Hello everyone, I feel the time has come to make an announcement, which I've been avoiding for a few weeks now, but which I can't put off anymore.
As some people are aware (and some maybe not) for a little over a year I have been doing development work and community support as a full time job, with the generous support of 3DRobotics. This represented a major shift in the direction of my career from industrial mechanical engineering, to this burgeoning field of aerial robotics. However, 3DR has decided to discontinue that support. As the project has grown in scope and complexity, the overhead time requirements just to stay on top of program development and developer group communications has grown with it. Prior to the support provided by 3DR, I was moonlighting while maintaining a "day job" outside of the industry, but that ended up burning me out and I won't go back to that situation.
This has led to me scrambling to figure out what to do next in order to pay the bills, etc. I actually saw the writing on the wall several months ago and began working in the direction to solidify my future in this industry, but the change happened a little before I was ready.
In any case, regretfully I must announce that my efforts to directly support the community, answering questions, analyzing logs and doing general improvements to the code will be ending. I am committed to seeing AC3.3 finally go Stable for Tradhelis, and updating the Wiki to be up to date with this. But beyond that, I'm not sure how much I'll be able to help out.
My hope is to produce revolutionary new turn-key UAV helicopter systems and use them for professional services, or 3rd party sales. If I can make this transition, I will likely continue to do Alpha testing on new AC code and contribute new features. I have almost completed design of an excellent new 700 helicopter design, and hoped to do a proper 500 size helicopter for mapping after that, but likely won't have the cash to bring these to market on my own at this point. Unfortunately, due to timing, I may be entering into development contracts with other companies which will be a bit more "closed" in nature.
It is my hope that the community can get to the point where it is self-supporting. We already have several members who are quite successful with helicopters, and are still coming around to help others, and that is encouraging. Hopefully this can continue to grow. I don't want to see Helicopter support in the code die. Helicopter mechanics are largely a "solved problem", and this platform delivers superior combination of flight performance. VTOL of a multirotor, range and speed similar to many of the foam airplanes used today, and absolutely unmatched stability and weather tolerance, while carrying payloads greater than any platform of similar size. And Ardupilot is the only full-featured autopilot system that will fly a UAV helicopter, unless you purchase military-grade autopilot systems. I feel, and I think the worldwide UAV community agrees, that Ardupilot is the most reliable, full-featured, high performance and cost effective autopilot system available. The fact that it is truly open source, distributed under a GPLv3 license means that is flexible, extensible and capable of meeting the needs of all users, big and small, while reflecting a share-and-share-alike mindset. And I don't need to tell you guys that Ardupilot is flexible and cost effective on the hardware side as well, capable of running on a wide variety of hardware systems from many companies offering differing capabilities to meet various user needs. This could be anything from tinkerers and hobbyists, to small service companies, UAV builders, academics and research groups. It's important that the project continues to keep UAV technology accessible to all, and not be locked down by big corporations or governments.
Given the situation where some of the developers now have more "free time" what exactly can we put on the ground to support them going forwards as a business case?
We can point fingers until they're blue, but I doubt that will help.
Currently there's a bunch of Open Source work that needs to see the light of day to move the whole thing forwards. How do we rally behind this and get the ball rolling? We can't just leave it up to the few developers that are still eager to do something.
If the we could somehow stimulate enough interest in the cause maybe the community could help set something up. There's already been some movement but it needs more support.
I also agree that as a community we need to grow beyond DIYDrones, where the focus is just one making drones fly...we need to expand into the support of real life applications of UAV's, whatever which way benefits the cause. There's so much "need" that can be solved if we really want to.
Open source is great but open source devs have mortgages and kids too!! I would gladly support a $10-20 a month donation to continue the development of the code. How many others?? But even if that would work... how would I know I'm not just subsidizing R&D for 3dr or others who are no longer willing no to support DIY products? If in fact that is the case...
Should we start a concerted effort to bring new controllers and their GC's to the forefront? Seems it would be in our best interest to have some choices in a year or so. Or do we think PX4 and APM will be with us for the next 5 years?
Many people would like to see the Pixhawk 1 design remain in production for some time. Particularly commercial hardware vendors who have incorporated into their own products. They don't want to have to go through reengineering for nothing.
The design is stable, and not obsolete in any way. Not really any good reasons to discontinue it.
It can co-exist with other newer designs such at the Pixhawk 2 and PixRacer, just happily.
Wish you all good. I have learn a lot from reading your posts and I would like to thank you for that. As a aeromodeller and open-source enthusiast, I understand your point of view. This has come to me as a warning. I live in Brazil, have dual european citizenship, have been working with IT for all my live and those two last years I've been planning to change my career path to startup an Uav consulting company, probably in Europe ( bad time to startup anything in Brazil).
The warning is because I have learned ( with your experence ) that it could be more difficult that what I was expecting, putting into perspective that you were working in the "eye of the hurricane " for 3dr etc. and in USA that the market is hot.
I have tried to contact 3dr a few times to see if they had any interest on Brazil but got no response at all.
I'm not giving up to the idea to have the "perfect job" to work with something that I love, but for sure it will not be easy.
Linux and open-source were to me kinda the same thing. I started using linux, building datacenters, ecommerce solutions and ISPs back to the 90s but felt that it was very difficult to me to carry this on in Brazilian market ( fighting micro$oft) . I can say that I almost give up, but not this time.
Hope we still see you around :) !
There is a lot of UAV development in Brazil. Intel has a dozen people in Brazil working on Ardupilot.
Hi Rob, sorry to hear that, hope it will end up better chance to chase your own interest in aero robotics. I had same interest in heli uavs, i have a goblin urukay 3 blade as my testing bed. hope you will bring something new to px4.io since there is no out of box helicopter support.
Good luck Rob, thanks for all your help over the years.
Good luck mate, your contribution was awesome. Some have tried to contribute to Dronecode also with no luck. It seems that 3DR has closed the "Loop" so to speak on many opportunities. Not to worry, open source and lean startups are the key to future innovations and the era of new markets. I think the aerial robotics landscape will be very different in the next 10 years with new key players replacing the old. All the best!!!
Sad to see you go Rob! The end of your support will be noticed hard, but Family always comes first!
I wish you the best of luck with your life!
Good people over there. Cool stuff.
Yes their stuff is pretty cool too.