Blogs

3D Robotics

Autonomously navigating warehouses fast (with Pixhawk)

Posted by Chris Anderson on February 12, 2016 at 12:36pm

Interesting proof-of-concept by DARPA researchers using a Pixhawk-powered drone to autonomously navigate a cluttered warehouse environment at 20 m/s.

DARPA's Fast Lightweight Autonomy (FLA) program recently demonstrated that a commercial quadcopter platform could achieve 20-meters-per-second flight while carrying a full load of sensors and cameras. The FLA program aims to develop and test algorithms that could reduce the amount of processing power, communications, and human intervention needed for unmanned aerial vehicles (UAVs) to accomplish low-level tasks, such as navigation around obstacles in a cluttered environment. If successful, FLA would reduce operator workload and stress and allow humans to focus on higher-level supervision of multiple formations of manned and unmanned platforms as part of a single system. Through this exploration, the program aims to develop and demonstrate the capability for small (i.e., able to fit through windows) autonomous unmanned aerial vehicles to fly at speeds up to 20 m/s with no communication to the operator and without GPS. 

This testing was carried out at Joint Base Cape Cod / Otis Air National Guard Base.

For more information on the FLA program, visit: http://www.darpa.mil/program/fast-lig...

E-mail me when people leave their comments –
Follow

You need to be a member of diydrones to add comments!

Join diydrones

Comments

  • LightWare - Laser Developer February 15, 2016 at 12:39am

    @JB - For this project I was thinking in terms of some indoor navigation but I agree that there's lots of overlap with the CC project. There's a surprising amount of interest in having things buzzing around factories and warehouses.

  • JB February 15, 2016 at 12:24am

    LasDev

    Might have to take you up on that with the CC project... 

    I need an edge for the OBC quadplane. ;-)

  • LightWare - Laser Developer February 14, 2016 at 10:36pm

    Anyone want to try this using an SF40/C as the detection LiDAR please PM me. Let's see if we can beat DARPA. The SF40/C is very easy to interface using a serial port and can provide navigational vectors without any external processing. It even has hardwired alarms that can be used for basic keep-left, keep-right avoidance control.

  • Jack Crossfire February 14, 2016 at 10:48am

    The autopilot goes even faster with Curtis Youngblood at the controls.

  • Rob_Lefebvre February 13, 2016 at 9:42am

    Is this the end result of a $3M of contract?

    https://www.fbo.gov/index?s=opportunity&mode=form&id=ac1efe...

    Fast Lightweight Autonomy (FLA) - DARPA-BAA-15-16 (Archived) - Federal Business Opportunities: Oppo…
  • Rob_Lefebvre February 13, 2016 at 9:32am

    I think it's this headline that really sets the stage for the entire thing to be misinterpreted.

    "Interesting proof-of-concept by DARPA researchers using a Pixhawk-powered drone to autonomously navigate a cluttered warehouse environment at 20 m/s."

    That should really read:  Quadcopter manually flown through an open corridor at 20 m/s.  Then demonstrates ability to navigate through a warehouse full of obvious optical tracking targets at 1 m/s.

    Not nearly as impressive.

  • JB February 12, 2016 at 8:15pm

    +1 Matt and Rob

    It clearly says "tele-operated" on the video on the first run. 

    Don't DARPA engineers have youtube behind their firewall or know how to google FPV racing?

    Impressed? Not.

  • Darius Jack February 12, 2016 at 4:38pm

    http://www.theverge.com/2016/2/12/10981740/darpa-drone-autonomous-f...

    "

    DARPA used a hodgepodge of different drone parts to create the UAVs, including a commercial DJI Flamewheel 450 frame and a 3D Robotics Pixhawk onboard autopilot system. In its first successful test, DARPA got its drones flying autonomously at the desired speed and also tested the drone's ability to "see" obstacles using cameras, inertia measurement devices, and LIDAR and sonar sensors. "

    More from Las Vegas CES 2015

    http://paper.li/Solarpvpanel/1451778116

    Global Drones

    http://paper.li/Solarpvpanel/1438605902

  • Tobias Witting February 12, 2016 at 2:53pm

    yes Rob, this is very misleading. I thought: oh wow. 20m/s obstacle avoidance...

  • Patrick Poirier February 12, 2016 at 2:36pm
    Well, all boxes have same size with 5 stickers perfectly aligned and the middle one is green. Yep I can certainly try it with Randy's Balloon_Finder !!
This reply was deleted.