Wheeled robots are nothing new. Flying bots are old-hat, too. But this is new: a dumbbell-shaped hybrid bot that can roll on thin, end-mounted wheels … and fly with two sets of counter-rotating chopper blades.
Switching between the two modes involves a Rube Goldberg-esque series of maneuvers and contraptions. It ain't elegant, but it works. And as a result, the foot-long automaton - designed by a team led by Nikolaos Papanikolopoulos at the University of Minnesota - boasts the freedom of movement of a flying drone with the long endurance and stability of a ground bot.
The hybrid bot's creators unveiled their experimental machine at the recent International Conference on Robotics and Automation in Shanghai.
The transforming robot represents the latest attempt to solve one of the most vexing problems in the world of robotics - variable terrain. Before, a robot could be optimized for one kind of ground: Drones with tank treads were ideal for mud and sand. Wheeled robots worked best on roads. Legged bots could traverse rubble. Segmented machines similar to snakes were ideal for slithering through grass and scaling poles.
For dual roles on varied surfaces, roboticists have tried to combine one or more modes. "Chaos," the U.S. Army's search-and-rescue robot has treads that double as rudimentary legs, for handling the toppled walls and piled debris from collapsed buildings. A snake-bot from a company called Sarcos features treads on its end segments, allowing it to twist into the rough shape of a miniature tank, making it road-worthy between vertical ascents up poles.
Papanikolopoulos' transforming drone is the first to tackle variable terrain by simply flying over the rough patches. At least, it's the first to "leap" over obstacles under its own power. So-called throw-bots - either dumbell- or ball-shaped - can be tossed over walls and hills by their human operators before rolling out on their own wheels or, in the case of the Israeli EyeBall, rotating to point its onboard camera.
The major downsides to the University of Minnesota transformer are cost and complexity.
"It was actually more efficient to design the robot with two completely independent motor systems than to try to design a transmission that would allow the low-speed wheel motors to power the rotors or vice versa," IEEE Spectrum notes.
The rotor system alone cost $20,000 to build. "With that in mind, future developments for this platform will focus on making things simpler," the magazine notes. But not too simple: That's a market the unwieldy throw-bots have cornered.