sensation in robots faces weight and complexity challenges.

Purpose of the Study:

• To propose an avian-inspired embodied perception approach for biohybrid flapping-wing robots.
• To develop a low-weight, accurate perception system for robotic flight states.

Main Methods:

• Designed a feather-piezoelectric mechanoreceptor using vibration structures and piezoelectric materials.
• Employed convolutional neural networks and grey wolf optimizer for tactile perception.
• Integrated sensing of airflow velocity, flapping frequency, pitch angle, and wing morphology.

Main Results:

• Achieved accurate tactile perception of airflow velocity and wing flapping frequency.
• Successfully sensed pitch angle via airflow direction and wing morphology via feather collisions.
• Validated low-weight, accurate perception of flight states using motion capture.

Conclusions:

• Constructed a biomechanically integrated embodied perception system for flapping-wing robots.
• Demonstrated potential for reflex-based control of complex flight maneuvers.
• Highlighted promise for natural bird flight surveillance applications.