This year team 1640 is going with 4 wheel independent steering. This is one of the modules. 4 window motors control the steering and each module is driven by a CIM. Magnetic absolute position sensors provide the feedback. Our programming team is having a rough time. Mechanicals are done on the drive system.
I’m curious what drove the decision to drive each one by an individual CIM and also to use all 4 window motors. Seems like your motor allocation could have been better used compared to other crab/swerve systems. No?
The module looks very robust, I like it.
I agree. It could sort of depends on the needs in other places.
I can’t really vizualize the size I’m looking at here, is that 25 chain and a 4 inch wheel?
Nice job though, the module itself looks really nice and compact.
Why did you build a coaxial module if you planned on independently powering the wheels anyway?
I like the design, I’m just curious what decisions were made and why. Maybe you had experience or wanted experience with coaxial swerve, I dunno. Maybe it’s part of your robot’s awesome features we can’t see yet
I love it! Independent steering gives some quite interesting abilities to a coaxial swerve system.
Nice work, and I can’t wait for the pics of the whole system!
looks like #35 chain and a larger wheel to me, planning on going over the bump with this?
The wheel is a 4" Andymark Plactraction and the chain is #35. The wheel dead axial is a stud with nuts on both sides of the plates which provides more stiffness to the cage. We put ball thrust bearings on the bevel gear shafts to control the thrust forces. There are roller needle bearings for the vertical bevel shaft. The entire unit rotates on a combination of thrust bearings and ball bearings. For the pivot purists, having 4 steering motors allows proper wheel angle setting when turning. For chassis rotation the correct angle and velocity for the inner wheels is different than the outer wheels. We choose the window motors because they are the only way to get 4 matched motors and we wanted the FP motors for the ball processors. The kicker is pneumatic. Getting the gears and angle sensors set up on the window motors took some effort. We are using Cherry Electrical AN8 sensors and they seam to work well. The actual area on the robot taken up by the pivot assembles , steering motors and cims is relatively small. As some teams have found out the window motors are a little weak to drive 2 pivot assembles. We are driving ours 1 : 1. The programmers are having a tough go but on Tues. things were going better. This is our first try at this and it has been allot of work. Hopefully the agility will make the pivot drive worth it. And yes it drives very well over the hump.
We built a prototype using VEX parts this summer. This design is an outgrowth of those efforts. You can see more details on our website under DEWBOT VI Drive Train and the control under Steering There are two good papers that are linked to on the steering page that describes the math to make the wheels track smoothly.
As Gary pointed out the wheels don’t use the same angle on the inner and outer wheel base. The four steering motors allow us to do both “Crab” and “Snake” drive steering. It also allows us to treat any side of the robot as the “front”. Yes, it’s a ton of motors, but we think the high agility will allow us to be effective players.
The engineering on the pivot along with the extra bearings is to support a full weight of the robot in a crash and bash environment. Last years robot spent very little time in the repair bay, we are hoping that will happen again this year.
Moving the code from last years base to this years has been a problem for the programming team, refactoring isn’t easy. They have the drive working to where it will drive smoothly over the bump. If we ever get back into the shop (snow day here) they will so some tweeks.
The decision to use (4) independent drive and steering motors ((8) drive motors total) was based on the desire to operate in several different pivot modes and to be able to also incorporate an overt chassis “twist” ability into Crab mode.
See http://wiki.team1640.com/images/4/46/Pivot_-_Crab_with_a_Twist.pdf as a reference.
Wheel diameter is 4". Chain is type 35.