We are trying to get a wind readout from a multirotor to groundstation. We are having difficulty figuring out how to do this. We want to fly or hover and be able to get the wind direction and velocity. We are not finding how to do this with latest version. If it can be done does the aircraft have to be in motion or can it be hovering?
1 Like
Wind estimation with a multi rotor is possible but non-trivial. I’m not aware of any open source system doing this yet (I have seen it in commercial systems though).
We do support wind estimation with a pitot probe in aircraft.
Any guidance would be great. We are looking for a digital solution over a pitot tube. I would like to get direction for my team members that would be doing the work. We want to get a readout similar to commercial aircraft flight management systems.
Well, the commercial FMS use multiple pitot tubes and very often an AoA vane to generate that data.
Without those you would need to model your vehicle in terms of dynamics and drag, it’s propulsion system and estimate the the wind based on your force input vs. a wind still hover condition. That’s what people have done successfully.
Yes, the second option. It seems that with GPS course and ground speed that could be compared to IMU various pixhawk data. I just need direction on how this could be done by smart guys at our university lab.
I gave you just that. Anyone with a controls background will understand what that encompasses (and also that it involves a reasonable level of work and experimentation). Then a literature research on http://scholar.google.com is a good start. I would point you to a finished algorithm if I would know of an available implementation.
Thank you very much. We will be using an older 3dr quad with the pixhawk on A to B missions. We will be building flight planning software and need wind readout to ensure actual and flight planned winds are similar and have enough energy to complete the mission or divert. Thanks for your help.
As Lorenz pointed out, you need to model your vehicle in terms of dynamics and drag. This past summer some of our students worked in that area and created an aerodynamic model of a sUAV in a controlled facility. They used the RotCFD tool and below you can see the approach they took for their research.
This paper might be helpful to you too since they used a 3DR quad as well.
