MIT Mapping System Makes Drones SelfAware

first_imgLet us know what you like about Geek by taking our survey. Stay on target If drone deliveries are the future, NanoMap is the present.MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) developed a system that allows drones to consistently fly at 20 mph through dense environments without incident.How? By embracing uncertainty.“Overly confident maps won’t help you if you want drones that can operate at higher speeds in human environments,” graduate student Pete Florence, lead author on a related paper, said in a statement. “An approach that is better aware of uncertainty gets us a much higher level of reliability regarding being able to fly in close quarters and avoid obstacles.”Researchers use depth-sensing to stitch together measurements about the drone’s immediate surroundings (via Jonathan How/MIT)Existing approaches—like simultaneous localization and mapping (the prescient acronym SLAM) or “occupancy grids”—often use complex geographic charts to inform unmanned aerial vehicles where they are relative to obstacles like trees, buildings, or power lines.But what if someone builds a new house or erects more high-voltage cables that aren’t mapped before Amazon sends its unmanned aerial vehicle on a dispatch?That’s an expensive accident waiting to happen.NanoMap, meanwhile, relies on the fact that change is the only constant; a drone’s position, much like humans’, is unpredictable, and cannot be neatly planned.“It operates under the assumption that, to avoid an obstacle, you don’t have to take 100 different measurements and find the average to figure out its exact location in space,” according to MIT. “Instead, you can simply gather enough information to know that the object is in a general area.”The simple concept requires complicated technology: A CSAIL team use depth-sensing to stitch together a series of measurements about the drone’s immediate surroundings. This allows it to make motion plans for its current field of view—and anticipate how it should move around the hidden fields of view it has already seen.“It’s kind of like saving all of the images you’ve seen of the world as a big tape in your head,” Florence said. “For the drone to plan motions, it essentially goes back in time to think individually of all the different places that it was in.”If a drone’s estimated location is off by even just a small margin, it can easily crash (via Jonathan How/MIT)CSAIL’s NanoMap software brings a certain awareness to the drone; the self-aware machine understands that it doesn’t know its exact orientation at any given moment.Without the element of uncertainty, if the device drifted just 5 percent away from where it was expected to be, it would crash more than once per four flights.When accounting for uncertainty, that rate is reduced to 2 percent.NanoMap is especially useful among small drones traveling through small spaces, for the purpose of search-and-rescue, defense, package delivery, and entertainment. But it can also be applied to autonomous navigation.A paper co-written by Florence and MIT professor Russ Tedrake, with research software engineers John Carter and Jake Ware, is available to read online.center_img Climate Activists Use Drones to Shut Down Heathrow Airport Next MonthUPS Wants to Bring Drone Deliveries to U.S. Hospitals last_img

Leave a Reply

Your email address will not be published. Required fields are marked *