3D Robotics

ROS interface for PX4Flow board


The PX4Flow smart camera board will be released in a few weeks (designed by ETH, manufactured by 3D Robotics). There will be a ROS library to drive it, along with the PX4 libraries, Lionel Heng from the ETH Computer Vision and Geometry site announced on the ROS news site

I would like to submit a new repository for indexing: https://github.com/cvg/px-ros-pkg

This repository currently contains a ROS package for interfacing to the soon-to-be-released PX4Flow optical flow board (coming soon from 3D Robotics). We have plans to add a ROS interface to the PX4FMU autopilot, and software for MAV autonomy as showcased in our recent IROS paper (http://www.cvg.ethz.ch/MAV ) in the coming weeks ahead.

Best Regards, Lionel Heng


E-mail me when people leave their comments –

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

Join diydrones


  • Curious if there is a status update (from Chris) on the release date.  I know how small engineering problems can turn out to be bigger than first thought.

  • Just an update on this... Chris Anderson contacted me to let me know that while they haven't finalised a release date or price, they're still hoping to get them out before end-of-year. They have a few hurdles, but are apparently close to being ready to release these boards.

  • 3D Robotics

    Tim, I'll check with the team and see what the current ETA is. I'm not sure who you were emailing, but as you probably guessed the sales team is not authorized to discuss unreleased products or potential release dates. 

  • I was in contact with the sales team at the start of November and they said they expected to release them late November, early December. Obviously that hasn't happened, so I tried contacting them at the start of this week. Unfortunately they've chosen not to respond... I'm guessing they think I'm pestering them... which I am, since I *must* spend my research funds before the end of next week or I'll lose them.

    Perhaps if someone at 3DR sees this they might get some information for us and stop me having to buy something inferior from somewhere else (I will spend my money by the end of next week).

  • Tim, I was wondering the same thing. The website has been saying that it "will be released in a couple of weeks" for the past two months. 

  • Any update on the release date for these?

  • ramboky: the on-board gyro is typically used to identify (and remove) the rotational component of flow about the axis orthogonal to the camera imaging plane.

    Todd: the benefit of this sensor is that it can not only give more accurate flow estimates, but also handle more complex image/flow analysis tasks, such as image dilation estimation. This can be used in auto-landing systems to control vertical speed relative to horizontal speed, estimate approach speeds to objects, etc. This is very useful for doing vision-based guidance, but less necessary if you're closing your control loop with eyes and thumbs (i.e. manual control).

  • Sorry typo, should have been "flow sensor with the current APM".  

  • What advantages would it offer over using the current Optical flow sensor on the current APM.  It just seems to me like it would be putting a high-end sensor cluster and dashboard in a late model Ford Taurus.  Not saying that its not possible, just don't see the point.  Enlighten me.       

  • Looks like a camera...?

This reply was deleted.