Hello, FIRST/CD community! Team 3596 Rocket Robotics is excited to announce that we are joining the #OpenAlliance for the 2023 season, Charged Up!
We were founded for the 2011 season and this year marks our 13th season! We are from South Milwaukee, a city of about 20,000 and is a close suburb of Milwaukee, WI. Our team comprises of 2 mentors and 12 students. Only 3 of the 12 students are returning students, so our team is relatively green for our age. With this, we plan on building a #everybot for the competition, and will document what we liked, didn’t like, and changed from the initial design.
We plan on sharing design, prototyping, strategy, CAD, programming, competition updates, and other media. For CAD we are using OnShape, and Programming is using Java. I (Adam Fleming|Lead Mentor 2) plan on sharing most of the updates, but will try to have other students post if time allows.
We will be competing in one regional this year, the Wisconsin Regional in Milwaukee, WI, March 22-25.
Preseason Shop and Inventory Improvements
Last year we identified some issues with the equipment we had available, the layout of our robotics room, and our robot part inventory.
For large equipment, we used to rely on what the high school’s metal and wood shops had. However, a few years ago we were moved from our small room in the High School Tech Ed area into an larger empty class room in the Middle School Tech Ed wing. The good news was we had more space for working and storage – we can keep the previous season robot now! The bad news was room is on the opposite end of campus (about a 6 min walk) and the middle school only has some basic wood working tools. Because of this, we have decided to invest in a Omio X8 CNC Router and a WCP tube jig! We have slowly been figuring it out and hopefully I’ll have it figured out enough by the time we need it this build season. We needed to modify the nut strips that came with the tube jig as Omio had changed the extrusion, so the included nut strips were too thick; we used the Omio to cut it into a t-shape and now it all fits. We also purchased an 3-Ton Arbor Press, and a Metal Cutting Miter Saw. We are also looking to invest in a different drill press and possibly a mini-lathe.
3 main issues we identified with our room layout were:
- The build spaces were spread across the room, so cutting, drilling, etc. were occurring close to computer work areas, which increased our safety risk and risk of damage to computers other electrical equipment.
- Tool, parts, and material storage were not located close to their respective work areas, and the tool chest and some part storage required passing through manufacturing spaces. This was causing tools, parts/ parts bins, and cutoffs to not be put away and cluttered the room.
- Poor layout helped create a lack of collaboration between subteams, and a general lack of knowledge of what is going on with the robot unless directly working on it. Students were also able to hide away in some areas and get distracted talking to other students, which ate into the already limited time we have. (we only meet 3hrs per day 2x a week with a handful of 4-6 hr weekends during build season).
To fix these we:
- Split the build space into separate manufacturing and assembly spaces. We placed manufacturing in the very back part of the room, and made it so all cutting, drilling, anything that makes chips needs to be done in that area. Assembly was placed in the middle of the room.
- A new tool chest was purchased and placed in the assembly area, on the side closest to the manufacturing area to keep them as close to the groups that need them the most. The main large parts storage was also placed in the assembly space. We also purchased a large amount of small parts organizers for bolts, nuts etc. for which a cart still needs to be built which will be placed in the build area as well. We also created a more compact material storage area and placed it close to the manufacturing space.
- With the robot assembly in the center, everyone now has eyes on the robot progress. The design and programming spaces were combined into a joint space, and placed close to the assembly area, and our marketing and admin spaces were combined. All tables and desks are laid out so everyone can turn towards to front center of the room and see our TV cart or chalkboard if we are having a group discussion, and most face towards or can easily turn and see the robot assembly space.
When it came to robot parts, historically we only made purchases when we needed an item, used what came in the KOP, and scavenged parts from old robots that we already disassembled. This was OK for our first few years, but 13 years on we were using a mix of ecosystems and using some really worn out motors as we didn’t track what was used when – we even have some nearly 20yr old CIM motors from when our high school briefly had a team from 2003-2004 . But now with the advancement in COTS motors, gearboxes, etc., our additional storage space, and sporadic part availability, we realized we needed to build up an inventory of the latest motors, motor controllers, and gearboxes. We decided to invest in the Rev Robotics eco system and stocked up on Spark Maxs, Neos, Neo 550s, and and assortment of Max Planetary gearboxes. We are also planning to track motor usage by at least logging what year it is used and what it was used for.
Mock Kickoff with 2506 Saber Robotics
On Saturday, December 17th, our friends over at Saber Robotics in Franklin WI invited us to join them for their Mock Kickoff. The goal of the mock kickoff was to introduce new students to what to expect from kickoff on January 7th and practice interpreting the rules, creating game strategy and developing initial robot design. 5 Rocket Robotics students and 50+ Saber Robotics students acted as one team for this exercise.
For the game, the mentors chose 2016’s Stronghold. After watching the game animation, students split off into 6 groups: 5 groups to discuss and make presentations for the robot rules (dimensions, components allowed), game rules, scoring strategy (maximizing points, cycle times), tournament strategy (scouting, maximizing rank score, making playoffs), and initial drivetrain strategy (researching what drivetrains would work for the game); 1 group worked on the field layout (taped out the field on the floor). These groups presented their findings to the whole team and discussed.
With the rules information and initial strategy, the team split off again into different groups to create some initial design concepts for climber, ball scoring, defense manipulator, and drivetrain. The field layout was also used at this time and groups rotated in an played a mock match with the students, 3 acting as “robots” and 3 acting as “drivers” telling the “robots” what to do. This allowed the students to understand how fast the game is and see how the game is played in person.
Rocket Robotics Strategy for Kickoff
We are planning on joining Saber on January 7th for Kickoff, and Rocket Robotics decided to modify how we will approach kickoff to better suit our needs as our capabilities and goals are different. Our team will join the rules and strategy discussions, but our students will be taking separate notes on what applies to our team and what we might need to consider differently. 3596 will then convene separately to create our own strategy and design concepts. We are still planning on using the KOP chassis and using the Everybot design, so we will be considering those in our discussions.
Our strategy is similar to what the Everybot goals are this year.
In the Autonomous Period:
- Score a pre-loaded game piece into a hybrid node
- Score Mobility points
In the Teleoperated Period:
- Score Cubes in the Bottom and Middle Rows
- Score Cones in the Bottom and Middle Rows
- Acquire Cubes and Cones from the Substations
In the Endgame:
- Dock on and Engage the Charge Station
- Be narrow enough to allow room for two other robots on the Charge Station
With needing to be narrow, we decided on the long configuration for the KOP chassis with Neos for the drivetrain motors.
We are planning to use a wheeled intake type grabber with claw adjustment (pneumatics or motors) to grab both game pieces.
Initial strategy is to use a single stage elevator with a single joint fixed length arm. Elevator is planned to be variable position. Arm is still being decided on if it should be variable position or just 2 position: stowed and deployed.
Drivetrain and Grabber Prototype Progress - Week 2
KOP chassis is complete in the long configuration! Only issue we found is the drive wheels have a wobble to them. We’ve checked the shaft, chassis rails, and wheel assemblies and couldn’t see any issues; we will be looking into swapping the inside rail hex bearings on Saturday to see if those are causing any issues.
Yesterday we worked on prototyping a wheeled grabber, similar to the Zoukeepers Ri3D design. It uses two 4in compliant wheels on top, and 2 3in compliant wheels on the bottom. The final design would use pneumatics to open and close the claw. The big takeaways are:
It works very nicely for the cubes, and works acceptably for the cones; using it as a claw vs roller intake is a good backup if the rollers cannot suck in the cone.
We will need to have a bigger back stop and maybe some small horizontal end plates to keep the cube from ejecting out the top of the grabber. Also would be good to have a time of flight or limit switch for object detection.
The cones need a lot more force to grab and hold well, and placement of the cone in the grabber is key. We may want to have a second set of roller behind the intake rollers to give more gripping surface with the cone, if it doesn’t effect the cube intake.
Comparison to Everybot roller intake
With our prototyping of an intake we have been able to compare it to the Everybot design:
Our small opening vertical roller intake will require more precision to grab game pieces vs the Everybot full width horizontal roller intake. We don’t see this as as much as an advantage as substation provides some side to side aiming from the human player.
Our protyped design might have a slight advantage on consistent scoring position as the intake is the same for cones and cubes, and cone position in the claw looks to be more consistent vs Everybot, This may be negligible with the wide opening of the cone.
Our prototype does not need to reverse rollers for scoring cones, it can just open the arms. This give our design a slight advantage as we can have full placement of the cone before releasing it. Everybot needs to be above the node before releasing.
Everybot will likely have less bounce out chances for scoring cubes as it rolls out the cubes top down, vs the our roller design that needs to roll it forward to score. This could be mitigated by have a 2 stage claw that can release the cube fully without needing to reverse the rollers.
The plan before the start of the year was to use the Everybot design, but as the season progressed have been able to draft some ideas up and prototype quicker than expected. With this comparison we are more likely going with the roller claw design.
Week 3 Updates - Angled Elevator Build and Grabber Design
We’ve finished the majority of the work on the elevator super-structure and carriage. We did run into an issue during assembly where the base width of the elevator triangle was 25.825 instead of 25, and they were slightly twisted. We spent about 45 minutes reviewing the drawings and inspecting our parts and assemblies, we found that the the top angle was about 1° over spec at 31° instead of 30°. We had already measured the tube cuts prior to assembly and those were 30°; closer inspection of the joint showed a gap on one side of the butt joint. We’ve determined the root cause to be that while adding the second plate, we had clamped it wide and spread the joint. This caused the twist and the increase in angle. Going forward we will make sure to add a shim - about the same height as the bracket and rivet heads - under each tube section before clamping it down if we need to add a bracket to the other side. Luckily the fix for the completed assembly was just to adjust the CAD to match as we had the room.
Week 4 plan is to add bracing to the super-structure, add the rope and pulley assembly to drive the elevator, and get started on the arm pivot to attach to the carriage.
Still working on the grabber CAD design, we have it roughed out at the moment with most parts placed. For opening and closing the grabber, the plan is to use a 3/8 bore, 6" stroke piston that we have on hand, but we may need to purchase a shorter stroke one to better fit the design. To spin the intake wheels, we are going to use Neo 550s with a 5:1 reduction running at a low speed. Currently have Max Planetaries for the redution, but an ultra planetary or belted reduction might suit this better.
Week 4 plan will be finalizing the design, to then start manufacturing it by Week 5.
Quick Mid Week 4 Update
**Elevator Structure is (basically) Complete! **
Seems like we weren’t going to get this done last night, but we pulled it off! We still need to add our end stops/bracing, but those should hopefully go quick.
It was close about 1 hr into the meeting, but then when we added the carriage, it was too short by about 1.5 inches. The tubes on the width were cut too short, so the only build team student that was there drilled out the rivets while I cut new tubes. Before we riveted back together, we made sure to check the dimensions by measuring with a different device, and lo and behold, the one I cut were even shorter! (Cut lines were marked with a machinists square, check was done with a tape measure). Turns out we had both likely used the machinist square wrong, in my case I had set it to 13.875, but had the edge flipped the wrong way and I accidentally measured from the 24 in edge to 13.875. With the square setup correctly, we cut the pieces a third time - to the correct length this time. And with about 15 minutes left for cleanup we just barely finished.
Learning from this, we will now be checking all the first cuts with a tape measure in addition to the squares for shorter parts to triple check we got it right and don’t waste time and material. We also need to get better squares not from Harbor Freight that only have inch measurements, as the couple we have are a real pain to flip around and could also be a cause of miss-measurement.
Week 5 Update
We’ve completed the design of our grabber! It uses 2 Neo 550s connected to Max Planetaries with a 5:1 reduction. We are using the shafts of the gearboxes as our pivots for the arm as well. A piston changes the grabber from a cube pickup position to a cone pickup position. The intake wheels are a 4" Compliant at the top, and a 3" compliant at the bottom.
Our initial teleop programming has been completed! This was done in Java using the command based framework. We are now working on autonomous mode and setting up closed loop control with encoder feedback on the elevator.
Our elevator is fully complete now with our rope winch and pulleys attached to the robot. We also started using the CNC to cut out parts for our grabber. It looks like there are some issues with that still, so we unfortunately had to scrap the first batch of parts as bearing holes did not line up.
Week 6/7 - Nearly Complete!
Arm and grabber have been built and assembled to the robot this past week, now we just have some finishing touches and wire it up this weekend to get it moving! There are a few things we’d like upgrade/change before the Wisconsin Regional, chief among those being the arm joint.
We decided to use the REV Pivot Joint (REV-21-2395-PK2), but after assembling it it has more play than we would like and could result in the piston getting damaged if the arm has a side impact.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.