[Racer26]

Building a test bot with mecanums can be a valuable learning experience.

Put it up against your 6wd kitbot, and you'll see quite quickly why most teams refuse to use them. You lose every time in a pushing match, if you've got mecanums, and they're a tank.

IMO, the only "omnidirectional" drivetrain worth building for a competition bot is a swerve. The problem, is that the mechanical and coding complexity of swerve drives tends to make them unreliable, AND it makes building one eat a significant portion of your build season.

During my 9 competition seasons with 1075, we built:

2003: Steered track unit, unpowered trailing wheels. A bit like a snowmobile in reverse, steering the track. Allowed turning at full speed to reach the top of the ramp first.

2004: 14" bike tired 4wd, the trailing wheels articulated vertically with the arm, so that hanging from the bar was primarily a "lift the wheels, rather than the bot" motion.

2005: Didnt compete

2006: 2+2 Coaxial Swerve, All 4 wheels driven from same gearbox, using 1 turn pots for position feedback. I couldn't tell you how many pots we snapped with this robot. It was awful. For the offseason, we built the first iteration of our belted drives.

2007: Belted drives (motors inside the belts driving a shaft running the length of the unit, through the guts of a supershifter, a bevel gear, then chain to the drive cog). I believe we competed with the 3rd iteration of this drive. 4th iteration in the offseason. We kept having problems with belt breakage, many of the iterations added deflection (sometimes variable) to the bottom surface of the track.

2008: 5th iteration of the belted drives. Partway through the offseason, we switched to a more standard 6wd drop center, and walked away with BE7.

2009: 2+2 Swerve again, only this time, with encoders, and independent motors for each wheel. MUCH better than 2006.

2010: 2+2 Double Sided Swerve. Essentially the same design as 2009, only with an extra set of wheels on top, in case we flipped going over the bumps.

2011: Another iteration of the 2+2 Swerve.

2012: Another iteration of the 2+2 swerve.

Then I left 1075 to go work with 4343.

2013: 1075 built a 4wd, front two omni drive, and 4343 built the kitbot drive.

Each year 1075 did the swerve? The drivetrain ate up so much of their season, that they were scrambling to get the end effectors working properly.

Working with 4343 in 2013 using the kitbot drive? That made life SO much easier.