(upbeat music)

Hi, I'm Jason McKenna, Director of Global Educational Strategy with VEX Robotics. Welcome to our Moments in Engineering series, where we talk about engineering concepts that apply to both classroom and competitive robotics. We talk with VEX engineers to really bring these concepts to life.

Hi, I'm Art Dutra, Director of Mechanical Engineering here at VEX. Today, we're going to talk about wheel scrub. Most robots are built with four wheels for stability, but they don't always have the same dimensions. You're also not always driving on the same surface, and you have different types of tires to choose from. We have traction wheels, Omni wheels, and slick wheels. Even among traction wheels, there are different types of traction.

This is a standard robot that has all traction wheels. The force when the wheels are spinning is only exerted in the direction the wheels are spinning. So if you drive forward, all the wheels perfectly spin forward, and the robot goes forward. In the other direction, it goes backward. Unlike an automobile where the wheel can pivot, these wheels have to go forward and backward to turn. This creates a situation where the robot wants to tear itself in half. Luckily, the robots are strong, so instead, the wheels slightly scrub at an angle.

The angle they scrub at depends on all the wheels on your robot. There's a center point in the robot between all of those. In this case, with four wheels on the robot, the turning point will be right in the middle, between all of the wheels. You can basically draw an X right there in the middle. This is what people might refer to as a wide robot. The wide robot has wheels that are close together but far apart, making it wider than it is long. If the robot had wheels positioned differently, it might be called a long robot.

In general, wide robots can turn easier. This happens because when the wheels try to spin, they go at an angle relative to the turn. Most of the force is still trying to make a turn and isn't wasted. In contrast, a long robot has a more severe angle, causing most of the energy to scrub sideways. Whether a robot can turn depends on the traction or friction between the wheels and the floor. For example, on ice, there's very little friction, so even oddly sized robots can turn easily. On carpet, which has high friction, turning can be difficult.

Generally, wider robots with all traction wheels can turn easier than long robots. There are ways to improve turning for long robots. One method is to swap out some of the traction wheels for lower friction wheels. For instance, this robot has Omni-directional wheels or, in the case of VEX GO, slick wheels. The slick wheel is all plastic, has very low friction, and functions similarly to the Omni wheel.

Thank you for joining us in this episode of Moments in Engineering. We hope you found this discussion on wheel scrub informative and helpful for your robotics projects. Stay tuned for more insights and tips from our VEX engineers.

This Omni wheel has rubber wheels, so when it goes forward, it gets normal traction like a regular wheel, but it has all these side rollers on, so it has no resistance to going sideways. In this case, the robot can very easily slide sideways. Because these wheels have no friction going sideways, instead of the turning point being an X in the middle of the four wheels, it's only between the two friction wheels, right here. Oh, okay. So this robot will turn easier.

In this case, because the center of turning of this robot is between the two drive wheels, the directions when these go, the forces of the wheels are exactly in line with the tires. That's really interesting. So in this case, there is no wheel scrub at all if you have Omni wheels on two of the wheels on your robot.

If I was first thinking about designing a robot, if I was trying to design a robot for my classroom, or if I was trying to design a robot for my competition, if I was thinking about the drive terrain first, I could think about, "Oh, I wanna build this drive terrain because I wanna add this stuff on top of it." But if I'm not thinking about something like wheel scrub at the very beginning, when I actually begin driving my robot, I might find it's actually difficult to turn and do those types of things. So it's important to keep this into consideration as I'm first designing my robot and think about how I wanna build the actual drive of my robot.

Or what would be a good way, you think, to almost investigate this and test this? Because one of the things I hear you talk about in engineering all the time is this idea of trade-offs. So I might wanna use something like an Omni wheel to eliminate wheel scrub, but then maybe it's more difficult to code a robot with an Omni wheel or things of that nature, or whatever those particular trade-offs might be. So how do I balance eliminating something like wheel scrub while also trying to accomplish the other things I wanna do with my robot? That's a great question.

One of the things that can happen when you put Omni wheels on your robot is that if you're trying to drive perfectly straight, because there's lower friction here to turn, your robot can actually drift easier. It has an easier time drifting versus if you have all the traction wheels, your robot, if you're driving straight, tends to track much straighter. So, if you're trying to do long autonomous runs, having all traction for going straight can help you. However, if you have a robot with a lot of wheel scrub, that can be inconsistent in turning, depending on what floor material you're on, whether you're on a consistent tile or changing onto a ramp or a platform, you can have inconsistent results trying to turn, so there are some trade-offs there.

Another trade-off to keep in mind as you're working and trying to decide, "Do I want a wide robot or a long robot?" might also depend on the competition and what mechanism you're trying to build. For example, a wide robot for VEX IQ, if you want a really wide intake like fling, it can be as wide as possible. That basically constrains you more towards this wide robot. But let's say in VRC, you're working on a four-bar or double reverse four-bar linkage, you want more length so you can have all the room for those linkages that go through your robot. That prevents you from doing a wide robot; you have to have a longer robot.

Thinking ahead to some of the mechanisms you're trying to build does basically push you in one direction of, "Do I want this wider robot? Do I want a longer robot?" Also, try to think ahead of how much programming am I gonna do? How much driver skills-? Yeah, coding like autonomous versus driver's skills, something like that, yeah. That's really good insight.

And obviously, with something like wheel scrub, I would imagine you would reach a point of diminishing returns, right? So you don't necessarily have to eliminate all the wheel scrub, but you just wanna eliminate as much as possible to make it successful with whatever the challenge you're trying to do, right?

Oh, 100%. 'Cause you could have a robot where, let's say, if you try driving on carpet, it has a lot of wheel scrub. And a lot of times, if you will see robots with high wheel scrub, if you tried driving them, they either might stall out or they might hop around or get bouncy. But then you realize, "Hey, I'm in a competition. I'm only driving on the VEX IQ tiles. I'm only driving on the VRC foam." It's good enough in that situation. So there are certain applications where like, "Hey, in the competition, it's great here, but hey, if I tried driving in the carpet in my classroom or something, it doesn't quite work there, but it works in the field."

I'm thinking about friction, right? So this could be like a good introduction to students or a good way to talk about friction with the students.

Exactly. Yeah. That's a great point. 'Cause the wheel scrub is heavily dependent on the friction between the wheels you have and the floor material. So you can try doing tasks of the same size chassis with different flooring materials to see like, "Hey, between all these different materials, how well does the robot turn?" You could also try of the same wheels tried doing different aspect ratios. 'Cause again, when you're trying to turn and this wheel's trying to go forward, but because the turning radius is here, so it's actually trying to go in an angled direction, you can actually do the math and basically calculate the difference of how much percentage, the vector math, to basically figure out how much of the wheel scrub is happening and whether that's gonna be a problem with your robot.

Got it. And in terms of the friction itself, I would imagine it'd be an interesting discussion that you can have again with students in a classroom or in a competition because friction, as we're talking about in the context of the wheel scrub, could be bad, but without friction, your robot also wouldn't move. So it's, again, with that idea of trade-offs where it's like, "We wanna eliminate friction." Like with the Omni wheels, like you were saying, "I want to eliminate friction, but then that also might cause my robot to drift if I eliminate too much friction."

Oh yeah. Cause I mean, sometimes if you tried building a robot with only Omni wheels, it's gonna turn incredibly easy, but let's say, for example, you're trying to drive up a slope or a hill or a ramp, all of a sudden now your robot might slide sideways without your control and you can't control that because there's no friction to going sideways.

I think a lot of that, like when you were talking before about running VEX IQ on the VEX IQ field, or I'm running VRC on the VRC field, what I'm constantly thinking about as you're talking about that is it's always first step is always understand what it is you're trying to do. So understand your constraints, understand what it is you're trying to understand the game or understand the classroom competition and then work from there. So talk about that just a little bit moment because as I admire engineers, how they always think very systematically like that. With students, students oftentimes just always want the answer. But there is no one answer because there's always trade-offs, there's always constraints. So talk about how you would just think about approach it from that perspective.

Yeah. So as you're designing robots, obviously between VEX IQ and VRC, there is always you're trying to design for a certain situation. You have to look at the problem at hand.

There is never one unified solution, "If I do this, it works every time." Engineering is always a series of trade-offs. It's about looking at the problem at hand and trying to come up with the best solution. Depending on your problem, the game, and the challenge, there can be changes from year to year. You have to consider, "Hey, what is valuable in this instance? Do I need to have very good friction? Am I trying to drive up a hill? Do I not want to slide sideways downhill?" Or, "Hey, do I need to put this mechanism on my robot, so I need a certain size of robot?"

It's about evaluating the game and trying to determine what is and what isn't important. From there, you optimize and make choices, either living with, "Hey, my robot doesn't turn great, but otherwise, 90% of the time it works great," or deciding, "Oh, I need a lot of maneuverability. I'm going to ignore this challenge and focus on this." That's my strategy for this game. Document all of that in the engineering notebooks. You can make those decisions, but we'll talk about that in a separate segment in our Moments in Engineering series.

So thank you, Art, very much. I learned a lot, and I hope you learned a lot. I hope you're able to apply this with either your classroom or your competition robot team, or both, when we're talking about wheel scrub with Moments in Engineering.

Thank you very much. I look forward to talking to you again soon.

(gentle music)