Scouting strategy/structure could be a good one in terms of what data you decide to collect and how you use it. Looking at how you motivate people to scout would also be a really useful video for a lot of teams as that seems to be a pretty consistent pain point.
2 Likes
The Auton Post
We thought covering autos and our design process for them would be useful. Before we go through each auto, we should explain the nomenclature we use to describe it.
Naming
Each auton follows the following naming scheme: (Rush) ##.# (Drop)
Rush: When the auton has the rush prefix, it rushes to the midline.
###.#: We number the center game pieces 1-5 from the amp to the source side. For Example, this is Rush 12
(Drop): This is the cool one. When we use the drop suffix, we drop one of our game pieces mid-routine and come back to pick it up and score it later. Here is Rush23Drop:
Process
Our Auton design process starts by making a priority list. For our first event, we knew most teams wouldn’t be grabbing notes from the centerline quickly, so we prioritized ensuring we were hitting every shot and only scoring five notes. Here is our Belton Auton:
For Houston (District Event), we knew there were at least three teams that were going to be competing for the centerline notes, so we made minor improvements, turning the five-piece into a six-piece and getting rid of a code bug where we would sit for the first second of auton. We also started using the entirety of the auton period; most of the time, we were able to squeeze 15.3 seconds out of it. Here is what that looked like:
After Houston, we increased the speed and consistency of our auton, turning it from a 6 into a 6.5 for DCMP. We didn’t think the centerline would be as competitive as it was, and guaranteeing the front 4 was more important than the last two on the midline. We also had an IQM of 5.2 Auton Gamepieces during qual matches. The next highest was an IQM of 3.0 Auton Gamepieces. Here is a video of that auton:
Champs Autons
This is where it gets fun.
Going into Worlds, we were confident in our autonomous routine while recognizing the prevalence of rush auton strategies more so than at DCMP. We designed multiple autonomous routines starting from the same spot to counter potential strategies from other teams that would arise if we were starting from different locations. All of these were designed with batch 1 notes, so some modification was needed to accommodate the Championship notes. We were on the tethered practice field whenever we weren’t in a match to adjust for the differences. Here are the different auton routines we brought to Champs:
Basic Rushes
We knew we needed these rushes to be compatible with other robots in our alliance and ensure we could always beat the opponent to centerline notes.
Rush 12
Rush 23
Rush 13
Rush 31
Rush 43 Drop
Rush 23 Drop
Here is an album with our recorded auton testing from the Saturday after DCMP to the Friday of Worlds.
Offseason
Rush 3-4/5 Double Drop (1323 drop Auto)
We have been working on this auto for TRI for the past week. We are bringing our Practice (8515) and Comp bot (3847), so we need something compatible for both robots to run.
Thank you to 118, 5414, and the practice field FTAs for allowing us to use their fields.
36 Likes
New off-season robot
We’re halfway through our summer camp for new students. The camp students get to build and compete with a new robot, which we designed to be as simple as possible. This year we took inspiration from 9496 LYNK. Here’s a link to their OA thread, they have a very cool robot.
- Onshape link
- Identical drivetrain and intake to all our other robots this year, and same indexer as Ultraviolet
- Fixed-angle launcher (hopefully) capable of launching into both speaker and amp by altering the speed of the top or bottom launcher wheels
- Simple telescoping climber (We will add this late next week)
- One limelight
- The robot was built, wired, and powered on in just 4 days! (sans the climber)
- Summer will be competing at TRI using the off-season number 9985
- Photo Gallery
32 Likes
This leads to 9496 OA post instead of CAD.
fixed, thank you!
3 Likes
@EmmaS @AllenGregoryIV question on your tube plugs:
Do you guys generally just hot glue these in? Epoxy? How well do these hold up and have you experienced failures of them coming loose?
To quote an older post,
On our practice bot, some of the rollers had only hot glue securing them, and some of them did end up coming loose. For the competition robot, we drilled holes and screwed in bolts after gluing, one bolt per hub. We never had a roller hub issue on the competition bot.
5 Likes
I love to see what other teams are doing with this architecture that is a sweet looking bot. I really like how the shooter seems to be detachable from the base frame.
11 Likes
Strategy Overview
This season, we’ve refined our strategy documentation and data management to enhance clarity and efficiency. Here’s an overview of the key components:
Internal General Strategy Training
All team members are required to review this slideshow before attending any event. It covers essential strategy principles, ensuring everyone understands our approach and objectives.
Data Collection Sheet
This sheet is where our QRScout data is scanned and stored. It is the primary repository for all raw data collected during events, organized for easy access and offline use.
Data Analysis Sheet
This is our most frequently used document for deciding strategies during qualification matches and alliance selection meetings. It provides in-depth analyses of the collected data, helping us make informed strategic decisions.
Robot Database Sheet
This sheet is used primarily during strategy meetings. It offers a quick summary of each robot’s data, providing a concise overview that enables fast and effective references during discussions.
Alliance Selection Sheet
Created during the strategy meeting, this sheet includes our pick list and alliance selection simulations. It prepares us for alliance selection by ensuring we have 24 (or 32 at worlds) robots on our pick list, ready for any scenario.
Please ask any questions you may have, I’m sure that there is more we can elaborate on about our process.
29 Likes
Just out of curiosity in addition to I’m assuming you use quantitative data for making your pick lists do you use any subjective criteria such as who you worked well with in matches or who played good defense according to scouts?
We actually had a system for qualitative scouts (or super scouts, depending on what you call them) in place and used it once before we didn’t have enough people available to maintain a comfortable schedule. One of the new roles we implemented this year were Teams and Robot Experts (or T-REXs) and they were in charge of taking notes on a group of robots assigned to them (more details are in the Internal Scouting Training slideshow), in addition to the text field that scouts are encouraged to fill in while inputting quantitative data. - Our scouting lead
So yes, we definitely consider qualitative data during the pick list meeting and alliance selection. This includes info from T-REXs, data from qualitative scouts (if we have them), and comments made by quantitative scouts (this slide covers what we ask scouts to put in their comments.)
3 Likes
Off-season robot and PM updates
Summer robot updates
Today is the last day of the summer camp, and Summer is mostly complete.
- Working amp shot
- Mostly working auto-aim (works up to five-ish feet from the subwoofer)
- Climber CADded, built, tested, and working
- Note blocker panels made
- Switched to black nitrile tread
- Simple shoot n scoot auton
We still need to make more autons and add LED indicators.
PM updates
PM (Practice Machine) will compete at TRI using Photon’s number, 8515. We updated its climber to match Ultraviolet. We’re updating its Amp-Trap to match UV too. There’s still a lot of work to be done to get PM up to speed.
13 Likes
I really like the T-Rex Idea, however does that rely on your team doing all the scouting and not scouting with others. Because for the last few events we’ve been to, we’ve scouted with other teams to help them out/because we don’t have that large of a team.
Yes we do always scout with just our own people because we have to train them before the events. Even then it’s still hard to get consistent data.
2 Likes
What technique did you guys use to get that range? Varying overall power, top vs bottom power, or did you guys do something else?
Yes it was just top vs bottom differences. For the subwoofer it was 6000 and 3000 rpm.
2 Likes
Texas Robotics Invitational Event Recap - 3847
We ran three robots at this event, so we will be making three separate posts, one per robot.
Thank you to all the teams who attended TRI this year, and a huge thank you to all of our wonderful volunteers who made the event a success!
Last weekend, Spectrum 3847 competed at TRI with our in-season robot, Ultraviolet. We ranked 1st, won throughout the upper bracket and finals, and went 14-1-0 overall. Thank you to our incredible alliance partners: 3005 Robochargers, 5431 Titan Robotics, and 5414 Pearadox.
Fun fact, this is our second TRI win, with our only other win being the first TRI in 2014. The fact that we almost always rotate drive teams at off-season events, but made an exception for this year’s TRI and didn’t rotate on 3847, probably has something to do with this.
Pre-event
The only change we made to Ultraviolet since Champs was replacing the cracked polycarb launcher and indexer plates. The robot somehow worked perfectly fine with the old ones (pictured below), although I don’t know how much longer that would have lasted.
With the fresh plates, the robot’s range increased by about a yard, and it can now launch consistently from the wing line on our practice field, which we’ve never seen it do before.
Failures and Fixes
Failure: In practice matches, we had radio/lag issues.
Fix: Secured the radio more tightly and turned off logging.
We had no failures on Ultraviolet during real matches. In fact, nothing broke at all.
Controls Update
-Autonomous-
We mostly ran our Madtown and 43 rush. The 43 rush consistently bumped the truss and shot low, and we never adjusted up enough.
-Other stuff-
Since we turned off logging, Elastic wasn’t updating to show whether the robot could see notes for auto-intake. Auto-intake is useful for cleaning up notes hidden behind the subwoofer when you’re in the middle or source-side driver stations (which we were for every qualification match, by the way), so we eventually turned logging back on just for auto-intake.
We will compete with Ultraviolet at more off-seasons this fall. We don’t have any changes planned for it.
41 Likes
Hey, did you guys do like a strategic season overview? Not like how you did strategy, but about the game strategy overall this season
I am making one right now and I want to see if I am missing some content or not
3 Likes
I don’t think we ever made anything like that. The strategies during the season evolve as we see new strategies play out around the world.
There are basic things like how to play when trying to score more points vs trying to just get RPs, etc. I do wish HQ would make those things more the same, Note points vs total notes, etc.
1 Like