For roboticists working in the field of biomimetics, copying a bat's complex flight patterns has been a difficult problem to solve. Or, as Caltech professor and Jet Propulsion Laboratory researcher Soon-Jo Chung put it during a press conference, "bat flight is the holy grail of aerial robotics." And according to a new research paper published by Chung and his JPL colleagues in the journal Science Robotics this week, that holy grail has officially been discovered.
When it comes to building flying robots, there's a lot to be learned from the beasts of the air. And researchers have -- we've seen robots inspired by the birds and the bees. The bat, however, is a different story, an animal whose flight seemed perhaps too complicated to unlock.
Unless, of course, you're a team of three roboticists from the University of Illinois at Urbana-Champaign and Caltech. They've just built a lightweight robot inspired by bats that can wing its way through the air. It's called Bat Bot, and the research has been published in the journal Science Robotics.
"Bat flight," said researcher Soon-Jo Chung of Caltech and the Jet Propulsion Laboratory during a press conference, "is the holy grail of aerial robotics."
This is because bat flight is complex. The animal uses over 40 active and passive joints on its wings to fly, as well as the flexible membrane that stretches over its phalanges.
Bat Bot doesn't have quite so many joints, but its advanced construction makes one of the most flexible autonomous flying robots to date. Other biomimetic robots are often clunky, or limited in their movement, but the 93-gram, 47-centimetre-wingspan Bat Bot is able to replicate bat-like flight patterns.
There were several ways the team achieved this. One of the major ones is that the robot can move its wings independently of one another, using asymmetrical motion to be able to make sharp turns and fly a twisted path.
They also pared the number of joints in the robot down to just nine, five active and four passive, compared to the over 40 of real bats, by identifying which joints were the most important for the wing stroke. This is because copying the bat's anatomy exactly results in a robot that is too heavy with joints to fly. The bat's shoulder, elbow, wrist bend and side-to-side tail movement were determined to be the most important joints for flight.
The skeleton was then made from carbon fibre bones and 3D-printed ball-and-socket joints, which was covered by a soft, lightweight, durable silicone membrane just 56 microns thick. This membrane allows Bat Bot to change the shape of the wing structure in flight without compromising its smooth, aerodynamic surface. It also generates lift as the wings flap, conserving the power supplied by micromotors on the robot's back.
This energy conservation gives it one advantage over fixed-wing flying robots, such as quadcopters. Another is that it is much more quiet, which would make it a much less disruptive tool for tasks such as environmental and wildlife surveillance, while its ability to perform tight manoeuvres would give it an advantage in close proximity to buildings, for example, for search and rescue purposes or monitoring construction sites.
There's a little work to do yet before Bat Bot is ready for field work. Its electronics are quite delicate and need to be made more durable so that if Bat Bot crashes, the electronics can be salvaged. At the moment, it can only fly for a short time, too, since battery technology is limited, and the robot needs to be as light as possible.
The team also wants to work on perfecting perching manoevres, another holy grail of aerial robotics. A flying robot that can perch is able to land to conserve energy, which decreases battery load.
No comments:
Post a Comment