Replies to This Discussion

ModeratorPermalink Reply by Gary Mortimer on April 12, 2012 at 11:06pm

Heres the first quad http://en.wikipedia.org/wiki/De_Bothezat_helicopter

Will follow this thread with interest as I would like to fill the sUAS News FB timeline with more info.

http://www.facebook.com/suasnews

The Paparazzi project would have to sit high in your list along with RCAP they both happened pre Arduino.

Also you must add Jack Crossfires work, to my mind several high profile projects have leant heavily on his work.

Oh the list is endless I will let others carry on.

• ▶ Reply

Permalink Reply by flyinghiho on April 12, 2012 at 11:12pm

Thanks Gary.  I appreciate it.

• ▶ Reply

Permalink Reply by Brad Hughey on April 13, 2012 at 8:13am

That's an interesting question.  The first electric multicopter I ever saw was the DraganFlyer, back in the late 1990's.  Since they don't claim to be the first on their website, they probably aren't.  However, they can't be far from it. 

It is very difficult to make NiMH or earlier battery technologies the center of a viable design, due to their low energy densities.  Lithium-ion cells weren't generally available before 1997, so that's about the time I'd expect to find the "first" electric VTOL craft.

• ▶ Reply

DeveloperPermalink Reply by John Arne Birkeland on April 13, 2012 at 3:57pm

The first DIY multicopter I know of, was a wood stick tricopter with plain head-hold gyros to keep it in the air. There is a rcgroups thread, but not able to find it any longer.

• ▶ Reply

Permalink Reply by David Conger on August 17, 2012 at 12:58pm

"What were the key moments in hobby Drone history." Let's see how long this post lasts. It is 100% fact so follow the links if you doubt any of it. 

 

I'd say Paparazzi was definitely the World First Open Source and Hobby autopilot in any measure. First to fly fixed and multirotor (Lisa/L). First to fly self navigated using IMU attitude sensing. First open source. FWIW: They still are the only ones that publish not just the designs, the complete source code, but also the Bill of Materials Parts list with part numbers. Building your own is still encouraged and 100% supported.

 

Paparazzi Project Started in early 2000s (2002/2003). Their orig. website is still online (see below).

 

More Facts/Firsts: 

First Arduino Based Autopilot: Paparazzi Arduino based (yes, first Paparazzi controller was Arduino) was circa 2003. Website from that era is still online here: http://www.nongnu.org/paparazzi/gallery_v0.html 

 

First dual processor hobby autopilot (Classix had dual LPC2148, an Overo Gumstix headers and is still supported). 

 

First to support multi-uav (TCAS logic).

 

First Quadrotor: Paparazzi Quadrotor (Antoine Druin/ENAC) was circa 2005. 

That first autopilot had GPS, IR attitude sensing, and autonomous self-navigation. From the orig. site: "The aircraft is able to perform autonomous takeoff ( hand-launch or bungee ), accurately follow a flight plan and even do some sort of rough landing. At any time it is switchable to manual or assisted control, for example to increase landing precision"

 

Just in case people are wondering Paparazzi is still very much alive and well. The ENAC team won the IMAV2012 (again) with their latest multirotor controller what has the processor, IMU on the same board 35mm x 35mm. More on NavGo here: http://paparazzi.enac.fr/wiki/NavGo_v3

 

More info here: http://paparazzi.enac.fr/wiki/Wall_of_Record

and here: http://www.nongnu.org/paparazzi/overall.html

Take note "DIY" play on wording in the title of the page above. Making the term Do It Yourself UAV up back in 2003.

Anyone who speaks for the whole hobby UAV community must mention Paparazzi. It's like talking relativity without mentioning Einstein. 

• ▶ Reply

Permalink Reply by Jake Stew on August 17, 2012 at 3:00pm

Good info DC.  To the OP... It depends a lot on who you ask.  There's a lot of fragmented projects out there and they all accuse everyone else of stealing their ideas.

In general, as far as I know, Paparazzi started the whole thing, the APM borrowed heavily from Paparazzi and so did OpenPilot.  The new PX4 project from PixHawk is a fork of the OpenPilot project.

The OpenPilot guys claim that APM stole a lot of their ideas and code, and that PixHawk forked their project without any attempt at cooperation, basically taking the code and running.

Paparazzi would have a good claim on the whole thing except that their dev team suffers from a lot of problems and their hardware is expensive and hard to obtain.  Basically, a number of major errors and fundamental strategy flaws have doomed the project to obscurity.

OpenPilot has done a lot of good work, but isn't an open hardware license.  They suffer from hardware availability problems, making it hard to get going with their project.

The APM runs on outdated processors and is somewhat limited by using simplified platforms and tools like Arduino.  However, picking a well supported processor with a huge dev community has allowed them to progress very fast.  Despite the terrible forum software here they've got a lot of users and devs because of the reasonable hardware price and the ease of getting involved with the development.

Hard to say how the next generation will do.  My prediction is that OpenPilot will improve, PixHawk will do well, the APM3 will stall and lead to the PX4 being the next most popular hardware platform, and Paparazzi will continue to fade into obscurity due to their continued lack of strategy or attempts to attract new users.

So the project to watch will be the merged PixHawk/APM platform and OpenPilot.  

• ▶ Reply

Permalink Reply by David Conger on August 17, 2012 at 3:37pm

My may reconsider some of those things when you read about the QuadShot Paparazzi project here: http://thequadshot.com/

STM32, VTOL, Hardware is easily available at prices comparable to anything else out there when comparing Apples to Apples...

 

• ▶ Reply

Permalink Reply by Jake Stew on August 17, 2012 at 11:34pm

The hardware situations may change now and again.  Their code may work well and is pretty advanced in many respects, but it's also a terrible mess and very poorly organized.

They didn't support NMEA GPS gear last I checked, and only supported overpriced IMU boards.

But their biggest mistake is choosing to only run on Linux (1.4% market share).  Excluding 98.6% of your potential users is a fatal flaw that cannot be overcome, even without the other problems.

Even if the Mac support they've been working on turns out to work well enough you're still talking about excluding about 90% of the market.

I ran Paparazzi on Linux for a bit and got tired of that, then ran it in a virtual box for awhile, but in the end it's just not suitable for the vast majority of users.  The GCS is ugly compared to APM and OpenPilot and it is graphically buggy.

Their code is scattered among as many files as it possibly could be, you have to edit all the settings in XML files, the settings are poorly documented and without proper examples, there's no IDE support, and every setting is compiled into the firmware.  Recompiling every time just to change waypoints or tuning settings is simply a major design flaw.

The worst part is that none of the devs see any of what I mention as a design flaw, or even a potential area for improvement.  They've missed the boat with user friendliness and standard, modern graphical controls for the interface and changing settings.  The computing world changed to drop down list boxes, checkboxes and the like, instead of editing settings in text files way back in 1995.  This serious failure to grasp what constitutes a modern interface is just the surface of underlying problems with the entire project.

This banging away at a command prompt attitude may demonstrate their skills and make them look cool at a programming convention, but it ultimately dooms their project to obscurity.

• ▶ Reply

Permalink Reply by David Conger on August 18, 2012 at 6:32pm

Well, I see where this is going. Dissapointing the moderators allow such toxic talk but I guess it's ok as long as it's not taling about APM. Sad really since Paparazzi project people have only positive things to say about APM and the good work that goes on here. 

• ▶ Reply

Permalink Reply by David Conger on August 18, 2012 at 8:21pm

Jake, I'd like to see the good side of what you write but your style comes across as you know the best way and Paparazzi isn't it.

Something I've learned with open source is it's a majority thing. If the majority feels something is important then it gets priority. A great many people use Linux. The US Govt is rethinking Windows as a platform. You may find to your horror someday Windows is not used by any project.

I really don't understand this statement: "banging away at a command prompt attitude". Such a negative sounding statement. The programs you do love to use was probably written by someone just like you describe. 

People here and on other forums for Open Source are a very understanding and tolerant bunch mostly. They spend a lot of their own time to try and meet the majorities needs so you have something as amazing as Ardupilot or Paparazzi or the others you mention. Trash talking them because they did not meet your needs is bad form. 

• ▶ Reply

Permalink Reply by Jake Stew on August 18, 2012 at 11:30pm

I'm not trash talking any one.  The OP asked for a history lesson on why things are they way they are and how they became that way.

I think the Paparazzi code is superior to the APM in performance.  Their hardware is also much more powerful.  They were also the first players in the game.

So now I've just said three positive things about them.  I take it that you're not disputing any of those statements.  Now we have to reconcile why they are a distant third in the autopilot game, at best.  If you have a better explanation than I've laid out then I'd love to hear it.

> I really don't understand this statement: "banging away at a command prompt 

I've played with Paparazzi plenty.  I've got no problem banging away at the command prompt to a certain extent.  But at some point it gets very tedious to open up a text editor, edit an XML file, recompile your firmware, and upload the firmware to your board for every minor setting you want to change.

The OpenPilot founders/devs actually started with Paparazzi and wanted to improve some of these things.  The Paparazzi devs made it clear that they had no interest in changing the "everything compiled in" structure, didn't want to change the GCS or help with a windows version, and so on.  That's why OP was started as another STM32 based autopilot.

I had essentially the same experience with them.  They have their own agenda with regards to what they want to work on, they don't see any need to change anything about their design philosophy or priorities, don't want anyone to work on making it easier to use, don't care about the cost or availability of the hardware, and aren't interested in expanding their potential user base or attracting new users.  Nobody is outright hostile there, but if your goals conflict with their ideas or threaten to expand the user base to a group they don't like then they offer no help whatsoever.

They still use an email list for their communication for goodness stakes.  If that doesn't clue you in to them being stuck in the past then I don't know what will.  I can't think of a much more intimidating tactic than making new users jump on the dev list for basic help.

In any case, if you have a better explanation or want to argue any of the facts I've stated as I see them then feel free.  I complain about certain aspects of the DIYdrones projects all the time here.  It is a testament to their project that I'm allowed to speak here as freely as I do.  

That's especially important because there are people *some* people that immediately cry anytime something is mentioned that they perceive to reflect poorly on their particular project.  These people immediately cry wolf and act like they're being personally attacked or that the statements they don't like are somehow vulgar or inappropriate.

Trying to win an argument by claiming the other person's discourse is "trash talking" or "toxic" is a very weak and disingenuous tactic which usually will lose you points even if your opposing point of view is actually correct, especially when it's clearly not true.

 

• ▶ Reply

Permalink Reply by Gautier Hattenberger on August 20, 2012 at 9:03am

Hello Jack and other people around,

I don't really like to do that, but I think I need to clarify some points. I'm one of the core developer of Paparazzi and a teacher at ENAC (a french public school). Our UAV Lab is composed of more or less 3 people not working at full time on this and some master/phd students helping from time to time. The other developers are also mainly from academic institutes. Like many other Labs working in this field we are developing our autopilot for academic research, educational purpose and competition (because it is a good way to meet other groups, find research fundings and it's fun). And like some other Labs we are just sharing our work by releasing SW and HW as open source.

You are complaining about the difficulty to use/configure/understand paparazzi. You are right, I do that myself almost everyday. We have a lot of ideas to make it better, but it will take a lot of time for a simple reason: time and people. We are missing both to make it "user-friendly" as you said.

But the most important thing I want you to realize, is about all that market share stuff. What market share ? We have nothing to sell. ENAC, as a public school, will never try to make money out of our work. We will never have any financial gain from Paparazzi, only research grant for the Lab to continue to exist. And as long as this is true, we will continue to work and develop on Paparazzi as a scientific and technological UAS platform.

Paparazzi is about flying. It is at the cost of the fancy commercial aspects. We are glad that people like David are investing money to provide hardware to a wider range of people, and other companies are doing the same with new hardware and software like Transition Robotics. The main code aims at providing a stable and reliable code for people using our system. By doing this, introducing big new features takes a lot of time. But the system is actually pretty modular. Feel free to add nice configuration tools around that. That's what the companies working with our system should do or have done in the past.

It is sad that you and probably a lot of people dropped Paparazzi because of its complexity or because it is a Linux based system. That's not a choice we have made to discourage people. But this is the only one we can support for now.

We will continue like this until more people are investing on the development, not only opening issues.

And again, we will continue to fly, because that's the reason this project started almost 10 years ago.

Happy flights to every one and we wish the best to the other Open Source UAS projects

• ▶ Reply

• ‹ Previous
• 1
• 2
• Next ›
• Page