Giving autonomous machines 'spidey senses'

Researchers are building animal-inspired sensors into the shells of autonomous machines, such as cars, planes and drones, to help them process data faster.

Animals such as spiders, bats and birds have nerve endings linked to special neurons called mechanoreceptors which respond to stimuli. The nerve endings, or mechanosensors, only detect and process information essential to the animal’s survival, which means data is processed very quickly. For example, when a spider’s web vibrates at a frequency associated with prey, the mechanosensors on its legs detect it, prompting a quick reaction. However, mechanosensors wouldn’t detect a lower frequency, such as dust.

Purdue University researchers want to give machines these ‘spidey senses’ to help them process sensory information faster and better detect and avoid objects.

“There is already an explosion of data that intelligent systems can collect — and this rate is increasing faster than what conventional computing would be able to process,” said Andres Arrieta, Assistant Professor of Mechanical Engineering at Purdue University. “Nature doesn’t have to collect every piece of data; it filters out what it needs.”

The researchers developed and integrated artificial mechanosensors into the shells of autonomous machines which can sense, filter and compute quickly without needing a power supply.

They are made of stiff material that can change shape when activated by an external force. Changing shape makes conductive particles within the material move closer to each other, allowing electricity to flow through the sensor and carry a signal that informs how the autonomous system should respond. This could help drones navigate dangerous environments and self-driving cars avoid hazards.

“With the help of machine learning algorithms, we could train these sensors to function autonomously with minimum energy consumption,” Arrieta said. “There are also no barriers to manufacturing these sensors to be in a variety of sizes.”

The research is published in ACS Nano.

Image credit: ©stock.adobe.com/au/autonomous06