Single-player operator interface?

[PneumaticFlea]

My team is interested in a control system that is controlled by a single player. We'd probably still have two joysticks, one for driving and one for everything else. There'd just be one person operating both joysticks. Some people on my team advocate for this approach, but I think it's a terrible idea.

Argument for: Total control of movement and shooting from one person is the most intuitive, as anyone who has played videogames will know. As long as the controls are mapped out efficiently enough, one person can do the job just fine, and without any of the synchronization issues that a two-man driving team can and will end up facing. I know we did a two-man driving team last year, but the controls for this year is a lot more complex and requires a lot more coordination if we were to use two people, and I haven't been on the team long enough to comment on other years, so I think it might be beneficial to try one person control system or at least not be scared of it.

Argument against: As they say, two eyes are better than one. Coordination between driving and shooting isn't incredibly important. If the shooter has control to pivot, they can do their job independent of the driver. In regards to surmounting defenses (particularly opening the sally port, drawbridge, and portcullis), the amount of coordination required to manipulate the arm and control the drive train would be too much for one person to do; two people would have an easier time at it. Additionally, this avoids people having to use their left hand.

So, Chief Delphi Forums. Has any other teams tried this before? Has it been successful? Is it a good idea at all?

[Jarren Harkema]

I've seen it done both ways as a drive coach. Typically the only times I have seen it was when the robot was basically just a drive train.

It's hard to say about your specific scenario without knowing the details about your robot and its functions. If there are two logical ways to break it up. (controlling pivot/shooting and driving) as in your example, I would encourage you to practice that way and see how you like it. I know last year there would have been no possible way for us to do it. We ran out of buttons on the main drive joysticks and barely had enough on the secondary gamepad! I think it's going to be similar this year. You really need to find two drivers who are totally in sync with each other and don't argue about what needs to be done to get from point A to B.

Now you might have a rockstar driver who can do it all without missing a beat. In that case, what is the other "driver" going to do? Think about the ways the second driver can be used rather than a warm body behind the glass.

[Sperkowsky]

Well as an experienced driver I think it depends on the robot. If the robot is a drive train plus a fixed angle shooter then 1 driver is enough. but, 2 is still probably a better bet.

If your robot has a variable angle shooter or other more complex mechanisms then I think 2 drivers is a necessity.

Yesterday our 2nd driver had to go home a little early. When I was playing around it noticed it was very hard to drive forward while controlling the shooter.

One major variable is how well the 2 drivers work together.

[joelg236]

I have an interesting perspective here. Last year, we had 3 sister teams with virtually identical robots, 2 being a 3-person drive team, and 1 being a 2-person drive team.

We decided it based off of personality & skill - the drive team with 1 driver was like that because he had excellent coordination and only needs small input from the coach (who could spend time focusing on the other teams).

Overall, the 3-person drive teams did a bit better competition wise, probably because they had more heads to solve problems in a match, but certainly coordination wise it was a tie. There can be a great advantage to not having someone else to direct and communicate with.

[PneumaticFlea]

Thanks you all for your speedy replies!

The wording of 3.1.1 is ambiguous; but it seems to hint at a single driver being legal. So theoretically, you could have only one person behind the glass.

I think the idea of trying both ways is good. We don't really have a dedicated driver or drive team, and I'm not exactly sure how other teams decide who they are, but last year we just gave everybody a crack at it and whoever was most confident got the gig. Maybe this year we should have something a bit more formal, and do some tests on different configurations of the operator interface (multiple IO classes) and decide which is most efficient.

To clarify on the design of the robot, we have a tank-tread drive base, a two-wheeled shooter that pivots with a servo, some sort of manipulator arm (the specifics of which I'm not completely sure of), and a scissor-lift mechanism for the endgame. The drive team would really only be operating two of these systems at a time at most. Is this too complex for one person? Or would it require too much coordination for two people?

I think it's a good point about having more brainpower to solve problems. Joel, your situation is very interesting to me (same robot, different amounts of drivers). It seems like for most things, 1 driver might be sufficient, but 2 doesn't hinder anything and the extra brainpower can be useful. I'll convey your findings to the rest of the team.