Researchers at Cornell University, New York, are developing tiny, insect-inspired robots that don’t just look like the real thing. They think like it too.
Taking inspiration from nature and mimicking it to the extreme are two very different challenges. Engineers designing intricate robots often seek to replicate the way that animals move, from Boston Dynamics to EPFL’s electronic, pollution-detecting eel.
A more complicated hurdle is copying the way that animals and insects think and process information. Overcoming that could eventually reduce payloads, free space for more computation and make tiny, insect-like robots a lot more convincing.
Researchers at Cornell University are doing exactly that with RoboBees, 80-milligram robotic insects manufactured by the Harvard Microrobotics Lab. With a wingspan of just 3cm, they offer the ideal base unit for new programming that could help them react and adapt to the world like the creatures they were inspired by.
Neuromorphic computer chips offer ‘event-based’ processing
These new developments are enabled by neuromorphic computer chips, which process spikes of electrical current rather than binary code made up of 0s and 1s. These complex electrical combinations work in a similar fashion to how neurons fire inside a brain.
Silvia Ferrari, a professor of mechanical and aerospace engineering and director of Cornell’s Laboratory for Intelligent Systems and Controls, has suggested that neuromorphic computer chips could lessen the need for the dense computers that usually form a robot’s payload.
The Cornell lab is developing ‘event-based’ sensing and control algorithms that mimic neural activity in response to external stimuli. They are being tested with RoboBees.
“Getting hit by a wind gust or a swinging door would cause these small robots to lose control. We’re developing sensors and algorithms to allow RoboBee to avoid the crash, or if crashing, survive and still fly,” said Ferrari.
“You can’t really rely on prior modeling of the robot to do this, so we want to develop learning controllers that can adapt to any situation.”
Removing weight from the equation
As part of the project, the RoboBees have been outfitted with vision, optical flow and motion sensors. The ambition is to soon remove the need for a tethered power source thanks to the use of “event-based” sensing. Cornell’s algorithms could allow RoboBee and similar small robots to become more autonomous and adaptable without being weighed down by bulky power sources.
“We’re using RoboBee as a benchmark robot because it’s so challenging, but we think other robots that are already untethered would greatly benefit from this development because they have the same issues in terms of power,” said Ferrari.