Our team had discusstion over a rectangle base (like 2729) but mounting the swerve module on the 2727 square in the middle to make our bot narrower but still be able to fast turning (we assumed that a rectangle base with modules on the corners would slow down the rotation, although we have not tested it yet) What would be some Pros and Cons for this design? As I have never seen a bot going with this approach. We are not sure about its statability and ability to go up to the charging station.
2 Likes
Wouldn’t be too unheard-of. I can think of a couple teams offhand who operated a drivetrain smaller than max size with some overhangs.
That said, you’re talking 2" on a relatively small frame. I don’t think you’d have all that much of an issue with turning anyways, particularly given that you’re already running swerve (which can effectively nerf any turning scrub a 6WD would have).
I don’t see any issues at all, except possibly getting stuck trying to get onto the charging station (low-centered)–but there’s ways to spot that early and deal with it.
You do not need the modules to be on a square to turn quickly. You may be thinking about how a narrower west coast drive turns worse than a wider one does. The same principle does not apply to a swerve drive. Your wheels will turn into a diamond shape when you want to turn, and that diamond will be shaped appropriately for wherever the wheels are relative to the center of rotation of the robot (assuming you’ve programmed it correctly).
Once they are in that diamond shape, the bot will be able to turn very quickly. Consider if your robot is geared to15 feet/second linear speed. Your wheels are sitting on a circle whose diameter is about 76”, and whose circumference is about 20 feet. So you’ll be able to make a full rotation in about 1.3 seconds, which is reasonable fast.
Note that moving the wheels around by a couple of inches between rectangle and square makes basically no difference in this equation.
As for the downsides of moving your modules away from the edge: there is one very big downside to this. The smaller the rectangle is made by your wheels, the more likely your robot is to be tippy-toppy. If this is your first time doing swerve, you need to be cognizant of this fact and design for it. In particular, consider driving full speed in one direction, and then your driver quickly turning the joystick 90 degrees. This creates a lot of momentum which wants to tip over what is now the “side” of the wheel. If you’ve got a big arm or elevator to solve for this year’s challenge, and your center of gravity is quite high, you’ll very likely have this problem. Keeping your wheels as close to the edge as possible mitigates that.
I’m sure someone out there can explain the exact math behind “tippy-toppiness”, but ultimately it is driven by a few main factors:
-The height of the COG (higher is worse)
-The horizontal distance between your COG and each edge of the rectangle formed by your wheels (smaller is worse, so you want a big rectangle with the COG roughly centered).
-Your acceleration (higher is worse)
Each of these factors may be enablers of some other aspect of your approach to the game (such as a small frame helping to fit on the charging station), so this is a good example of an engineering trade off that you will have to make.
1 Like
I’ve coached three students driving four different swerve robots now, going on the fifth. In my experience, swerve never turns too slowly; if anything, you’ll need to slow down the turning for the driver to make it easier to control (or implement a stronger exponential scalar if you don’t want to lose the option of max speed).
I believe you’re solving a problem that doesn’t exist. I would put the modules as far out as they can go on the frame for stability reasons and try it.