New algorithms will help drones fly through small, cluttered spaces without hitting any objects.
Drones are effective in completing monumental tasks that are beyond human reach. But one of the biggest challenges is how to fly them around smartly without letting them hit an object and how to deal with unexpected circumstances like wind and weather.
Buy Now: Sony PlaysStation VR In Stock Here
Researchers from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed software that can help drones avoid obstacles but in a different manner. The new algorithm does not focus on the objects that a drone has to avoid but on the empty, navigable space between them where it can fly through.
“Rather than plan paths based on the number of obstacles in the environment, it’s much more manageable to look at the inverse: the segments of space that are ‘free’ for the drone to travel through,” said Benoit Landry, graduate researcher and lead author of the study.“Using free-space segments is a more ‘glass-half-full’ approach that works far better for drones in small, cluttered spaces.”
MIT has released a video where a small quadrotor drone is seeing flying over, under and around a ‘forest’ of strings, doing donuts and figure-eights and maintaining a speed of 1 meter per second while flying through the 10 square foot space. Algorithm segments space into obstacle-free regions and then link them together to find a collision free rout to negate object-filled environment.
Another approach was demonstrated by Ph.D. student Anirudha Majumdar where fixed-wing plane is shown plotting a course to avoid obstacles in real time. It searches a library of pre-computed funnels with 40 to 50 trajectories and selects the one that does not intersect with any obstacle even survives a gust of wind.
“As the drone flies, it continuously searchers through the library to stitch together a series of paths that are computationally guaranteed to avoid obstacles,” said Majumdar. “Many of the individual funnels will not be collision-free, but with a large enough library you can be certain that your route will be clear.”
For having a clear sense about surroundings, motion capture optical sensors and an on-board inertial measurement unit (IMU) have been attached with quadrotor drone that help estimate the exact positioning of obstacles. While algorithms used for fixed-wing plane can generate a number of flight plans and can switch between at will.
Now the industry will determine which approach is actually suitable and mature enough to be used in real drones. Jingjin Yu, an assistant professor of computer science at Rutgers University talks about the ingenious technique of combining on and off board sensors to determine the drone’s location.
Don't Miss: See the first leaked Black Friday 2016 Ad
“Enabling dynamic flight of small, off the shelf quadcopters is a marvelous achievement and one that has many potential applications,” said Yu. “With additional development, I can imagine these machines being used as probes in hard to reach places from exploring caves and to do search and rescue in collapsed buildings.”