The Open Lidar Project - Hack the Neato XV-11 Lidar for a $400 Bounty!

Hey everybody,
I was inspired by Ladyada's awesome Open Kinect hacking bounty, so I decided to have a little fun myself. I thought, "what's something else that needs hacking?", and it occurred to me that everyone has been talking about the Neato Robotics XV-11's sweet laser rangefinder (, but nobody has hacked it yet (or at least talked about it) .
So, to get things started, I'm offering up $200 of my own money (plus RobotNV is offering another $100!, and Matt Trossen of Trossen Robotics has offered another $100!) in order to get someone to hack it and publish open source documenthttp://www.trossenrobotics.comation/drivers for using it on a robot.

You can find more details here:

Tell your friends! Let's get this puppy hacking and getting useful data back.

Views: 4912

Comment by Jarek Ignas-Menzies on November 15, 2010 at 11:30pm
Looks promising, too bad people have to get the hole vacuum just for the sensor.
Comment by William Cox on November 16, 2010 at 5:17am
True, but keep in mind that it's only $400 for a sensor who's next closest rival is $1200. I'm also hoping that interest in this contest might help spur a community and we can roll our own lidar for cheaper. I already have rough plans.
Comment by Patrice Mainville on November 16, 2010 at 2:53pm
Mmm, that a point a view, The neato vacuum is real slow on the ground, so, probably the sensor is slow too and not enough fast for a UAV need, but, we still have to see! the Kineck cracked seem more promising to me!
Comment by William Cox on November 16, 2010 at 4:08pm
It'll remain to be seen, but all reports have indicated the sensor scans 360 degrees at 10 Hz. That's pretty fast, especially if we can hack it to only scan a portion of that range. There's also a possibility we can increase the scan rate in exchange for a lower accuracy.

The Kinect is an awesome projet, and I'm not comparing the two in terms of "epicness of achievement", but it is worth noting that the Kinect will almost always require a high power computer in order to do anything useful with it. It's also large, bulky, and requires a lot of power. It seems much better for larger or stationary systems to me.
Comment by Daniel NIkolov on November 16, 2010 at 4:32pm
the only bad thing is that this is scanning in only one plane while a lidar can scan in many planes at the same time. In order get any type of usefulness from this you would need a pan and sweep system. Panning about the y-axis. It is possible to dump the readings within a certain range and just focus on 180 degree or smaller. But you would need to buffer the correct readings in through an external timer. I guess if someone can figure out the best way would be to do all the corrections on the back end in the computer so you could narrow the field of view.
Comment by Ritchie on November 16, 2010 at 5:45pm
Please search "Colin Ord Scanner" you need a webcam, laser and clear cocktail stick (cost me £3 in all).
I've updated it to .net 3.5 and emailed it to Colin so he can go through the code for his next incarnation. Whilst his uses outside apps mine uses all internal. Im trying to get the plotting he has on his working but as mine goes in realtime it seems to hate it.
I hopefully will be getting a faster embedded setup to try and port it so its small enough for the UAV but never can be sure with Santa.
Comment by William Cox on November 16, 2010 at 7:53pm
@Daniel, I'm not sure what your referencing, but very few sensors are able to capture a 3d field. Most scanning lidars, except for the newest systems from Velodyne, only capture 2d data - distance info on a plane.
Comment by William Cox on November 16, 2010 at 7:55pm
@Ritchie, that's a neat setup. Not sure how well itll work with longer distances, though. Have you tried an eye safe IR laser yet?
Comment by Ritchie on November 17, 2010 at 5:57am
eye safe :D never believe IR lasers are safe. Even the ones in DVD players which are low class are dangerous and worst of all you can't see them working or not.
I've scanned out of my window before with it. Which is about 30m/90ft full distance and it worked pretty well but there is a big downside to the ColinOrd approach. Scanning inside is easy but long range become hilarious due to the 30° laser needed to go across your screen. I walked across the lawns at night and did it.
The approach I'm adapting uses the camera's individual setup to measure the distances. This reduces variables to how just the shown variance on screen after a calibration. Distance from laser to camera is set and parallel. Calibration shots are needed though before real world use (wall pictures at set distances and level).
Comment by William Cox on November 17, 2010 at 7:32pm
Just got an email from RobotShop. They are upping the bounty (again!) and offering to refund the cost of the robot, if you buy it from them!
No excuses now.


You need to be a member of DIY Drones to add comments!

Join DIY Drones

© 2018   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service