Mission Planner "Makeover"

Ground Control Software Mission Planner
67

Oh I’m happy with the heat. I personally don’t have any skin in the game, but others do, especially new people, and I think the project does too. I’ve dealt with MP for years now, and I know where just about everything is… unless it’s hidden like most of the advanced features. The simple features that I use are fairly accessible, even if the logic to get to them doesn’t make sense. Most advanced programs with tons of features start to make more and more sense as you continue to use them. Unfortunately I don’t think this is the case with MP - the logic doesn’t flow in my mind, and I agree with the following:

On the topic of me doing it myself (with no Visual Studio, C++, or Windows API), I don’t think that’s ever going to happen. Right from the developer wiki, and the complexity is a topic all on it’s own:

Mission Planner is a very complex including and making changes is not for the faint of heart. Here are the basic skills you will need to make reasonable progress with MP changes:

  • C# programming skills and experience (at least or C++ experience).
  • Experience with Microsoft Visual Studio (VS) development environment. MP is not the application to begin learning VS.
  • Experience using Windows API (Application programming Interface) - including understanding of streams, processes, threads.
68

Where are the “numerous users above”? This thread has been alive for almost a year with 5 or 6 people posting in the affirmative. As you said in your post a couple days ago that resurrected this thread from the dead “Overall, there has been little attention given to this topic”. Right.

Submit your desired enhancements as PR’s and wait and see if they are taken up.

1 Like
69

https://www.facebook.com/groups/ArduPilot.org/permalink/1701371449886041

70

There have been 2.8K views, not bad, actually one of the most viewed in the all forum. Maybe people don’t feel like writing anything because of the “everything is already perfect” mood that is around here.

1 Like
71

Yes a lot of views but it might simply be due to the Reality Show title. Maybe if it was titled “Mission Planner Extreme Makeover” it would generate even more views. But about the same response I would guess.

72

I guess you couldn’t work out how to cancel your account then?

I there is by no means a everything is perfect attitude, everyone agrees there is room for improvement. I think it is simply that devs have more interesting things to do. If we forced all the devs to do things they weren’t interested in for free there would be no devs and you would still be using your flight controller to play wii sports.

1 Like
73

I for one have just decide to stop using Vector due to lack of features that are available in M.P.
It may take some time to learn & work thru however worth it.

1 Like
74

Yes, didn’t figure out how to, probably there was no interest in people having me cancel account. It is really miserable behaviour from you.

I see you take it personally, like everybody else when someone comes up with a critic on something to make it better.

That is the way it works around here, it goes down to personal attack whenever people have some critics to make.

75

I guess without thousands of users feedback even devs would look at their great piece of software in front of their pc. So, as asked before, please some respect for all the people that tested stuff countless hours at the field, like we all have for devs.

I tested a bunch of things that i was not interested in, just because i wanted to see if there was any bug for the community, so i guess devs can do the same from time to time, no?

76

That was harsh of me, I apologise.

If your staying maybe you could un-edit your posts here, it benefits no one. (and annoyed me more than perhaps it should have)

Your are of-course right testing is just as important as developing, it allows the latest features to reach the whole community.

I can only speak for me but for me if I’m not interested in something it feels too much like proper work. I struggle to get motivated and either never finish the job or finish it to a poor standard.

Maybe the way to go with this makeover is to do some ‘story boards’? To show how mission planner could look give a clear end goal. Someone could collect all the issues together into a single issue to give a overview of what needs doing and how far through it is. Then it could be more easily broken down into more manageable jobs.

Maybe part of the problem in this case is just than Autopilot moves forward so fast, the mission planner guys are working hard just to keep up with all the new features.

2 Likes
77

Hi its a shame that things just dont happen there at eagle tree just so slow to do any thing and if there is a bug can be a while before it is fixed.

78

I think it is you that takes it personally the way you approach this,yes we are all beta testers no matter how we go about it,I am playing with some thing now as it keeps me busy and happy and I enjoy it.

For me all the developers do a great job,yes there are faults with the system but O my god its 99.9% good,Mission Planner could do with a tidy up but personally I would see new feature’s appear,but that is just me.

2 Likes
79

I’m nobody here, but I’ll proffer this:

Two times I wrote Michael by email, he replied in a reasonable amount of time, and fixed/added what I had asked of him.

One time, he didn’t. I was a bit sad because I thought that last one was a great feature that would benefit everyone, most especially me for flight planning, but such is life.

By numbers alone, most of the time Michael is responsive and willing/able to fix up MP when asked. I don’t begrudge him the some of the times he’s too busy, or possibly CBF to listen to an unsolicited ask.

I can’t code worth shit, I’m not smart enough to help with anything on the EKF/algorithm side, and I don’t know enough about the projects to document it. I have to work two jobs/7 days a week to stay afloat so there isn’t disposable income to fund things.

So what can I contribute? Like most above, I can test. I can think critically. I can make thoughtful and polite requests/suggestions.

What more can we do, as a community of end-users, to help contribute if we see a need? Perhaps we need to establish a community steering process where we consolidate, clean, order/rank, and propose changes/fixes such that they’re easily looked at and digested. What we likely lack is the knowledge of what is feasible, but it might be better than scatter-shot requests spread about in a forum.

ArduPilot has a great structure within the development team for this type of task-setting and vetting, and I think that mirroring that somehow within the community for requesting things might be beneficial to both sides. And yes, likely, community crowd-funding of certain things may even be necessary to sponsor big changes.

Or, that’s the best I can offer at the moment.

We’re all friends here, though at times things can get heated. That’s what happens when people are passionate/enthusiastic about things. Golden rule always applies: How would you like to be treated in the other person’s shoes?

7 Likes
80

Totally off-topic, but I wanted to throw in my 2-cents

I tried to get into helping out with documentation, because I’m not a prolific programmer (also I have very little time to devote to writing code), but more importantly, I have very little time (or desire) to learn the intricacies of git (or to setup an environment to do all the git work) so I kinda gave on on wiki updates after several of my PRs didn’t look that good after being merged. The whole site is difficult for me to update, so I gave up.

A nice WYSIWYG wiki page editor built into the docs/dev site would REALLY be helpful to me…and perhaps would lend itself to more people being active at adding/updating content.

That said, the wiki dev is in the same boat at MP. “Someone” would have to take on the task to completely re-design the wiki page, and so I recognize that it isn’t going to happen.

While I agree that this community is probably served best by politely and caringly taking the honest input of the community, I also see a 1 year long running thread that is basically jut arguing that no one will do what a few people want done. Personally, I think re-writing MP into cross-platform (like Java) or something makes TOTAL sense…but it isn’t something that 1 person in his spare time can handle, and isn’t something worth arguing over, considering the thousands of people who use MP regularly like it is. This thread is simply just a lot of bickering.

4 Likes
81

I think this would be an important improvement to the Ardupilot community. Unfortunately I think the nature of the beast does not lend itself. I think the people who have spent the effort to learn Ardupilot have a significant investment in the community even if they have never “given back”. I think because of this investment, as well as additional time spend helping others, testing code and hardware etc, should justify a “user advisory panel”.

The devs are smart capable people but they live in their own world. GIT is probably a terrific programmer tool but it creates a block for casual user contributions that is significantly reducing the effectiveness of a community with lots of volunteers contributing. As lordneeko pointed out the documentation editing system is barbaric. Something a programmer would come up with.

But to change these things would require a major contribution by a “civilian” to create a better documentation system and to manage a fair but organized user feedback system. Which would not be a minor effort to just bereferee to a large group of users with wildly differing needs. As an example I have zero interest in a change to Mission Planner. I really like it just like it is. But I would really like and need UAVCVAN/UC4H support. I have zero interest in racing drones and think it is a waste of dev time to play in that sandbox. Everyone’s needs are different. Who is going to run the "House Sub-committee on drone enhancements"and advance certain ones to the floor.

Then no matter what users want they may not get it because the resources, both money and people, along with some self interest may get in the way. The development process is, and always has been, influenced by moneyed interests. 3DR contributed a lot in the beginning to make what Chris envisioned happen. So each dev is driven by what they are interested in, what they can get paid to do and then some vague sense of what the whole group of devs think. There are politics internally, as there are in every group of humans, with conflicting interests.

But I personally believe that the commercial world is leaving Ardupilot in the dust because Ardupilot is so focused on porting flight controller code that the usefulness of the eco-system is being reduced. I just picked up a Parrot anafi which at 320g can fly 25minutes in 30 mph winds. HD video to the smartphone on the ground, 4K recording, 180deg pitch gimbal with Mission control and simultaneous rc controller override during missions. Optical/gps follow me on a 30mph vehicle. 3x zoom camera. It can automatically take and stitch 360deg panoramas etc etc

I could not build an Ardupilot based aircraft with anything close to those specs at anything close to the price. A lot of the features would not be possible at all.

My two cents

3 Likes
82

I think Ardupilot needs to understand what it wants to be, than focus and develop in that direction.

Now it looks like there is no path to follow.

83

I like your idea a lot. I think this would go a long way toward giving the community a voice.

I also see your point here for commercial multirotors. Fixed-wing and VTOL aircraft aren’t in that situation though. There is not one commercially equivalent system that can beat arduplane in my opinion. Parrot is as close as it gets, and their equipment is very closed, expensive, and not adaptable. Maybe PICCOLO, but it’s so expensive.

There are so many AMA folks and similar that fly totally manual RC airplanes where a cheaper flight controller will expand their horizons with what is possible, so I understand and value the porting to more boards. I think the number 1 way to expand to the largest community (recreational flyers) is to make things easier/simpler for them while still being inexpensive. I want to show that community what is possible, and that’s why I think a more user-friendly Mission Planner should be a priority. Most of us that have stuck it through can deal with MP without changing, however I fear that we will lose a lot of people who turn away because of what they view as complexity. I almost feel embarrassed to point new users to mission planner because it doesn’t follow the same logic as most computer programs.

For example, when you open a text editor program, you probably have a few expectations. On the top will be a ribbon or drop-downs with options for text formatting, etc. On the bottom might be a few buttons like zoom, viewing options, etc. The left and right are generally open or where you might find a toolbar or two. MS Word, WordPad, Notepad, OpenOffice all share this generic layout.

Unfortunately Mission Planner doesn’t follow the same generic layout as any other program that I’ve seen. There’s main tabs on top (I can generally follow that), but some tabs switch to have additional sub-menus on the left (Config/Tuning). Some have sub-menus/tabs on a second window (flight data). Some are on the right and bottom (flight planning). Very useful features are in a menu that’s only accessible by Ctrl+F. The main tabs seem to be flight-stage oriented, but post-flight tools are under the VFR HUD. etc, etc, etc. The “flight data overflow” of Mission Planner cited by Intel’s report on GCS’s is something that won’t be helping our new friends.

On a high level, I think Ardupilot should be looking to expand its community to everyone. If I’m misinterpreting the developer mission statement ArduPilot aims to enable the creation and use of trusted, autonomous, unmanned vehicle systems for the peaceful benefit of all., please let me know. To me, Mission Planner is only catering/focusing development on what advanced users need. That’s not “all” to me.

If Mission Planner’s intent is only to be an advanced GCS for advanced users, I don’t see how that falls in-line with Ardupilot’s developer mission statement or the Versatile, Trusted, Open slogan.

1 Like
84

Maybe we should have a community vote every month or two on issues to be fixed. Could have a ‘top 5’ user requested issues to be fixed in the code, mission planner the wiki, whatever. This would give devs things to work on if something takes there fancy. The autopilot foundation could then put up a code bounty (of some small amount so its sustainable) to get the top issue fixed.

85

Maybe there could be a list of topics and the possibility for the users to help with donations, than a dev could pick it up and get the amount once done. Like a list of topics with the total amount contributed to that time, so a dev could decide to dev that point.

Corrado

2 Likes
86

That’s not really a bad idea.
Right now, there is just the “Contribution of the Month” prize…which seems to be a bit hit or miss, but I think the devs tagging Issues with “Rewards” might get people (maybe even companies) working on some of them. I know people who make a great “second job” living on doing bug bounties. But what we have here is simply “if you contribute something awesome we might remember to vote on giving you a pat on the back”