A Comprehensive Guide to Competitive Driving

Competition
#1

Hey teams!

Looking back on the 2021-22 season, I thought it was a pretty good comeback from the pandemic craziness. Comps were fun to be at, teams were able to show up and compete pretty well despite having mostly new members, and most of all teams got a new batch of rookies to mold.

It seemed that most teams were able to get their rookies up to speed with regards to hardware and software, but one of the main aspects that I felt weren’t quite up to speed was driving. It’s totally understandable, it’s hard to teach someone how to competitively drive. I’d like to think I’m a pretty good driver, so I’m just going to summarize my tricks to successful driving. If anyone else has any cool tips then feel free to add!

The biggest thing: Don’t be afraid of the robot
I know it can be scary to drive a 120lbs brick of metal at 15 feet per second, but you shouldn’t be afraid of breaking it. Now, don’t go all gung ho and crash top speed into the field wall (although it’s probably the wall that would break, not the robot) but also don’t drive like an 80 year old grandma going 45 on the interstate.

That said, Respect the machine
Robotics is not a risk-free activity. Students are operating industrial production machinery, programming 120lbs death machines semi-autonomous vehicles, and competing them in very fast-paced activities. These things have the potential to seriously injure someone (just ask any mentor who’s almost been run over) or damage expensive equipment. Just don’t go too crazy with them.

To be successful: Know the rules!
To save your alliance potentially many dozens of penalty points, know the game rules! This doesn’t mean know the entire rulebook cover-to-cover, but you should have a pretty good handle on where your robot can or can’t be, or how your robot can interact with game pieces or other players.

Always be affecting the score
Nothing can kill a team’s chances like having an idle robot on the field. Obviously this can be unavoidable sometimes, like when you lose comms or a critical hardware piece, but if you can help it, try to be affecting the match score. Of course, the best way to do this is by scoring for your team, but if for some reason you are unable to do so, then you can shift to stopping the other team from scoring.

However, it’s important to note: Don’t default to playing defense! ESPECIALLY IN QUALIFIERS!!
If your robot is able to score points, then score points. At a competition, don’t expect every team to be scouting just based on W-L records. If that happens, something is going wrong. Any good team will scout based on your actual performance, not just a number. At least for scouting on 1895, the first thing we look at is the OPR (offensive power rating). The OPR is a pretty good way to get a rough estimate on how many points someone is scoring in a match. Now, if there’s some reason that you can’t play decent offense (whether it be robot limitations or hardware issues), then you shift to defense.

When playing defense know where everything is at all times!
You don’t need to know where every game piece is every second, but having a good idea on where there are big groups of them will help your decision making on who to defend against and how to do it. At one of our regular season comps, our team was picked to be in alliance because of our defensive skill. Our auto and general offense was okay, but our real power came from being able to control where the game pieces were and keeping opponents away from them.

Some helpful driving tools: A flip controls button and a quick 180 turn locked down
One of the things that separates good drivers from really good drivers is their agility. Last season, our robot had it’s intake on the rear of the chassis while the shooter was facing front. Now, a potential problem with that is if your driver isn’t totally confident with backwards driving or they are afraid of tight maneuvers. To remedy the reverse issue, I bound a flipped drive command to the left bumper of the controller (I naturally rest my index finger on the bumpers of the controller, but you should bind it in whatever way will let you use it without moving any fingers). It had the effect of switching where the front of the robot was. That way I could steer the robot going backwards with the intake down as if I were driving normally, which made picking up cargo easier. Also having a quick 180 on call (don’t bind it to a button, just have something that you can reliably do for turning around on a dime) is extremely helpful. Sometimes, a driver can waste precious seconds trying to turn around, then overshooting, then overcorrecting, and so forth. Just have a nice little 2-point turn you can quickly do and be on your way. Pivoting is also an option but In my experience I’ve found it not to be as reliable. A good time for a 180 (entrance, turn, exit) would be about a half second.

If this applies to you, it’s extremely important: A reliable vision targeting system is your best friend!
If you have a really accurate targeting system with your on-board cam, use it as much as possible! Part of why my team’s robot was so accurate (about 90-95% accuracy) was a well tuned, well executed Photonvision targeting system. We positioned the camera itself in a way that it could see the hub from pretty much anywhere on the field, as well as having a separate webcam on the intake for picking up cargo. I had buttons (I think it was A for hub, X for cargo) bound so that if I held the button, it would position the robot exactly where it needed to be. We calculated the range to the target, as well as how far off it was from the center, then steered/drove the robot so it was the perfect range and the perfect heading to land a shot.

As good as vision can be, don’t forget to get pretty good at eyeballing shots
Sometimes the vision system can be finniky at best, depending on the quality of the camera and the skill level of whoever programmed your vision subsystem. Try to figure out where your robot needs to be relative to field markings to make the right shot. Use features and lines on the ground to decide this.

That’s mostly it for me, feel free to add if you want!

17 Likes
#2

I disagree on this - if you are going to contribute more points in the negative direction to the other alliance than you’d score, play defense. If you’re a skilled defensive driver at the district/regional level, teams will notice, and you will be picked.

62 Likes
#3

I’m gonna have to disagree on the defense point-- there are teams that can contribute much more on defense. Especially the new teams who may stuggle to get an effective scoring system can still do really well on defense. The stats may not be the prettiest, but teams will notice if you do it well. We almost upset the 1st seed alliance in quarterfinals this year because one of our alliance mates was able to shut down their best bot.

Point being: defense can work even if you aren’t super high in alliance rankings.

14 Likes
#4

Disagree on the defense point as well. Unless things don’t go to plan (and improvising is needed), it’s generally best to do what the alliance agreed what was to be done (in terms of role within the match). Being uncooperative/rogue is generally a great way to end up on a DNP. FRC is a team and alliance sport after all.

How to develop match-to-match (alliance to alliance) stategy is a whole different discussion.

9 Likes
#5

Also, to the OP, if you have the manpower, please stop using OPR as your first look at a team. Please start scouting and using that data to better determine your picks. OPR is an okay but very very flawed statistic.

23 Likes
#6

Okay, so here’s my response to everyone all at the same time.
First off, when doing scouting we don’t solely use OPR as a field for scouting, it’s just a way to start separating the teams. We go into deep detail about auto and tele scores, driving styles, etc. OPR is just an at-a-glance way to get an idea for how teams are doing.

As far as defense goes, It drives me insane when a team with a really good scoring robot defaults to “I’m just going to play defense” because the driver isn’t confident with their skills. Win-Loss record doesn’t really affect how teams will scout you (at least with good teams), but points do. If you can score the points, score the points. If you can’t, that’s when you’d play defense.

Playoffs are a different case, where if you are going to play defense in the alliance it’s because you were scouted to play defense. That’s how my first comp last season was. I don’t have problems with defence per se, It’s just when people do it when they don’t need to.

-PS: I’ve been going to team meetings and events since 2015, and have been participating since 2019. Everything I put in my guide comes from my experiences at comps, when observing new drivers. If you have plently of experience driving, and are fairly successful, then do whatever works for you. I just shared what works for me (3 blue banners in 2 years) as the driver. I’ve seen too many timid drivers, too many overly agressive drivers, and too many drivers that are in the wrong place at the wrong time.

3 Likes
#7

The defense thing really depends, try and do whatever will make a bigger score difference. For example, if you are going to score like 2 cargo in a match, it might be more beneficial to play defense if you can prevent the other alliance from scoring more than 2 cargo.

4 Likes
#8

I think the most important thing may be driver Practice.

I can remember being involved in some pretty good robots that never lived up to their competition potential simply due to a lack of time on the sticks or spreading the driving experience too wide. With no bag/ship day this (getting practice time) is more realistic now than in the past. Still, I definitely witness robots at regional events that don’t seem to be driven to full potential. I believe it is mostly due to a lack of stick time.

This brings up another topic in some ways, but I remember once being involved in a team where there was a push (valid in some ways) to give many students the competition experience by constantly switching drivers. While well intentioned, it quickly became apparent that it was not the right move competition wise and in a way unfair to the entire team.

Other point: “Don’t be afraid of the robot”. It may be better to say learn and know the limitations of your robot and drive up to, but not beyond those limits. Obviously, the goal is to build a robust machine, but inevitably there will be some limitations no matter how good your design is. Knowing how to utilize the strengths of your team’s machine while protecting and minimizing its vulnerabilities goes a long way toward success in my opinion.

This is an interesting thread. I’m interested to see what teams say works for them.

13 Likes
#9

I remember in 2020 at Arkansas Bomb Squad mentors came by our pit and rubbed the pneumatic tires on our kop chassis and said that is good wear, we like that, it means you have been driving.

12 Likes
#10

How do you pick your defense bot?

I have seen so, so many teams keep trying to score when they should be practicing defense in quals. That’s another thing - driver practice should mostly be with another robot playing defense, and they should also switch roles and practice it from time to time.

5 Likes
#11

Does this happen often? I’m only familiar with the opposite; teams with robot cycle times 4x times as long as alliance members insisting to pick up cargo and shoot rather than picking up opponent cargo and trying to bother and prevent the opponent from scoring.

13 Likes
#12

We had to DNP a few teams this year for this exact reason. If you’re refusing to play defense or transition to defense when it was discussed as an option when it’s in the alliance’s best interest, how can we trust that you’ll have our playoff alliance’s best interest at hand?

5 Likes
#13

Right, what I’m saying it’s more rare to see people refuse to play offense and default to defense.

2 Likes
#14

At least in my region, I think a little bit of a team’s soul dies when you ask them to play defense (for a majority (citation needed) of the teams). Anecdotally, the amount of time and effort (sunk cost and labor) spent designing and mfg’ing a manipulator for the game piece, makes it hard to ask the teams to play D.

Not saying it’s wrong to ask, I just think it’s difficult.

8 Likes
#15

There are a couple of teams here in the KC region that plan on and design for defense. They love it and feel it is their niche. Though I did overhear at Heartland during quals last season, “IF another team tells to go play defense we are going to ignore them and show our scoring” I felt that maybe more than one team mishandled match strategy and dictated rather than negotiated.

2 Likes
#16

^^ This. FRC is fundamentally a game of resume building for 99.9% of teams in a season. At any given event only the top 8-10 teams really control their destiny and most of those teams are usually comfortably stronger than average in a given year. If you’re not one of those teams, you need to be ready to flex those specialist coming-off-the-bench skills and nail them.

Good defense is an exceptional way to climb the pick-lists of those 8-10 teams. If you aren’t practicing defense at every reasonable opportunity, you’re giving those teams reason to pick someone else. Good, clean defensive driving is going to be a deciding factor in most games and the top teams at every event are going to be scouting for it.

7 Likes
#18

It is difficult to stomach that decision absolutely. In 2018 the team was repeatedly the defense robot despite a lot of effort going into the mechanisms. 2019 was a little better, but for most of the official season we were tasked with playing defense despite our best efforts to improve as scorer and we did our best to be good at just two things. The team had drastically improved from 2018 but was still labelled as a defense robot at the Championship and that was hard for the team.

Those years were key motivators for our team to turn our priorities around and refocus our technical abilities.

Your point is one of the hardest things for a person to learn in life. Sometimes you pour a lot of time, money, or effort into something and it doesn’t pan out. Learning when to walk away and pick yourself back up is a better lesson to learn than continually trying to make it work.

4 Likes
#19

Looking back on the first time I drove a robot to now, it’s important to remember that nothing comes quickly. Teaching drivers on strategy and knowing the field, as well as knowing your opponents takes time to learn in depth.

On the topic of defense, it’s a very useful tool to learn, if you dont know how to defend or counter defend, your missing a large skill set in the task of driving a robot, but as I said learning takes time. The best time for you to learn how to defend or get yourself out of situation where your being defended, is either learning it while your doing driver practice outside of comps, or in quals when your robot decides it doesn’t want to run correctly. You can be a valuable asset if you learn defense, just watch out for accidentally putting your intake into other robots (knowing from experience)

1 Like
#20

I’m also gonna disagree on the defense point. Having driven a very capable scoring bot in the 2022 season, I still played defense on many occasions even when offense was the primary strategy because it was advantageous to play defense. If you are slowing down your opponent more than you are slowing yourself, you are gaining an advantage. It takes just a second for you to make your opponent miss two shots or rotate their intake away from cargo and knock it away. I took advantage of opportunistic defense and it significantly disrupted the opposing alliance while wasting very little of my time. Sometimes, there isn’t easily accessible cargo in which trying to score cargo is an inefficient use of time, but playing defense makes sense and is the best use of time. Additionally, drivers that know how to play defense are usually less susceptible to defense. I can’t remember a single time I got significantly hindered by defense in the 2022 season including champs because I know what the defenders want and didn’t let them get what they want. I’d be much more afraid of going against a bot that can score 16 cargo while throwing my alliance off their feet than a bot that could score 20 cargo but leaves us alone.

Additionally, one thing I kept on seeing this season is teams being afraid to drive aggressively. Obviously being in control of the robot is important and how aggressive one can be is very game dependent, but too many teams (including high-level teams) geared for high free speeds this season but yet didn’t drive them anywhere close to their top speed, nullifying the advantage of gearing fast in the first place. Rapid react heavily favored driving aggressively.

The teams that failed to take full advantage of their drivetrain’s capabilities were very suspectible to defense due to allowing the defenders to catch up and reach them. All the nice scoring mechanisms a team builds are worthless when a defender is on your bumpers controlling your actions. In contrast more aggressive drivers were less susceptible to defense due to forcing the defenders to react to the offense bot, not the other way around. Driving at top speed makes it difficult for defenders to catch up and defenders have to react to your changes which gives you more distance between the defenders and enough breathing room to fire off shots safely or make it to a protected zone.

2 Likes
#21

Ummm

But in all seriousness we don’t even look at OPR for anything beyond having fun looking at our OPR vs our actual point contribution.

In our experience it is usually the more confident drivers who are willing to switch to defense not the less experienced ones. We also have respect for teams who come to us and just say they want to play defense, if you know your limits and come to the strategy meeting with the best strategy for you and the alliance you will shoot up our priority list.

11 Likes