I understand what you’re saying, but wouldn’t it still be more beneficial to use what a driver is already familiar/comfortable with?
Given any particular driver, with the same robot and the same amount of practice time, wouldn’t their skill be higher if they’re building from a place they’re familiar with versus completely relearning how to drive?
Of course I can only speak from my own experience as both a driver and a coach on 279, but I’ve learned that what worked best for the driver before me wasn’t what worked best for me, and that, in turn, wasn’t necessarily what worked best for the driver after me.
When I drove a mecanum bot, my preferred layout was on a logitech gamepad (about a $20 controller), where the left thumbstick controlled translation and the right thumbstick controlled rotation.
To answer both of these questions: not necessarily.
Driving a robot can be like learning how to type. At first most people are most comfortable doing hunt-and-peck. Comfortable, but remarkably slow even if you’re the best in the world at it. Learning to touch-type is harder to startwith, but your ultimate typing speed for the same practice time will be much faster.
We rarely build mecanum robots, so I’ll use a tank-drive story that has repeated at least three times over. The driver wanted to use a single joystick to control the robot. They drove a robot for a while using a single stick, or sometimes ‘twin stick tank’ where one stick controls one side of the drive. Then we switched to ‘fps style’ with a gamepad. This splits forward/reverse to the right stick and turning with the left stick. By splitting up the major axes of movements to different input devices the drivers were able to cleanly execute their core driving moves: driving straight, turning in place, and arc turns. It took a little practice for most, but their proficiency quickly overtook their prior control method. While the starting point of the learning curve was reset their learning rate and performance ceiling increased so much that the end outcome was superior.
Taken a step further: I opine that a mecanum (or any holonomic) driver could be highly successful with a gamepad setup like: left stick forward/reverse, right stick left/right, analog triggers to rotate (or mix and match however you want, but 1 DoF per input device/stick/trigger set). They’ll be able to make crisp, clean, maneuvers without accidentally mixing in an axis they didn’t want. It will be weird at first, but faster in the end. There won’t be mistakes like accidentally adding in a little turning because your thumb slipped a tiny bit while trying to drive forward only.
Some teams (like 1678?) split up control axes by using steering wheels and pedals. ‘Twin stick tank’ splits up control differently than I like, but can have some of the same benefits as ‘FPS’ layouts, teams like 330 have been highly successful with this scheme and specialized hardware (IIRC they used two chopped-down joysticks, avoiding the slow/clumsy inputs of being ham-fist on two whole joysticks).
That’s the idea, but not the execution. We have a sort of box built around the base of two Logitech Attack 3 joysticks that acts as a palm rest. Rather than moving the joystick relative to your body with you entire arm (and occasionally upper body), you’re moving the base of the joystick relative to the driver station with your wrist and fingers. It’s like driving a car with your heel on the floorboard vs just floating your foot on the pedals. One approach gives you much more precise control and cuts out variables.
If you’re running mecanum, I would say a good gyroscope is almost a necessity.
Our team had a really good experience with the Kauai Labs NavX MXP when we ran mecanum in 2016. Easy enough to set up and worked consistently well over the course of the year.
Once you have your gyro set up, the best drive system we found is field-centric with robot motion on the left thumbstick and rotation on the triggers, leaving the right thumbstick open to control some other mechanism of your choice. It’ll take a bit of tuning to get the robot to stop rotating in a way that feels natural and to stop it from deviating too much from or overcorrecting for differences between where it is currently pointing and where you want it to be pointing, but getting it up and running is incredibly rewarding.
While I don’t think it will ever really be the best drive system, mecanum offers a lot of mobility for a pretty low amount of investment, and as long as you’re SURE your team is willing to put up with the downsides for a given game, it can be an incredibly rewarding project.