We started writing about AirBurr, the robot that would become Gimball, in October of 2009. Over the last five or so years, we’ve watched it change and evolve through what by now has to be more than a dozen unique versions until we were introduced to Gimball at ICRA 2014 in Japan. This is a robot with quite an academic development history, and that makes us particularly excited to see it win US $1 million in the first Drones for Good competition (an event organized by the United Arab Emirates government), not as a research project, but as a commercial one.
First, some background on Gimball:
And here’s the onboard footage from the competition finals, showing the robot locating simulated people in a simulated disaster zone, flying through very narrow areas outside of the pilot’s line of sight:
Really simplifies the navigation, doesn’t it? When in doubt, smash into stuff! It also simplifies dealing with laggy video or controls, since the drone is perfectly happy to not be under any sort of control for short periods of time. And smashing into stuff can actually be a good thing: in 2013, Flyability founder and CTO Adrien Briod was working on a way to use collisions to build maps.
Since Flyability is a company developing a new product, their plan is to take the million and use it accelerate the development of their first commercial drone with a focus on search and rescue by adding infrared sensors and visual SLAM capability. And, you have to figure that they’ll spend at least a little bit of the prize money throwing one awesome victory party.
[ Flyability ] via [ Drones for Good ] and [ RoboHub ]
Evan Ackerman is a senior editor at IEEE Spectrum. Since 2007, he has written over 6,000 articles on robotics and technology. He has a degree in Martian geology and is excellent at playing bagpipes.