Autonomous Multi-Floor Indoor Navigation with a Computationally Constrained MAV

This video shows our results on autonomous multi-floor indoor navigation with a quadrotor. We designed a system that is capable of autonomous navigation with real-time performance on a mobile processor using only onboard sensors. Specifically, we address multi-floor mapping with loop closure, localization, planning, and autonomous control, including adaptation to aerodynamic effects during traversal through spaces with low vertical clearance or strong external disturbances. All of the computation is done onboard the 1.6Ghz Intel Atom processor and uses ROS for interprocess communication. Human interaction is limited to provide high-level goals to the robot.

• Category

  Science & Technology

[iliasasdf]

great

[mortenrobinson]

Do a search on "autonomous flight with a kinect". They've fitted a kinnect onto a quadcopter and uses it to do 3D mapping of it's environment :)

[mortenrobinson]

It would compensate

[Jordan Liske]

you have some serious wobble there, more tunning?

[Anna French]

The real drones are actually pretty hard coded (they know exactly where to go and for the most part know what to do every step of the way a priori), so they don't really use online algorithms to figure out their position etc, which is what this one does.

[adam3176]

This is low level Bullshit.. The real drones are To secret and are 400 years ahead of this garbage

[Freedom1man]

How would it respond to a direct hit from a 12gauge shotgun?



[Brent Gerson]

Oh god those drones from Terminator 3: Rise of the Machines

[TheBenjaneer]

The issue is the small intel atom processor. Also if you watched the video closely you would notice that they do not have a 3d sensor on-board (2d camera and a 360 degree laser scanner as well as an imu); yes they could do 3d, but no, it would not be effective on this platform.... 2d mapping is not a waste of time; and it is no small feat on its own that they have effective 2d mapping working on this tiny robot using ROS.

Kudos to the team who did this!

[xxyy3003]

how does the robot know in which dircetion it is pointing?