Our new video shows position hold (the position can be moved on the fly, so its actually full position control) using the PX4FMU autopilot plus PX4FLOW on the AR.Drone 2.0 frame. We have created a set of example apps to show how flow can be used for position control - these are however just textbook like examples, no bells and whistles, no complete sensor fusion and no attempt to max out on performance.
This post is also to announce the availability of the PX4FLOW open source firmware. We are only releasing it now since we wanted to do all restructuring necessary before we release it, so that any open source contribution won't be left behind because the structure changes dramatically. We invested a full person month into getting things clean, nice and simple, and we hope developers will appreciate the effort.
A quick guide is available here for early adopters / testers interested to fly this immediately. As for any beta releases, you will need to have the PX4 toolchain installed (installation guide).
Comments
Hello! Please, do you know if px4flow works by default with flight modes position hold, loiter and alt hold? I just need to connect the i2c output of px4flow in pixhawk i2c input? I'm using a split because I have the external mag on that same entry ...
Thanks for the tip, I've switched to an STM32F4, that fixed the problem.
It processes at 200-250 Hz and outputs at 100 Hz steady. The latencies you see have to be on the receiving side, not on the PX4FLOW side.
Hey, I'm trying to replicate this in Arducopter. The strange thing is I get wildly varying times in between flow messages, sometimes they come at 0.01 second difference, but other times the time difference skyrockets to half a second.
In qgroundcontrol the flow message does come in at around 100Hz, is this due to the use of 115k serial, or does the PX4 Flow output at 100Hz there as well?
The tape markings are not part of the localization, those are there for a different project. The height is not limited regarding optical flow (the accuracy just scales linearly). However, if you do not provide a different altitude over ground sensor, you will be limited to ~4 meters by the ultrasound, we have not yet tested the full outdoor range though.
so this is position control using a downward facing camera right? The floor looks pretty non-textured too! (Or does the system require those tape marks on the carpet?) Is there a specific height above which the system doesn't work? Looking good!
Great work! We are moving forward.