Imex-Erik

I still think one of the best features is the on the fly inertia and momentum adjustments via potentiometers. I did not realize how much I would use this feature until I actually assigned them. The ability to change driving characteristics to any surface in ~2 seconds is amazing.
The applications are quite wide ranging for these components. The Scout I use almost daily for not only driving but motor testing as well and it stays cool in on 3S in the hot Florida heat with no additional cooling. I just want to remind people that this is an open source design AND code. Maybe the ESC and more will be combined later, the options are always there I personally like the modular approach as this is my favorite way to do things in RC. No need to buy or replace things that aren't broken and testing individual components is always easier than AIOs. Here soon I'll see what I can do about an install, lately I have been doing some streaming with some friends of mine on Twitch and if there is enough interest I might do some tank video streams and such. Once again I think you guys will be surprised at the ease of the PC app, go download it and take a look!

LukeZ

Erik is right - all the settings available in OP Config can be saved to a file on your computer (they are called .OPZ files and for you techies are actually just XML). You can do this with the application now to test, you don't need a TCB. OPZ files can then be shared and imported by anyone.

The only thing that will be unique to each user as Erik alluded to is their radio settings, unless I suppose people are using the same radios. So even if you import another user's OPZ file you will still want to perform the radio setup routine in OP Config so that the TCB knows the specifics of your particular radio. But this only needs to be done once.

You can watch a video of the radio setup routine on this page.

pedzola

I applaud this effort. Some of you may have seen my prior posts about creating an arduino tank controller. I didnt come up with anything nearly as full featured and elegant as this controller seems to be, but programming my own driving algorithm and setting up my own transmitter controls were infinitely rewarding. I was finally able to drive a tank the way I wanted to.

I am actually downsizing in the hobby and have a couple tanks for sale in the classifieds, but keeping my custom arduino and sabertooth-driven KV2. I'd like one of these new boards just to play around with, maybe will have to start a new project. Would be happy to contribute my driving algorithm to this project if it doesnt have a similar one already - mine simulated inside-track-braking type control with adjustable inertia. I was super happy with the scale realism of it. (Might not work with standard gearboxes tho, as it relied on a style of gearbox that would not roll unless power was applied).

Imex-Erik

I wanted to tell you SOO bad but I just couldn't and I apologize to those I couldn't! Oh and the Sabertooth works flawlessly as well via PWM or serial I've tested personally both the 2x5 and 2x12, but I like the Scout better IMO. The layout is very similar yet it is more user friendly to work with. Oh and who doesn't love purple?

superrod

My Feedback: (16)

I had a little time to do some reading on the site and really liked what I saw. The amount of information on setting up the board and the parts of the tank like the recoil is very well done. As soon as these are out I'm ready to try a couple. I am not that good at programming but I am sure I can find some help at work. Be cool to have a driving profile with the gear shifts programmed in.
Jimmy

LukeZ

Pedzola, I did see your thread when you first posted it and watched your video with interest. I like the simplicity of your approach and I mean that as a compliment. All coders know that simplicity is the greatest good and much harder to achieve than something which takes 800 lines and is a debugging nightmare.

One simple addition you could have done, and maybe already did, is map your steermargin variable to an extra analog channel on your radio for on-the-fly inertia adjustments. And if you had separate steermargin for forward and reverse then you would be a good deal of the way already to the functionality the TCB has.

I've used a slightly different time-based approach which is not nearly so simple as yours, but we have some other things going on that necessitate the extra complexity. The final result however is the same effect as you so effectively show, which is to say very fine control even at slow speeds and non-jerky movements. These driving algorithms were the very first thing I began with same as you, and I can say I know exactly what you mean about the great satisfaction it gave to finally be able to drive a model so smoothly and precisely.

I have not had the opportunity to use the ETO gearboxes but I can confirm that your algorithm would give surprisingly good results even with a cheap Heng Long gearbox, and yes even your drag brake implementation would also still work.

I have gotten to the point where I drive with very high levels of inertia and momentum because I like the smoothness, and also if the tank takes long enough to come to a stop one can actually use the "brake" to slow down, which is fun. The only drawback I found is that the greater the acceleration constraint, the less responsive the model would feel particularly when first starting to move from a stop, because of course it takes time for the speed to ramp up. Also because there is an amount of force required to start a gearbox moving that is greater than the amount of force required to keep it moving, so one tends to jam the throttle high at the beginning and then ease it off once the model gets going, but this results in a jerky movement which is what we're trying to avoid in the first place.

For that reason the TCB also has what I call the "nudge" option, which is a set level and duration of throttle (adjustable by the user) that will be applied when first moving from a stop, that overrides any acceleration constraint specified. When that amount of time is over (it can be as short as a small fraction of a second) then the user specified acceleration constraint comes back into effect. This lets you use a very high constraint during regular driving but also maintain some sense of responsiveness when moving from a stop because it is artificially "nudged" to get going.

There are many more adjustments that can be made to accommodate each person's driving style, including the option to drive your tank like a Heng Long maniac with full super spin. It's all in the Wiki, and I will try not to bore you with too much of it here. But feel free to explore and ask questions.

PS: Pedzolla if you want to read through how I approached the acceleration/deceleration constraints, check out the code in the OP_Driver class. The code is extensively commented, but whether any of my comments make sense or not I can't guarantee!

uglyduck

Great work! This is exciting news and the sort of thing I've been looking for. As Erik stated in a different thread I think its worth the wait.

On the sound board - maybe I'm trying to connect dots that can't be connected but Battle Armor has a DIY sound board mod for the RX-18 here : https://www.scribd.com/doc/277350656...een&from_embed . Could it be adapted to the TCB as a low cost alternative to the Benedini?

pedzola

Lukez I'm flattered that you looked at my code so closely. I did consider assigning the margin variable to a knob. I reviewed the class you linked and obviously you've put a lot more thought into it. Watching videos of real tanks they do jerk around a bit on quick starts/stops/shifting - someone could take what you've done here and make a very realistic model.

I am looking forward to getting the hardware. Will keep an eye out for further announcements.

Also, I came to really like the KV2 as a prototyping tank - Erik you should sell a kv lower half w/4:1 gearbox as a robotics platform.

heavyaslead

No computer programming required, now that has me interested.

What 'standard' set-up is configured 'from the factory' without computer tweaking?

Imex-Erik

For those interested in the Taranis there has been a large update to the FRSky line with the addition of a new Taranis model compatible with OpenTX and almost everything the normal Taranis has minus a few physical inputs. It runs OpenTX, has a module bay (JR), Sbus/PPM/PWM support with all current X and D series receivers, telemetry support, etc. I LOVE this new guy and with hall effect sensors now available for the Taranis X9D+ and these sticks looking like normal X9D+ sticks, it just might be possible to have an amazing TX with hall effect sensors, OpenTX, Sbus, and a rock solid radio system for under $150! Oh and self centering sticks on a Taranis take about 10m to install and are about $15 for a complete gimbal.
It might not be the best entry level TX, but WOW is it an amazing mid range TX. I will get one this year after I upgrade my gimbals first, I am very attached to my Taranis X9D+ as of now...

Imex-Erik

I was thinking the new Leopard lower chassis instead It is much stronger and will provide a much more stable platform for components to be mounted or an "upper hull" type body/cover. I have already thought of this and the lower hull will be offered in a few different variants. It will have the 4:1 and probably 390 motors too.
Well since it supports a wide array of stuff nothing is default out of the box. However as mentioned earlier a preset could be loaded and then you would only have to assign your sticks. People could setup a preset section for others and share configurations including audio later down the road. Stick assignment takes around 2-3m max (for 16ch) and is done 100% on the app, no need to adjust trim or endpoints on the radio.

YHR

I wish you good luck, and I believe open source is the way to go as it allows you to tailor how your RC device runs. In a way this is an advanced BARC. I will share my experiences with supporting the BARC. Not everyone knows which end of the soldering iron gets hot, and most can't be bothered with anything that takes anything over a 15 minute learning curve.

I compare this development to the development of the PC. IN the early days, it was the techno nerds buying and programming. This will be the same thing. You wont' have many customers that aren't already into programming and playing with electronics. However with the open concept and sharing of ideas, development should happen quickly, and the real prize coming out of this will be the software development.

I.e Consumer will buy the hardware, plug it in and then load up a "software program" that will configure this board for them. Whether it is tanks, subs, car, or planes.

The boards on the market today all leave something to be desired, but in the end it comes down to cost. People put up with stuff they can afford. So the hope is that development happens quickly, demand picks up where you can produce the boards for next to nothing and then you clean up selling easy to load "software" that configures the board for the consumer, with them having to do little more than plug in a cable.

For every good idea you will have a thousand people who will be critical and wish of its failure. However at the end of the day even the greatest ideas need to be marketed, and appeal to many. This is where the problem is in RC tanks and in Hobbies in general. There just isn't enough critical mass to ever allow a sustainable profit for the manufacturers So in the end the passion and the ideas die as soon as the principle is no longer interested . Tamiya is a big respected hobby company, and in hobby circles it is said the only reason there are 1/16 Tamiya tanks are because of the old man himself. When he moves on I doubt the line of 1/16 tanks will continue.

People move in an out of hobbies as their interests change. Dedicated tank controllers have a very limited customer base, and the base is saturated right now. Your idea to build an onboard PC crosses over to many hobbies and for that reason you should have a broader appeal with it. The big question is how many people want or need an onboard PC. Remember there was a time when people wondered why anyone would have a computer at home, and look at where we are today.

I am betting the improved sounds coming out of the New Heng Longs are going to be enough to keep most of this market happy, and that is the reason I pulled the plug. The people left over have their favorites, and are not really interested in anything else. They learn about one and once they understand it they are loyal regardless of what comes out that is better, because frankly most don't want to spend any more time then they have to "learn" anything.

Anyway I don't want to come across as one of the thousand, but I do want to point out a few things. Broaden the base of applications for this as quick as you can so the RC tank market is only a small niche of your customer base. The upside of this system is that its development and what it can be used for may not even be on the radar yet.

I recently added slot car sounds to 1/32 scale slot cars using the TBS, Not many Clark users have that ability. Point being, make the appeal as broad as possible, and don't rely on the tank market, Also think outside the box with this. With this open concept and many people working together, the ideas of how it can be used is only just beginning. For this I am hopeful

Edit

I just remembered that beautiful 1/35 tank done using a raspberry. The hull rocked back and forth, and the action was way more convincing then any 1/16 RC tank I have seen. That will be the beauty of this system. An intense few will develop software code, which can then be sold to the masses as a download.

Cheers

Dan

LukeZ

Yes, those boards could be adapted to the TCB in essentially the same way they are shown adapted to the RX18. One would need a great tolerance for poor hygiene and my personal view is the effort would really not be worth the result. I had mentioned earlier that I tried to make this project as easy to hack as possible, but I really meant that in a firmware sense. This mod is what I consider a hardware hack and while for some those are perhaps more intuitive than learning a software language, the risks of frying your board and the complexity that ensues are rarely worth the minor benefit. My view is generally that if the hardware needs to be hacked, then we need new hardware.

The TCB has extra serial ports and other outputs, a better approach would be to create a simple hardware interface that takes serial commands from the TCB and then controls an off the shelf sound unit. The hardware can be very simple and very cheap, the wiring a single cable, and all the work can be done in software. This is the approach I am taking in current development and of course others are free to undertake similar endeavors, and we will see who crosses the line first! Sound will be an option in time, but of course time is always the thing in short supply. For that reason I make no predictions as to the future, and for the moment Benedini remains the only option.

Or you can run with no sounds, as I often do. I would rather have no sound than a headache, which is what most of them today give me.

LukeZ

I will have to give the KV chassis a try! It is one I have been eyeing for some time, but then again, I pretty much have been eyeing every tank ever made and would buy them all if I had the money, space and time!

Speaking of test beds, I thought it might be interesting to show my own modest version, and perhaps sing its praises briefly. This is a bone stock Heng Long Panzer III chassis, all plastic running gear, that I obtained from Phil Pflueger seven years ago!

This thing has been a workhorse beyond compare, and until just last week has never given me any problems (one of the zinc gears has finally shattered). Who knows how many hundreds of miles this thing has run in circles around the living room and back yard during infinite testing. I've never thrown a track, had a wheel come off, or any other issues whatsoever except an occasional track pin come loose.

The first photo below shows it circa 2011 with an Arduino Mega and various shields, the second picture was taken today.

The biggest drawback to this guy is the lack of space. With the gearbox failure I realized it was long past time for an upgrade, and I am planning to try a Taigen Tiger 1 chassis with all plastic running gear. Although I am a complete sucker for all the metal bits on my actual models same as anyone else, for extended running on carpet as well as simply to avoid breaking my wrists every time I pick it up, plastic and light weight work best for this application.

LukeZ

As Erik mentioned it doesn't really make sense to throw the board in a model without at least connecting to the computer once. It's true all settings start at default values but these will almost inevitably be incorrect for any given individual. There are many optional settings you can choose to enable or not (like acceleration and such), and all that can be left alone. But you also need to tell the TCB what hardware you have - it needs to know, are you running airsoft or mechanical recoil? Are you driving the tracks with a Sabertooth controller or a hobby ESC or a continuous rotation servo or some other? Hardware selections such as these are the big ones that need to be specified. Of course it is just selecting a value from a drop-down box, so I really don't think it is too odious a chore. And in addition to motor selection one absolutely also needs to run through the Radio Setup routine once.

The point I was making is that once you have the model programmed correctly, you then do not need a computer in the field to perform the same selections that Tamiya can do in the field, which is to say, change your weight class.

LukeZ

YHR (Dan), I am really happy to have your comments. I have followed the development of Dave's DBC and your BARC from the beginning and owe much of my inspiration to you both. Dave's "General Notes" on the RX-18 and his research into the technical details of the Heng-Long and Tamiya IR systems were of great help to me. You two are the Original RC tank hackers for sure. Even now the BARC remains an excellent system and is probably almost as close as one can get to the plug-and-play that Eric prefers while offering advanced features. But I also understand why you are no longer selling it.

My project originated because I set out to make the system I wished existed, for my own use, with the only purpose to please myself and teach myself things I was interested in. That I have achieved.

Beyond that, I must admit I have seriously grappled with whether to release this to the public at all. It is important for me to point out that no one is, or ever has, paid me for my time, all of it has strictly been for the "fun of it" - not that much of it hasn't felt suspiciously like work at times. I do have a full time job and sadly it has nothing to do with tanks. Taking Open Panzer public will mean responding to support requests, fixing bugs, writing new features, creating documentation, maintaining the website and a million other time-consuming things for which I will not be compensated.

Even when someone buys a TCB from Hobby King, they are only paying Hobby King for their cost to produce the hardware. Not that I am ungrateful, in fact quite the opposite - Hobby King is also fronting entirely the very substantial amount of money required to produce a run of boards; in other words they are assuming all the financial risks if it doesn't sell. This is all very generous of them, though I guess they are big enough they can afford to take some risks every now and then. They are also taking care of inventory and shipping which saves me a lot of work. Finally they are also offering to donate 5% of proceeds back to the project, which they are not obligated to do. But I think when you see the final price compared to the raw cost of components (you can tally them yourself as the bill of materials is posted on our site), it will be clear even they will be making an insignificant profit, and the proceeds back to this project will be lucky to cover web hosting fees, let alone anyone's salary.

The very real possibility of burn-out does not escape me, and those of us in the hobby long enough have seen it several times already even amongst those who ostensibly were making money from their efforts.

So why bother at all? Ironically I think the open source, non-profit nature of this project may just allow it to succeed - or at least survive - where others fall by the wayside. Chasing profit in such a small market segment is a fool's errand to begin with. So let's dispense with that from the start. Instead I hope we can build on, and expand, the greatest asset already present in this small market segment - the community.

Dan whereas if you tap out modelers lose the BARC, which is a step backwards for us all; if I burn out, go bankrupt, or decide to leave the hobby to take up underwater basket weaving instead, nothing is lost. The work done so far can't be erased or taken away from public ownership.

Dan you are absolutely right to say the real prize from this project is the software development. There is nothing special about the TCB board itself, which anyone can see who examines the schematic. Anyone can also inspect the Benedini sound card with their two eyes and make out his circuit as well. You can clone it easily enough, which the Chinese have already done. But without his firmware, the device is useless. And only he possesses it.

The Open Panzer firmware belongs now to all of you, whether you know what to do with it or not, whether you care or not. A great deal of work has already been done, though there is a lot that can still be added. The good thing is that as new people arrive to add their contribution, they don't have to start at square one, recreating driving algorithms, decoding every IR protocol under the sun, writing serial communication protocols for Windows and all the other tedious baloney.

Furthermore, even if Hobby King decides the TCB is a waste of time, we are still not defeated. The software can easily be adapted to other hardware. If you really want a board with ESCs built in, you can build it. All the information is present right now on our site to do that. Raising the money to fund production is always a challenge with open source projects, but there are always avenues like Kickstarter. And if through refinement and experimentation the community arrives at a design popular enough to succeed commercially, places like Hobby King will always be happy to make a dollar from it.

And if we are insanely successful, the Chinese will clone our design and sell it for pennies at Banggood. Then we will have reached nirvana.

My wish is that in time this project is no longer intimately associated with me personally. I hope others will become involved and that we can all build together our better vision for RC tank electronics: not for the financial benefit of some company, nor with the restrictions imposed on us by the accounting department of the same, but for the satisfaction and betterment of us, the modelers and enthusiasts.

This process will take many years if it happens at all. The whole thing is a huge gamble, but I am glad to be here with you all for the ride. I know I'm in good company (grumpy as some of you may like to seem).



Luke

Afrikakorps

Sounds interesting. I would like to see it in person or know from someone who has used the RC tanks Australia ASP2 what they think of this. The ASP2 combines combines almost all the features of Elmod, IBU2+3, HL and Tamiya, not to mention dual sound cards, so it is an impressive and complete Multi Function Unit.
Although it is marketed online and in Australia, I think Rad_Schuhart has tried the Asp2.
I would like to know your thoughts.
Cheers
Afrikakorps

Imex-Erik

One of the many advantages of Open Source. Bear with me guys here for a bit (I fly and tank)....
I remember a few years ago when Cleanflight (open source flight controller) first started. People took their time with it, but after a few features were added things started snowballing and now has many forks and is used in almost all multirotors that are not APM or DJI controlled. Betaflight (a fork of CF) is now so popular it has over 2800 pages of comments on RCG.... pages, not posts. I personally couldn't imagine running anything else but BF on anything that flies anymore, I trust it more than the closed source alternatives out there that are a lot more expensive.
I think this project will snowball like that. I think once a few things have been worked out and the public has some in their hands they can see how enjoyable a fully configurable system can be. If you only want the general stick commands you are used to then np! If you want to spend 5-10m and assign tons of commands and such then np as well. Options are my favorite. Also, neither the software or hardware is complete! Always evolving and improving is the goal.
Do not get me wrong, if you want a plug and play this is probably not for you, but if you don't mind spending a few minutes working with it you might end up with something very amazing for the price. Considering the hours and hours we put into mods, I would think a few minutes on a self updating Windows app wouldn't be too bad. Utilizing advanced radio systems such as PPM, iBus, and Sbus are things I have asked about for a long time. 1 wire = 8ch on PPM, 14ch on iBus, and 16ch on Sbus. With almost all modern radios compatible with PPM at the least there is no reason we shouldn't use those channels. You also get increased control, less latency, and a cleaner signal. I have setup my TX to use all 16ch, I have my Taranis announce everything to me, from neutral drops to firing (with a countdown to fire again) and even lights it is quite neat.

ksoc

This looks great!. - Can the BlueTooth option to be used at a battlefield to confirm a tank has the correct setting?
- Is there the possibility of IFF / Tracking of friendly fire? Always nice to know if you worst enemy is on your own team

Rad_Schuhart

Yes, you are right, The ASP2 has everything that all the other boards have together and much, much more. I indeed have two, and I want to buy another two. Is the best board out there and people that say it is not is because they have not tried it, or they sell another one from the compertence.

My opinion about this new board is what I said at the beginning, and I agree with our colegue that said "KISS" (Keep it simple stupid) Heading back to the ASP I have seen plenty of times that people just wont buy it and preffer to buy another board with waaaay less functions and more expensive just because they think they are easier to play with when sometimes it is not.

Sometimes rewiring even one cable is enough to keep people away from one board, that why I dont see very clear the new open source board. Im not sure how much will this new board cost, but a benedini is about 100 euros and it plays low quality and short lengt sounds (I think something about 93 seconds maximum, all together!) meanwhile with the ASP I can play for literally days at CD quality just for a bit more money. If I need to spend 100 euros for the benedini, plus the ESCs, plus this new board it will be more expensive, I wont have enough room inside my tanks and when I open it will look like and spaguetti dish. Not counting people will have countless problems. (For example I've read plenty of threads about people triying to make their hobby grade boards work with the stock Heng Long christals controlers, so you can imagine...)

In my opinion (and again, I might be wrong) if this new board wants to succeed it should at least be able to play a lot of sounds with good quality, getting rid of another extra expensive sound board. And of course they should try to find their place in other hobbys, not just RC tanks.

Again, I think that the "KISS" is what most people wants. I must say I love the look of the IBU3 board, is very clean and you see where to connect everything.


Sorry, I dont want to put sticks on the wheels, is just my opinion as an RC tank user and plenty of forums in diferent languages member. The open source is an amazing idea and it might be the way to go in the future, but the board should be more or less plug and play...

ausf

Open source is absolutely the future, probably of everything. It makes sense, many hands get involved and make the product better. Look no further than Elon Musk for evidence of that. The days of sitting in your castle trying to defend the secrecy of design is over, too much information is the palm of everyone's hand.

In terms of adapting, I'm 50 plus years old, have been tinkering with RC since the 80s, can figure out some intricate stuff for animation, etc, using servos, stepper motors, gearing and linkages...That said, I started up with Arduino and took my GenX approach, bought a bunch of books, went through the learning process. My 15 and 17 year olds came along, did whatever the hell they do and in an hour surpass my progress with a few things they find on their iPhones.

Put this thread together with Brett's regarding the state of the hobby and there's your future, the next generation will be all over this, us dinosaurs either adapt or hoard Tammy MFUs and Rx-18 like canned food after the zombies come.

And as far as underwater basket weaving, I've done plenty of underwater pumpkin carving and that's a mess. Basket weaving at 30 feet sounds like a dream (as long as the water is warm enough to leave off the gloves).

LukeZ

Pedzola, I just remembered from your YouTube video that you are using an Arduino Mega in your experimental tank. The TCB uses the same processor as the Mega, so the Mega can run the TCB firmware, although of course not all functionality will be available. But with a few jumpers you can easily control a Sabertooth and Benedini. OP Config will be fully functional as well as all the driving adjustments we just discussed.

I posted a diagram in a thread on the Open Panzer forum if you, or anyone else, is interested in using a Mega to experiment prior to TCB boards being available.

uglyduck

Thanks Luke! Just so happens I have a spare Mega sitting around.

Have you looked at Squirlier's code for his Stabiwii barrel stabilization? My rudimentary understanding of Arduino programming doesn't reveal if its useful for the TCB or not.

https://drive.google.com/drive/u/0/f...lJOWGRRZGJoSG8

LukeZ

Yes it could, although one drawback is that every new bluetooth device your computer detects adds another COM port and Windows goes through a detection process, so if for example you were thinking there would be a referee with a laptop at a public battle, and he would sequentially try to connect to every guy's tank to check their settings, it could get to be a lengthy and tedious process. This is just a reality of the way bluetooth and Windows works, not anything to do with the TCB specifically.

But there is no reason the TCB can't pair over bluetooth with your personal cell phone. All we need is an app developer! This will probably ultimately be the better approach since everyone has a phone already.

Yes absolutely. You can battle with team based play already by having all participants select the FOV protocol in OP Config. But in time I believe it will be necessary for us to construct an Open Panzer Battle Protocol, but I have not yet implemented such a thing yet because I believe the community needs first to decide exactly what this standard should be, what should be included and what should not, etc... It would also not hurt to get the input of the other manufacturers such as are willing to be involved. I could have invented one myself but then it would just be one more among the many that already exist, and there would be no guarantee it would be any more popular than what is already out there. That is why I believe this particular topic needs to be addressed collectively if we really intend to create something better and with wide adoption.

The one thing to keep in mind - as soon as the protocol is changed at all, it is no longer Tamiya compliant. In my opinion this holds true even if you fire the original Tamiya signal but append an IFF tag to the end (as some RCTA devices do). There are common sense as well as technical reasons why this would be so. This inevitably means any protocol that is not Tamiya will require all battle participants to use electronics that are not Tamiya if it is to be considered fair. I think it will be a long time, if ever, that the TCB or any other board become as popular and prevalent as the basic Tamiya system, and for that reason I see the Tamiya standard remaining with us for a very long time. And I don't think that's a bad thing.

LukeZ

I took a look at the link you provided but most of it appears to be in German and if there is useful code there it eludes me, perhaps due to the odd organization of the site. That is not to say it couldn't prove worthwhile, perhaps this Squirlier fellow would want to become involved with our project. Are you acquainted with him or any English-speaking forum he may be active on?

If not I did find his YouTube page and I can try to contact him there.

I actually do have barrel stabilization working with an Adafruit BNO055 IMU; what is not working is the I2C library used to communicate with it. It works, but conflicts with other processes on the TCB because the standard I2C library has blocking calls. My attempts to create a non-blocking I2C library have so far not succeeded, but the source is listed in the GitHub repo if anyone wants to take a crack at it.

The BNO055 can also communicate over serial so another option is to re-write the sensor comms to use the TCB's Serial 1 port. This has not yet risen to the top of my priority list. I am also wondering if a less expensive IMU, paired with a small processor on a small board that simply plugs inline with the servo may not ultimately be a better approach.

This is a good example of an area of development we can use help with. A further list of To-Dos can be found here, and I expect it to grow significantly once people start using the product.

If you are a coder and want a project to work on where all the hard things haven't yet been solved, this is a good project and a good time to get involved.