Scientists at Purdue University in Indiana have reportedly equipped drones with artificial mechanosensors to improve their navigation capabilities. According to the team, the sensors act like real nerve endings that are linked to mechanoreceptors or the special neurons responsible for processing the information critical to the survival of animals.
According to Purdue University assistant professor Andres Arrieta, the biological mechanoreceptors play a significant role in filtering the data that animals need in their everyday living. For instance, the mechanosensors located on the legs of spiders give them the capability to react instantly to threats or if there are potential mates around.
Arrieta said:
“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. Nature doesn’t have to collect every piece of data; it filters out what it needs.”
Giving Autonomous Machines Artificial Mechanosensors
The biological mechanosensors of spiders tend to ignore lower frequencies like dust and other harmless particles since they don’t pose any threat to the animal. This unique capability inspired Arrieta and the team to create artificial mechanosensors that could be programmed to identify predetermined forces like objects that obstruct the way of drones and other autonomous machines.
These sensors could increase a machine’s capability to react instinctively against threats and avoid potential collisions. The team wants to integrate the artificial mechanoreceptors to the shells of drones and other autonomous machines.
Aside from sensing danger, the sensors developed by the Purdue researchers can also compute without the need for a power supply. The materials used in the sensors were designed to change shape rapidly when triggered by an external force. They utilize electricity to send the signals that could guide an intelligent machine to its proper course of action.
Arrieta added:
“With the help of machine learning algorithms, we could train these sensors to function autonomously with minimum energy consumption. There are also no barriers to manufacturing these sensors to be in a variety of sizes.”
Comments (0)
Most Recent