FRC 7125 Tigerbotics 2023 Build Thread

[DAflamingFOX]

Howdy!

We are super excited to officially join #openalliance this year! ( We technically joined at the end of last year so only posted on discord. ) As a team it has always been our standard to be open with our findings and development, mostly by sharing on Facebook. By joining the #openalliance we hope to more formally share our process, while also working together with other teams to be more successful.

Somewhat unfortunately, Tigerbotics is not always the most “successful” team, due to many factors such as: being student lead (mostly), rural small school, and time available to students; we sometimes struggle with organization, funding, and general effort. Not that this can’t be overcome, nor does it stem entirely from the symptoms above, but it is definitely something we currently struggle with. Due to this Tigerbotics does not necessarily strive to be the best team, but we simply strive to go above and beyond our reasonable goals, do better than we previously have, and most importantly have fun. If you’re not having fun, why do it.

With that here are our goals for this year:

• Attend DCMP again
• Be in an alliance at every event
  • Bonus: To be 1st pick or even captain
• Keep all members productive
• Have fun

We will be attending the following competitions:

• Week 1 - Waco
• Week 5 - Amarillo

Important Links:

• Socials
  • Facebook
  • YouTube
  • Instagram
  • Twitter
• Open Source Info
  • GitHub - FRC2023, TigerLib
  • OnShape (We are still learning OnShape bear with us )

Thanks for reading our spiel.
- Jeff M - Co-Captain

P.S. - Expect our build thread to be as unorganized as our shop.

[DAflamingFOX]

Update 0: “Drivers behind the line in 3, 2, 1…”

As a preface to this pre-kickoff update about what we have been doing… pre kickoff… I would like to summarize it in this convenient meme:

Anyway… no, we haven’t done all that much, but we have been busy.

General:

The majority of the time has been spent cleaning our shop, it has gotten so much better, while its still pretty un organized outside of the closet, I would say 90% of our tools and parts are in the storage closet and dealt with. The sheer volume of stuff we removed from the shop is astounding and while it doesn’t look that much different, it is way cleaner.

Storage Room Image

Marketing:

Marketing has been very busy designing a multitude of things that we hope to have, it is unlikely that all of these will be viable to produce, but the current ideas are as follows:

• Buttons
  • Personalized Team Buttons
• Banners
• T-Shirt
  • Hockey Jersey
• Stickers
• Flag

Marketing Design Images

We have also created our own custom colored version of the OA logo for use on printed products and electronic media:

Customized OA Logos

Mechanical / Build:

The build team has started the process of building our kitbot, Gretzky, to be used for defense practice against our real bot, and also a test platform for prototypes.

Gretzky Images

Somewhat lacking in the image department here, but as you can see Gretzky had eyes for a little while while programming was testing that the vision was working.

Programming

A reminder that Tigerlib & FRC 2023 are linked in the first post.
Updates to Tigerlib for 2023:

• Custom Trigger & TriggerGroup Implementation.
• Java 17.
• Removing sub-optimal practices and useless code.

FRC 2023 updates:
As we don’t know the game yet there isn’t that much we can do, but we have programmed a drivetrain and vision system in preparation for the game.
This year we hope to use a new control scheme, it could use a better name, but for now its called “face angle”. It does exactly what it sounds like, you point your turning joystick to the left and the robot faces left. You can see how it works here, and video of it working in sim here.

general programming:
We have began working with a RasPi & SnakeEyes hat running PhotonVision, and we will run it for AprilTags or retroreflective vision, which ever is more useful.
Feel free to use our AprilTag16h5 repo btw, there are pdfs for every tag or combined, all labeled and resized.

[kinahawi]

Interesting idea with “face angle”. Is that operating on a differential drive robot?

[DAflamingFOX]

We use mecanum drive, although the way it is design it would be pretty easy to port over to diff drive. Basically we just have our heading controlled by a PID controller and constantly feed it a desired heading, whether its derived by joystick inputs or a target on the field is up to the code to decide.

[DAflamingFOX]

Also thanks to openai, I’m gonna rename those enums to something less confusing now lol.

[kinahawi]

So what the major difference between this and field relative control? That the “yaw” control is position with your approach while it’s more of a “yaw rate” with field relative?

I agree the code for differential would be pretty easy, but I’m curious if you have experience driving it that way. Seems like it might be a nice way to simplify the mental transform required when the bot is facing you vs facing away from you.

[DAflamingFOX]

A typical field oriented approach only constrains your robots translational axis (x: forwards, y:left) to the field and your controller; forwards from you/field/robot is +x, left of you/field/robot is +y. By constraining the rotation, or yaw, you effectively constrain the yaw as well, pointing north on joystick makes robot face north, pointing east on joystick makes robot face east.

We had very minimal experience with it last year when setting up a PID controller on our heading, but never went with it as it was new and foreign, while we haven’t had actual robot time with it yet, the simulator has been very nice, and it seems to be very easy to drive and just click to those on our team who tried it.

Yes, the idea is to make it as simple as possible.

[DAflamingFOX]

Update 1: “Tell that to my screwdriver”

Game Analysis:

After the kickoff we immediately began reading in depth the rules and deciding on strategy and how we should go about scoring. We decided that due to the simplistic nature of the game we hope to be able to cover all tasks for once. Our design goals for this year are as follows:

• Score high & mid goals, both cones and cubes
• Be able to pick up cubes and upright cones from ground & double substation
• Be able to both dock & engage the charging station by driving or other means.

Kickoff Photos

Build / Mechanical

While we wait on our kit to be delivered, we have been looking at past years, watching Ri3D, and the other openalliance teams test the game pieces and we have pretty much settled on one design, while it might change, we hope to keep it pretty final.

Our design is heavily influenced by 254’s 2019 robot, and the general 2018 intake design; also due to our design constraint of only picking up upright cones, we will probably go with one of 7461’s tests, in which the cone is squished between two rollers and lifted by them.

We have not began prototyping yet, but will start shortly; as soon as we receive out kit.

Design Photos

Software / Programming

We have began the long process of trying to preemptively program and automate subsystems.

Most notably, we have on-the-fly path generation working, and should be able to allow the driver to hold a button and have the robot drive itself and align with zones and substations.

Programming Photos

OTF Path Generation

[robean]

Update 2: “Why is it doing exactly what I told it to do?!”

Heyo! I’m not jeff (obviously) I’m Robyn (or Maddison) and I’ll be doing the update this week!! I will try my best to sound like I know what I’m talking about.

General:

Not much of an update this week in any specific category BUT we were able to register for the FIT District Waco and Amarillo event :D!! (We switched Belton with Amarillo)

Software / Programming:

@DAflamingFOX “has been very busy working out many bugs that never came through during sim”. [Round of applause]

Build / Mechanical:

Our intake (prototype) progress has been going well and field elements are being built using the shop scrap; we finished building the substation but need to get more materials for the charging station and grid.

Intake

Dual Substation

Bonus:

Bumper Bump Test

[DAflamingFOX]

I believe we plan on doing more extensive bumper - bumper interaction tests with the two bots btw.

[DAflamingFOX]

Update 3: “7125 is doing strategical donuts”

Sorry in advanced for being so late this week, and apparently I just forgot to take pictures so were kinda lacking in that department.

General

Overall we finally got our kickoff kit and FIRST choice packages. We have are starting to finish purchase lists for our designs and getting those items purchased so we can begin building.

Build / Mechanical

We tested our claw prototype with the cube and code and we decided that it was good to go and we would purchase parts for it, and will finish up the cad soon. (I forgot pictures, oops).

We’ve continued building our field elements and got the april tag holders made with polycarb and vinyl.

Electrical

We got all of the REV control system parts (PDH, PH, RPM) and got those switched out on our bot. The REV parts are so much nicer, but we’re a little bummed that they dont accept ferrules.

We (its really my fault I assembled it) learned how strong vhb tape really is after it gets fully stuck. I’ll just say we won’t be doing that again.

(Mildly unsafe)
Removing the network switch from a mounting plate

Marketing

The marketing team has made many designs for jerseys and stickers / buttons.

Here are currently the highest voted stickers:

Software / Programming

We got our drivetrain to translate on closed loop pretty well, still working out some rotation bugs, as its controlled by PID.
We learned that feed forward is pretty much a requirement for closed loop control on the drivetrain.

We worked out some vendordep bugs in Tigerlib so now it will work fully without requiring vendordeps in the project its used in.

Currently working on finishing up drivetrain code, and starting on elevator / arm code.

---

This week has been hectic getting alot of stuff in just now. Hopefully we can start getting alot done and start building the actual robot soon.

[DAflamingFOX]

We were all stuck at home this week due to the weather, so no update this week T-T.

[DAflamingFOX]

Update 4: “Electronics run on smoke”

General

Thanks to lovely Texas weather, we lost an entire week of time due to school being canceled. We got in a pair of drills (impact & standard) which have been much needed.

Build / Mechanical

We are at the point where we are grinding out CAD as fast as we can so that we can get things ordered and built, we are super behind, but hopefully we are able to get something scrounged together for Waco.

This CAD is extremely in complete and almost just a placeholder for more mature CAD

Electrical

We attached the pneumatic components, so that they will be ready when we get everything else built.

Marketing

We finalized our T-Shirt and jersey designs and got them approved at todays meeting.

Software

Tigerlib 2023.3.1 has been released! This update includes basic support for generic Graph<T> objects which can be traversed with a similar AStar<T> object, as long as you can provide a heuristic function. See more here: code & docs.

Sorry for being kinda dead for a while, we are super behind and pretty anxious about having a robot by week 1, if anyone knows how to get their hands on more time and/or competent freshmen please let me know!

Keep doing what you do best; we hope to see you all at comp very soon :).

[DAflamingFOX]

Update 5: “We participate in RitL3D”

General

Oh how stressful it has become, none of our parts are here, everything is out of stock, and we wouldn’t be ready if they were.

Build / Mechanical

Intake is complete, we built a final prototype just to make sure it would work ( which I promptly broke because it couldn’t handle the Neutron style ). We should have all of the parts for our intake tomorrow, along with our Grainger order of tubing, and our TTB elevator!

We have very suddenly realized how bad of an idea our old vertical elevator and super long lever arm is. We will be changing to a slanted elevator with a smaller V4Bar arm. Unfortunately we have not gotten a chance to update our CAD to match ideas, but hopefully we will have it later this week.

Here are some videos of the intake prototype:
Intake Prototype working via manual trigger
Intake prototype “working” via an 873M ultrasonic sensor

While a higher psi may result in more grip from the intake, it also results in far more stress, as we discovered by upping it too much.

Lastly we are finally almost finished with our field elements, which has been very good thing to keep the newer members doing. Yes, we are in fact still building our field elements.

Electrical

We got all the electronics mounted and wire managed so that it’s all clean and ready to go when we get our rapid building going later this week.

Marketing

No updates

Software

We’ve got all major code written, and just waiting on physical hardware to test it on. I’ve been working alot on our dashboard so that everything runs smoothly. Just for instance here is the scoring selector that the operator controls.

Till next week
- The most stressed and behind the schedule team.

[DAflamingFOX]

Not a full update but we finally got our elevator in and we are in build asap mode putting it all together.

We now will have something for Waco lol

See yall at week 1!

[Skiddy4795]

You might wanna consider vertically mirroring the scoring selector, since that’s how the drive team will actually see the grid.

[DAflamingFOX]

Thats a good idea, we will have to test when we practice driving. Whatever is most intuitive we will go with

[DAflamingFOX]

We are kinda dumb and kept mounting things wrong on stage 1, so we spent most of the day re-building it, today stage 2 went together much quicker.

Stage 2:

We did notice one tiny problem with the spacers for the cross beam interfering with the bearings, they also seem to interfere on CAD, but others have said theirs went together fine so we’re not sure whose at fault here.
Anyway our solution was to just dremel out a groove in the spacers lol.

Its a bit stiff at the bottom because we didn’t get everything perfectly square but a bit of monkey strength and just running it up and down a bunch, we’re pretty sure it will wear in fine.

Your favorite “Robot in the last week” is going to be cutting it stereotypically short once again!

Just imagine those poor programmers… oh wait.

[DAflamingFOX]

Update 6: “Did you build it right this time? Ehh…”

I am trying something new, hopefully the videos are actually embedded in the post, if not, here is a backup link where all the videos are hosted.

General

We have made a great deal of progress, at least for the amount of members we have. It’s difficult to work on a single task for more than about 4 hours, but thankfully we just drink caffeine and struggle along in our great big struggle bus.

This is truly the closest we have ever been to not having a at a competition.

With that being said lets keep these wheels rolling and make it to Waco in… oh my 3 days, we really aren’t joking about being a “robot in the last 3 days” team.

Build / Mechanical

Current tally of how many times something on the elevator had to be taken back off and reassembled: 13.

Yikes… however in doing so I would say we are very well versed in how this stupid thing works now.

It is almost fully complete, we just need to lengthen our chains a bit, they were too short for our configuration when we made them initially.

Elevator progress

NO we are not running the elevator with NEO550s, it just happens to be what’s on the gearboxes right now. They will be switched out for 36:1 anyway, rather than the current 20:1.

We have also began work on the arm, which will add a DOF to the elevator, and allow us to reach out of the bot much more than the elevator does on its own. The arm is fairly simple, its simply driven with a chain and sprocket from within the carriage of the elevator. There will also be a bicycle brake to assist with holding the arm in place, which will be actuated by pneumatics and a brake cable (obviously).

Lastly our intake, we just got our polycarbonate plates cut and should begin assembly of them tomorrow, we will be using two 837M ultrasonic sensors to detect and automatically intake game pieces, as we tested and showed in a previous post, it works fairly well.

We also designed some dope tube caps. Thanks to 8177 for the inspiration to make them.

Electrical

No major updates.

Marketing

No major updates.

Programming / Software

We have vision pose estimation working finally! Getting about 10-15fps running a OV5647 on a Raspberry Pi 3B+ running PhotonVision.

Secondly our “Autopilot” feature is just on the verge of being ready to roll, just needs some PID tuning and it should be ready to roll. I’m pretty sure its suffering right now just because we are waiting on a NavX from AndyMark after our Pigeon 1 died.

As you can see in the clips, we also have been working on our dashboard, you can check the code for it here.

P.S. Please welcome to the family George and Jerald!

[DAflamingFOX]

Left about 20 mins ago for waco and pits clost at 10, and its an hour and a half away lol. Looks like itll be morning load in. Unf our very important andymark package got delayed until tomorrow so we will be kinda scrounging for parts until later tomorrow.

Almost eveything is ready to just drop in our parts when we get them, It will just take some rapid pit building time.