New System for Scouting?

HELLO! I am a new member to the first organization. I am on team 1126 (Sparx) and I am doing a project for the team to find a new way of Scouting. I am leaning twards the PDA systems. Any advice of personal experiance, or just common knowledge would be great.

:eek:

There are many threads that discuss scouting at competitions. If you use the search feature, you can find many of them. Here are a few to start you off.
http://www.chiefdelphi.com/forums/showthread.php?t=56433&highlight=competition+scouting
http://www.chiefdelphi.com/forums/showthread.php?t=55461&highlight=competition+scouting
http://www.chiefdelphi.com/forums/showthread.php?t=52269&highlight=competition+scouting
http://www.chiefdelphi.com/forums/showthread.php?t=46280&highlight=competition+scouting

Additionally, if you search the white paper section for scouting, you will find scouting sheets and programs that many teams have used.

Welcome to CD!

Welcome to FIRST. Rochester has a great FIRST community and you are on one of the elite teams in the area. Matt Starke, a mentor with me on Team 340 is amazing at scouting and reads CD often. I’m sure once he sees this he will be happy to go over some ideas with you. Good luck in this endeavor and again welcome to FIRST!

In our 10 years, The Pink Team has found paper scouting to be cumbersome.

My recommendation as a developer is against PDA scouting. While it seems convenient, its unlikely that enough team members will have PDAs (6) to scout fully, not to mention all of those PDAs running the same OS (PalmOS, Windows Mobile, etc.) Standard laptops are much easier to come by, or be borrowed just for competition periods (this is what we do), rather than purchasing team PDAs.

Another major disadvantage to PDA scouting is that they tend to be hard to develop for, not to mention the problem of getting the data OFF of the PDA into a usable form for strategizing.

If you want some ideas or more details on how we do scouting, don’t hesitate to PM me.

Team 1726 will be releasing a scouting system in 2008 which will be similar to PDA scouting, in a manner of speaking :slight_smile:

I think you should definitely go ahead with creating a PDA scouting system. It’s a neat opportunity for you to explore new technology and make something useful with it!

It depends on what you need. For example, robot functions and features will differ year to year depending on the game. So, a new scouting program should be created each year to scout specific features of a robot. The best way to go about in scouting is paper and MS Excel =D. Its actually quite easy and not cumbersome if you can get yourselves organized. Our team has said it again and again…“Scouting can win you regionals.”

Team 190 did “proper” (read: organized) scouting this year for the first time in many seasons. We used 6 sheets of paper per regional and one excel spreadsheet, and kept the data to a minimum. There’s no need to inundate yourself with more information than you can handle.

For example, drivetrains: Does it really matter (from a pit scouting standpoint) whether or not a team has 1 speed, 2 speed or 15 speed transmissions? Instead of asking that information in the pits, it can be more useful to just let whoever’s watching the matches make notes on driving skill.

I am not saying that the information is not needed, it’s just an example of something that could be considered extraneous

We collected more information from the stands, making notes on consistency, and what kind of strategy decisions they seemed to be making. We also took input from our coach on which teams she had good experiences working with, and which teams she didn’t.

It should also be noted that individual performance often does not agree with ranking, due to poor alliance partners, this year’s dreaded “perpetual opponent” and other factors. Rank teams based on how they are performing on the field, not necessarily how their alliance is doing.

Performance can very greatly between Friday and Saturday morning. Drastic changes to the robot or the code can positively or negatively affect performance. Don’t be afraid to shuffle around teams on your pick list based on the most recent data trends.

And lastly, for all the data in the world, don’t forget common sense and gut feelings. Statistics and trends don’t do much good without some context. If redabot can put up 5 tubes per match, make sure that they’re not being left undefended before you put them as your sole tube scorer.

I don’t know anything about scouting or PDA software.

What I do know is that I never really did any REAL scouting as a student. I was my team’s programmer for 2 years, so I was stuck in the pit doing last minute fixes. I thought it was great that I got to skip out on scouting duty. I was forced to do it a few times, which was kinda boring.

As a mentor I had a bit more time, and I did some non-official scouting. This made me realize how much I missed out on as a student. Casual scouting is really fun. It is a series of conversations with really cool people about things that they are really into.

What ever the tool set you come up with does, I hope it encourages long winding conversations about whatever. I therefore propose the following form:

Stand in front of their pit and look confused.
How many seconds until someone on their team initiates a conversation?
Was it a student or a mentor that initiated?
How enthusiastic were they?
How many students join the conversation?
How many students linger behind?
Ask for “the electrical person”. Do they redirect you to a specific person, or do they do everything communally?
Stare intently at something weird/unusual on their robot. How many seconds before they begin to tell the story of its creation?

And finally…

You have partnered with them. There are 10 seconds on the clock in the tie-breaking match to win regionals. Your alliance is down by 1 [ringer, ball, tetra, gerbil,…]. Their robot suddenly wigs out, smashes into you, bursts into flame and releases enough magic smoke to fill 30 magic lamps.

Are they cool enough to turn this into a story you will tell your great grand-FIRSTies?

I love it!!! This is an entire element of scouting that I have completely overlooked. It raises awareness of robot elements, team dynamics, and initiates interaction between teams that lead to learning.

<Adds to must do list>

We have a PR group whose responsibility is to do what you have just described above.

When you have 50 teams in a regional pit to drivetrain scout (or upwards of 80 at nationals), and no more than a day and a half to see all of them, learning everyone’s life story just isn’t practical.

The first concern of a scout should be performance.

The scout’s job is to gather and evaluate data on a robot’s construction and abilities, as well as its performance and that of its drivers on the field. The ability of the team to collaborate with others in a GP manor should of course come into consideration, but only to the extent that affects their ability to work together with alliance partners to win matches.

Our drivetrain scouts (and in fact all members of the team) are allowed and encouraged to walk around the pits, seeing robots and meeting fellow FIRSTers, but only when they’re “off the clock”. If you have field scouting duty, or if we have a match, or, in this case, you’re a drivetrain scout who hasn’t yet gathered data on all the teams, you should be doing your duty to the team, not socializing.

I guess what I’m saying is, in scouting, scouts should be scouts first, team ambassadors second.

I find that Thursdays are useful for pit scouting, and then Friday/Saturday for match scouting. Team8 did this at both of our regoinals, and it was awesome. It was a lot of work, especially because we usually had six people, one watching each robot. Then, when we got back to the hotel, we did data entry until 1am. Laptops would probably have been a better idea, but we just didn’t have enough.
I feel like PDAs can be a little bit awkward, just because you’re dealing with a small screen, plus they can be expensive, or the old ones just aren’t very good. The PC is just an easier platform to develop for.

Something I’d like to see a movement towards would be a massive data collection cooperative effort across all regionals. I think there was a system like this at Championships this year. I envision some sort of web-based system that would allow for some sort of redundant data collection and organization. Even match results are hard to come by for some regionals. Of course this would rely on reliable wifi at regionals, but I’m curious to know what kind of interest there is in a wide-scale data management system.

This year, I was one of the two scouting leaders for Cyber Blue, 234. We given the opportunity this year to experiment using Pocket PCs as a part of our Scouting system by our school district’s administrative office. First, we just experimented using the PDAs to see how accessible they would be. We found that one of the most important things about Scouting using a PDA system is that it must be done quickly, easily, and accurately. We eventually found that the Pocket PC version of Excel was the best program. We created a template in which we used a 1-10 scale on physical qualities and strengths to the robot. We also had a y/n part about penalties and other information that could be covered that way. For scoring, we simply said how many they scored on each level H/M/L. We saved each file under a special name correlating to their place on the field (ex: Red 1 or Blue 3) and the match number. Using Bluetooth on the pocket PC’s we had one that was used for taking the files from each Pocket PC. The problem with this is that we had to grab each file one at a time making the process slow. We did this periodically through the matches to gather information for our next match. When we had gathered the files, the master Pocket PC was then plugged in to our Scouting Laptop. We had an Excel database on the laptop that put the files into a nice Spreadsheet that gave us averages and ranges of the scores, and numbers that determined overall abilities of the robot. The problem with this system is that one small mistake could change the outcome or cause an error. With proper trainging of your scouters though, this type of system can be great though.

What you need:

At least 6 PDAs: 7 or 8 is better because they do run low on battery power.
A Scouting Laptop: Formats all the information into one spreadsheet and creates a better presentation for drivers and mentors to see.
TRAINING: Train all your scouters very thoroughly to make information as accurate as possible. It is very easy to press the wrong letter when typing with a stylus.

This year, we had a whole backup system of paper scouting forms to because we weren’t sure how well the PDA system would work. In addition these served as a better place for marking Robot strengths and weaknesses.

GOOD LUCK!

Team 1629 has tried to do this for the past two years, but the problem is that on Thursday many teams are still making changes or are at practice matches. Because one part of our scouting is pictures of each individual robot for the more visual members of the team, this is very difficult and at our last regional we were still taking pictures as part of pit scouting on Saturday.

Our team has six scouts watching the field and recording information on paper and then the papers are delivered to two other team members that are inputing data to our Acess scouting database on a laptop. These members are also in charge of videotaping the matches, both for scouting and team videos.

Scouting is the most tedious part and the most annoying job that gets done for all the matches. My team doesn’t have Pocket PCs, PDAs, or an extra laptop that can be used for scouting. So all the scouting has to get done with pen and paper. Keeping track of how each team did and how their robot is made is hard to keep track of whan going through two bingers of information. It is no fun to have to watch a match and record everything, when everyone is there to have a good time. Scouting in the pit was more exciting because you get to see the other robots up close and get an idea of how the different teams build their robots and how the work…

I am honored to be known as such a person. Welcome to the scouting world of FIRST! Below, I have put together different posts from previous threads of what Justin is talking about. If you would like more information, let me know. This year, I am passing off the torch, for I will be away at college. I will continue to help out where I can. Know that I think about it, SparX is not that far away from Oswego! :yikes: I wonder what Dylan thinks about that…

I think there are a lot of teams that would like to see a single, unified scouting system.

As a driver this year, here’s what I cared about:

  • Strategy used (What do they do? Score? Defend? etc.)
  • How good they were (Do they actually do what they aim to do? How effective?)
  • Any special abilities (eg, 111 is good at avoiding defense)

From a strategizing standpoint, we care about what they do, not the means by which they do it.

Having a centralized system could be greatly beneficial for teams. If data were collected from all the teams in a regional in a uniform manner, it is feasible that the data would be better than the data a single team could collect.

One idea is to borrow the “rating” system seen on many websites (YouTube, eBay, etc) and apply it to robots in various categories (like those mentioned above). If a significant number of teams participate, meaning there are significant numbers of contributers, you can start applying statistics to it.

The reason this works is the “wiki” principle, the wisdom of the masses. If you have a large number of people watch a match, and rank the robot on a scale (1-5, 1-10, 200-800, w/e), there tends to be a center, an average, of the ranking. (I read once that this happens at about 30 data points.) All that would be needed would be that for a match, you give a robot 5 or 6 ratings in different categories and submit it.

From discussions in IRC, there seems to be significant resistance to a collaborative scouting system amongst teams. (esp the successful ones.)

Does any of that make sense?

I personally like the idea, however a unified system might take away from the “spirit” of scouting, which is an important part for each team.

Personally, it seems odd enough to me that people share their scouting strategies (but I’m not saying bad, I’m willing to give my thoughts, in exchange for others) but I cannot imagine any “common” scouting system working. When it comes down to it, actuall scouting data is collected to help your team compete, and even if someone does create such a system, I have a feeling that successful teams will still rely on their own scouting.

Again, I don’t oppose inter-team co-operation, but as long as we have robotics COMPETITIONS, there will always be some competive spirit.

My advice is that paper is better. I believe it is way easier and managable for a newer team. At Indiana Robotics invitational (iri) pink had about 8 labtops lying around and it took up space and you have to bring all kind of cords and adapters to make sure you can get the laptops going and laptops can be distracting as well, distracting members from the copetition because members can be playing card games. If you use papers, you can have a sheet for each team in a small file folder run the 6 sheets needed to the pits to discuss stategy for the next match. I believe it is all around easier process. (we have used pdas as well)

CYBERBLUE234 WELCOMES YOU TO FIRST AND ALL THE EXCITEMENT OF COMPETITIONS,BUILD SEASON, AND ETC!

any questions on how we run things feel free to ask.