Watch: Drone programmed to duck, dive and avoid obstacles

3 Nov 2015 | Author: | No comments yet »

MIT Researchers Say Their Drones Can Safely Navigate Forests At 30 MPH.

An MIT researcher says he’s developed a system for flying drones safely around areas thick with trees, and to do so at speeds of up to 30 miles an hour.A team of researchers from the Massachusetts Institute of Technology have just released a video showing a drone autonomously navigating its way through tightly clustered trees at 30 miles per hour, the most advanced system of object detection and avoidance we’ve seen to date.

We may still be playing the waiting game for Google’s self-driving car, but in the meantime, there’s been an autopilot development in the drone world. It’s called Pushbroom Stereo and was developed by Adam Barry and Russ Tedrake, both with the Robot Locomotion Group at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL), a hotbed of innovation in the world of robotics and computer vision. “Everyone is building drones these days, but nobody knows how to get them to stop running into things,” says Barry, who developed the system as part of his thesis with Tedrake, an MIT professor. “Sensors like lidar are too heavy to put on small aircraft, and creating maps of the environment in advance isn’t practical. The craft was able to do this using a stereo-vision algorithm that rapidly detects and avoids objects immediately in front of the craft, according to a release from MIT. If we want drones that can fly quickly and navigate in the real world, we need better, faster algorithms.” The results look like lost scenes from the Top Gun special effects sequences.

Andrew Barry, a PhD student at MIT, has developed a computer program that allows the drone to autonomously detect physical objects like trees and enables them to perform maneuvers to avoid these obstructions. The problem itself is fairly simple: Small-scale UAVs like the ones many amateurs and tinkerers own aren’t designed to autonomously avoid obstacles because they aren’t capable of carrying the weight of the processors they’d need to analyze the world around them and react to it. Earlier this year, the Defense Advanced Research Projects Agency funded another MIT group and a team at research outfit Draper Lab to develop software that could run on hardware carried by a miniature drone but powerful enough to allow it to navigate through a cluttered room at high speeds.

A drone’s camera might record hundreds of frames per second, and analyzing the depth of field for every object in each frame takes some serious firepower. There are currently a number of efforts to develop obstacle avoidance systems for drones, and Qualcomm has built a reference design for drones using standard components. This new system maps only 10 meters out, making it much less computationally intensive, although what happens if a squirrel suddenly enters the previously empty path between you and that 10-meter mark is not exactly clear. But Barry only sees more opportunity there. “Our current approach results in occasional incorrect estimates known as ‘drift,’” Barry said. “As hardware advances allow for more complex computation, we will be able to search at multiple depths and therefore check and correct our estimates.

But because it’s flying at 30mph, it doesn’t need to spend time analyzing all those extra frames and objects. “While this might seem limiting, our cameras are on a moving platform (in this case, an aircraft), so we can quickly recover the missing depth information by integrating our odometry and previous single-disparity results,” they write in a paper published on ArXiv.

Here you can write a commentary on the recording "Watch: Drone programmed to duck, dive and avoid obstacles".

* Required fields
All the reviews are moderated.
Our partners
Follow us
Contact us
Our contacts

ICQ: 423360519

About this site