Computational creativity and the future of AI

Goshawk flight could inspire next-gen agile UAVs


January 23, 2012

Analysis of the way a goshawk flies through cluttered forests has revealed a critical safe...

Analysis of the way a goshawk flies through cluttered forests has revealed a critical safe speed that suggests a roadmap for next-generation UAV research (Photo: Ben Fredericson)

Research into goshawk flight could inform the design of next generation UAVS. Where prior research into bird flight has focused on steady flight, new research from MIT examines the patterns of birds adept at flying in "cluttered environments" to find principles applicable to robot motion planning. It's research that might one day find practical applications in engineering, including fast, agile UAVs.

The research involves the random generation of simulated forest environments with a known, quantifiable density. It has identified that for any density, there is a critical speed above which a bird in a forest is absolutely certain to crash. Below that speed, the bird is able to safely plan a trajectory indefinitely. The margin between absolute safety and critical failure is narrow.

MIT's research not only suggests that goshawks must fly below a critical speed to avoid crashing, but that even just below that speed, the "inevitable collision regions" are sparse enough that the goshawk can fly straight for the majority of the time. In other words: flying just above the critical speed is impossible, flying just below it is easy - for a goshawk, at any rate.

MIT mathematician Emilio Frazzoli thinks that a goshawk gauges the density of the forest and adjusts its speed to suit, an intuition shared by downhill skiers.

"When you go skiing off the path, you don't ski in a way that you can always stop before the first tree you see," he said in an MIT News Office release. "You ski and you see an opening, and then you trust that once you go there, you'll be able to see another opening and keep going."

It's a different approach to that adopted by current-generation UAVs, which have a stopping distance within the range of their sensors.

Applied to robots, co-researcher Sertac Karaman asserts that this critical speed will prove decisive regardless of the navigational algorithm employed. Instead the problem seems to be assessing the density of the environment and identifying that critical speed, in order to fly below it.

MIT is now collaborating with Harvard to analyze actual bird flight in cluttered environments, assessing whether an environment can be too dense for a bird's liking. Preliminary research, using pigeons, is "very encouraging" according to Frazzoli. The researchers are additionally developing a computer game to assess how adept humans are at identifying, and flying close to, critical speeds.

The following video shows simulated bird flight through a random, cluttered environment at close to the critical speed:

About the Author
James Holloway James lives in East London where he punctuates endless tea drinking with freelance writing and meteorological angst. Unlocking Every Extend Extra Extreme’s “Master of Extreme” achievement was the fourth proudest moment of his life.   All articles by James Holloway
Post a Comment

Login with your gizmag account:

Or Login with Facebook:

Related Articles
Looking for something? Search our 31,561 articles