Hello teams
At conpetition, do you use one or two drivers who control the robot? Also, how do you train the drivers to work well together and what are the pros and cons for having one or two drivers. Any feedback is great!
We have always used two drivers. In our opinion, it is easier to have one person, the driver, operating the movement of the robot; while the other, the operator, operates all mechanisms. This spreads the workload over two people.
The pros of having one driver is that there doesn’t have to be as much communication. If the controls are simple enough, one driver would suffice.
It’s also better to get that extra member out there just for the experience of driving the robot. From my perspective, driving at competition is one of the best experiences I’ve had in my lifetime. It’s an awesome feeling being able to compete, especially at worlds.
This school year our team competed with 2 robots.
The first a Recycle Rush robot had 2 drivers. One to drive the other to control the lift. It allowed the lift to be controlled more easily while driving.
The second a Stronghold bot had 1 driver. We found it was easier for a single driver (myself) then for 2 people.
I say do what works for you and you can never be wrong.
For our regular competition season we used one driver. It worked alright, except the accuracy and speed of boulder intake/outtake left something to be desired. Thus, we will be trying out 1 driver and 1 operator at our offseason events. Essentially one person will drive and line up shots and the other will intake and shoot boulders. I personally think that will work better as there is less thought involved on the one person, but to each their own.
We’ve found that you may want to give the shooting to the main driver. This is because the driver knows when they are ready to shoot and are done moving.
I would say a single skilled driver capable of controlling the entire robot will always be quicker and more reliable than two half-skilled drivers. There are basically two ways to accomplish this: Driving practice or code complexity. If you allocate more time to driver practice, your driver MAY eventually be able to control an entire complex robot by themselves. I say MAY eventually as this is a pretty difficult task.
To compensate for driver skill, or if you have an exceedingly complex robot, you will need to automate some tasks to lighten the burden on the driver. Perhaps instead of controlling 3 different actuators with 3 different inputs that result in a shooting motion, you have a single “shoot” button that makes the life of the driver much easier.
Both driver practice and complex software will require a lot of time to get good, so you need to manage your time well during the build season and be realistic about what can be accomplished. This means finishing the robot extra early, and potentially building two robots so your mechanical folks can keep working/tuning the primary bot while your coders/driver works on the secondary bot.
So how would you code multiple drivers for one robot??? We used one driver this year but we only had three buttons that were used.
We only use one driver (me). It is simply a lot faster for me to do everything, than trying to communicate what I am trying to do with another person, and tell them exactly what to do.
But I can only say do what your team finds best for you guys and your bot
We change it up based on the robot’s functionality. In 2015, our robot was simple so we only used one driver, since the only mechanism on our bot needed to work well as the driver moved around the field. This year, we used two, since our shooter mechanism worked independently of our driving.
Basically, if a feature or mechanism needs to work at specific times while moving (intake or A-defense manipulation for example), our base driver gets control of it. If it works independently (e.g. shooter or hanger), it’s given to our copilot.
We use two drivers and have found it to be very reliable for us. While two drivers will never be able to communicate as well as one driver, there are often a lot of tasks that need to be taken care of at once.
I have found that, by using a two-driver setup, I can focus on the task of getting around defense and not worry about forgetting to turn the shooter on or raising our arm over defenses.
There is also an objective bonus to using a two-driver setup: you get more analog inputs. This isn’t necessary for every robot, but most controller setups are limited to two analog inputs (one for each thumb). Some manipulators really like having analog control, and you probably want analog control for your drive train too.
While it might seem like two-driver is madly inefficient because of communication issues, a little practice can seriously mitigate that, and makes a lot of complex multitasking maneuvers much easier.
Remember too that while technically having one guy do it all can be more efficient than two people working together, the advantage of the latter is that you’re teaching at least two of your team members to work together. There’s serious value in learning how to work well and communicate effectively with other people, learning to keep their cool in heated and stressful situations, and so on.
What we did this year was had the main driver drive, use the breach arm, and shoot. The second driver was intake up down, and actually Intaking, also he had for a brief time the capability for climbing.
If you only use three buttons, one driver is usually best. If your robot has multiple continuous manipulators and little automation, you may want to add a second driver (or gunner). In this case, the drive controls will be on one joystick/controller and the manipulators on another.
If anyone wants to know. I attached a image file of our controls this year. While it seems like a lot, I had no problem keeping up. But then again (no offense intended to anyone at all) not every person is capable of multitasking etc… I simply just find it a lost easier to have only one driver. Because I do a lot of stuff “on the spot” and it would be impossible trying to explain to another person (like a shooter) what I’m about to do…
I’ve found in my experience that a little faith in your drive team partner can go a long way. Our team uses a few keywords to communicate. “Acquire” means we’re going to go intake that ball. “Raise” means raise the arm. “Fire” means… fire.
I’ve found that most of the time I can just rely on my partner to do the right thing “on the spot” so long as we communicate about what our goals are. If we’re stuck on the moat, for instance, and I we need to swing our arm in such a way that it shifts our CoG forward, a simple “we’re stuck” will suffice, though usually I don’t have to say anything in the first place.
The notion that making snap judgments on the spot is impossible with two drivers is a fallacy. (See 330 righting themselves on Einstein if you don’t believe me)
That doesnt look too too bad. Im the driver and I can most likely keep up with that. Using a joystick with that many buttons would be SUPER confusing.
If you are looking at switching to two drivers, you have to know that they can communicate. We used two drivers this year and had some matches where my manipulator knew exactly when I would be lined up to shoot without me even telling him to shoot. That in sync behavior is what you need, very similar to sports. However during some matches we would shoot before we lined up due to bad communication. Another huge point for two drivers is that if something went wrong, our manipulator would have overrides to the limit switches and things like that. As a driver, it would be a nightmare to try and switch to override and manually control the device being used. Also just the amount of buttons was getting ridiculous for us and they would not all fit on the driver controller.
To the best of my knowledge, Lightning Robotics has never fielded a single-driver robot in our 15 years of existence:FIRST is about the Inspiration, after all, and limiting the drive team to a single student-- especially if your reason is to avoid the valuable learning experience of working closely with your peers in a high-pressure environment-- strikes me as a rather uninspiring policy to have.
…With that said, we usually divide the labor as follows:
“Driver”[ul]
[li]All drivetrain, all day![/li][li]Closely related subsystems (like ramps/wedges, whose entire purpose is to help the robot to navigate the field obstacles) are sometimes bundled in as well.[/li][li]Interface is always composed of full-sized joysticks (one or two, depending on the style of drivetrain).[LIST][/li][li]I don’t care how much of a hardcore gamer you may be; a thumb stick on a hand-held game pad will never offer the kind of fine-grained control that you want for an FRC drivetrain, period.[LIST][/li][li][EDIT] lol, I knew that I’d get splashback for this assertion![/li][li]I acknowledge that many students, having grown up with access to game consoles for their entire lives, can indeed reach great heights in terms of skill with a gamepad. With practice, a good driver can learn to compensate for the robot’s inertia, resulting in a very competitive performance.[/li][li]Regardless, I stand behind my original statement. The difference in scale between a robot and a joystick is significant enough as-is, but at least a joystick still gives you some sense of the robot’s inertia. There’s a good reason why gamepad thumbstick extenders are a thing: range of movement is kind of a big deal, and thumbsticks have very little of that. [/EDIT][/ul] [/LIST] [/li][li]Above all else, we depend on the driver for consistency; on top of the usual selection requirements of driving skill and cooperative attitude, the ideal driver is completely impervious to pressure and distraction when they are “in the zone” during a competition match.[/LIST]“Co-pilot”[ul][/li][li]Controls everything else: arm, intake, shooter, what-have-you.[/li][li]Back in the days of serial ports, the co-pilot’s interface usedto be custom-made: an array of knobs and buttons chosen specifically for that year’s subsystems.[LIST][/li][li]These days, we usually just give the co-pilot a handheld game pad and a printout of the button mappings.[/ul] [/li][li]The co-pilot is expected to multi-task like a champion, managing several subsystems simultaneously; they must also exhibit exceptional adaptability, in order to cope with mechanism failures and improvements changes made by the programming team in between matches.[/LIST]“Drive Coach”[ul][/li][li]Pretty much self-explanatory…[/li][li]Studies the large-scale goings-on around the field, coordinates with alliance partners, and issues high-level commands to the drive team. Examples…[LIST][/li][li]If the clock is running out: “One more shot, make it count!”[/li][li]If we’re defending against a powerhouse, and a ref starts counting down for a pin: “Countdown: 3… 2… 1… back!”[/li][li]If we’re crossing a defense, boulder in tow: “Batter shot, left side!”[/li][li]If an ally gets stuck on a boulder: “Hey [####]! You need help?” [/li](And if the answer is yes) “Ok! After this shot, we go help [####].”[/ul]
[*]Selection criteria: preferably an alumnus with competition driving experience. Alternatively, someone with experience coaching or refereeing any sport will also do nicely.[/LIST]
Let me give some context for our situation. SO we are a small team. 13 students, but only 3-4 that show up on a constant basis. Every build season we barely finish the bot in time for bag day (in the case of last year, not even finish, and have to finish the thursday of our first regional. We don’t have time to actually practice on the bot before our first regional, and definetly don’t have a practice bot. That in addition to the fact that I’m the only person on the team (not bragging here, just facts) that has the required skills to be a driver, makes it simply not possible to have a 2 drivers that communicate flawlessly. Don’t get me wrong. I love my team members, and we know exactly how each one of us thinks, and i trust them 100 percent (especially us 4 that show up constantly), but 2 drivers is just not in the question 
I’m curious to what particular skill you possess the others don’t.