There are plenty of drones designed for flying indoors. You know what those drones can’t do, however? Open doors. At least, that’s true in 99 percent of cases. Fortunately, the diminutive Flycrotug microdrone is the exception to the rule — and all it needs is a partner drone and its trusty lasso.
In a proof-of-concept demonstration, Flycrotug’s creators showed how two drones could attach their onboard cables to a door handle, anchor themselves to the floor using adhesive feet, and then pull the door open using on-board winches. This is only one example of how the drones could interact with their environment. Other examples might include removing a piece of debris blocking their path. The objects the drones are able to pull can weigh up to 40 times their own weight.
“We came up with a design for micro air vehicles that allows them to anchor onto their environment — similar to insects — and tug loads much heavier than what they can fly with,” Matthew Estrada, a graduate student at Stanford University, who worked on the project, told Digital Trends. “This addresses one persistent limitation with small, air vehicles: They are very mobile, but can only exert forces about twice their weight with thrust from their propellers. It is exciting progress because it could allow them to accomplish tasks that we usually send in larger robots to tackle.”
Perhaps unsurprisingly, Estrada said that an application for these tiny, scenery-manipulating drones might be in search and rescue tasks. They could be especially useful when these involve confined or remote environments, where other robots would struggle to fit or reach. That might mean slipping through a crevice or the window of a partially collapsed building to clear a pathway or retrieve an item of interest for rescuers.
“This is really a beginning, and there are many interesting questions in making this a practical method in unknown environments, where they would encounter a variety of different attachment surfaces and obstacles they would need to interact with,” he continued. “Planning maneuvers for manipulation tasks and generalizing the hardware to attach onto ever-more varied surfaces are two interesting thrusts for future work.”
A paper describing this research was recently published in the journal Science Robotics.
