Mosquito-inspired scientists develop perception system to avoid obstacles

Inspired by the mosquito’s ability to fly and land in the dark, researchers recently developed a perception system that can evade collisions and tested them on quadcopters. An international team of scientists, led by Professor Richard Bomphrey of the Royal Veterinary College (RVC) in London, studied the sensory mechanisms of male reservoir mosquitoes and found the ability to simulate the insect’s ability to detect obstacles using air currents

Many campers often sleep outdoors without knowing that mosquitoes have landed on themselves, and one important reason is that mosquitoes can land lightly on you even in the dark at night. They do this thanks to mechanical sensing, a response to mechanical stimuli that allows them to sense obstacles without using the eyes. Unlike bats, which navigate through a biosonar system, mosquitoes use a combination of wings, tentacles and airflow to navigate.

Mosquito-inspired scientists develop perception system to avoid obstacles

According to the team, mosquitoes fly by flapping their elongated wings very quickly, producing fast jet air to provide lift. If these jet streams encounter obstacles, these air flow patterns change shape, which can be detected by an array of sensors called “Zhuang’s organs” at the base of the mosquito’s tentacles. This allows the insect to use “aerodynamic imaging” to create images of its surroundings, allowing it to map the location of the ground and other obstacles.

Mosquito-inspired scientists develop perception system to avoid obstacles

The team used the findings to provide it with a miniature four-rotor aircraft with aerodynamic imaging by installing a bio-inspired sensor device. The unit consists of an array of probe tubes connected to the pressure differential sensor, which are placed on maximum sensitivity. After a series of test flights, the quadcopter was allowed to fly autonomously.

Mosquito-inspired scientists develop perception system to avoid obstacles

The team found that the quadcopter could detect surfaces on the ground or walls at sufficient distances, with little need for data processing, and that the new system was said to be lightweight, low-power and scalable.