Robots seem to come out of a powerful futuristic technology, but in reality they are constantly evolving. For this reason, engineers often turn to insect observation for inspiration. Studying the response to stimuli from flies or locusts helps to teach robots to do the same.
Flying cage for flies
Scientists have designed a smart, flying cage to follow flies when they avoid a collision.
This « cable lab » can predict the fly’s movement and follow its flight through a room. Researchers say it is accurate to within 1 cm. « Understanding how the miniaturised brains of insects control sensory processing and flight behaviour could serve as a source of inspiration for future developments in robotics, such as micro air vehicles that mimic gliding on the scale of insects, » explain the authors of the study.
Backpacks for beetles
The scientists behind this study wanted to find a way to reduce the powerful vision capacity to the size of a beetle.
To do this, they designed tiny Go Pro backpacks and placed them on the backs of beetles they released in a university car park. The camera, which transmits video images to a smartphone at a rate of 1 to 5 frames per second, is based on a mechanical arm that can rotate 60 degrees. This allows the viewer to capture a high-resolution panoramic image or to follow a moving object while expending a minimal amount of energy.
Insects and Virtual Reality
These privileged insects have had a virtual reality environment created just for them. The aim of the researchers was to imitate what their free flight would be like and to observe their reaction and their approach to different objects in the environment, such as a tree in bloom.
By placing a fly, as well as other flying insects such as mosquitoes, in a controlled virtual environment, researchers are able to better understand how they feel and react to different sensory stimuli such as wind or odour.
By understanding how these insects use what is called parallax to see their environment, researchers hope to design search and rescue robots that can do the same.
Locusts and collision avoidance
In this study, scientists focused on understanding a particular neuron in the locust brain to see how they could mimic this behaviour for safer autonomous vehicles.
Because the neuron has two branches, the locust calculates the changes in these two inputs and realises that something is going to collide, » explained one author of the study. The researchers plan to expand the stimulus environment of their cars to include objects of different speeds, light intensities and trajectories. They hope that these new experiments will refine their device and improve its usefulness. This information could be a means of ensuring the safety of cars, passengers and pedestrians in traffic.