Wow, that accually looks really good. I might spend some more time looking into this design.
1 Like
Joe - do you see an advantage for X-Drive for this year? IMHO it seems speed would be more important than sideways-maneuverability.
1 Like
X drive increases speed by sqrt 2 in non orthogonal directions
3 Likes
yes, you could zip and zoom around the feld and cycle the balls with the speed of a cheeta
1 Like
thanks all - not sure I understand this.
If non-ortho moves sqrt 2 faster wouldn’t going straight be 1/sqrt 2 slower (25%) ?
Q then is how much time is spent doing either. Some matches being able to strafe is great but not sure I see that this year.
It travels at the same free speed as a tank drive in orthogonal directions but with half the power (the vectors cancel and effectively only two motors worth of power are active).
Thanks - beyond my intellect to grasp this. If I have a wheel at 90deg & one at 45 the distance covered in 1 rev would be pi()*d vs (pi()*d) / sqrt 2 - or ~25% slower
Sure strafing it’s be faster but once a regular drive’s turned it will be 25% faster?
I saw a YouTube video seemed to confirm this - going straight seemed X-Drive seemed slower.
Sorry I’d so dense - but trying to understand
You’re thinking of the force vectors, which in this case is the inverse of the speed. It has 1/sqrt 2 the torque of a tank drive and sqrt 2 the speed of a tank drive in non orthogonal directions (diagonal directions) and 1/2 the torque of a tank drive and the same speed as a tank drive in orthogonal directions (forwards, backwards, left, and right).
3 Likes
Going straight on an x drive is slower than going diagonal on an x drive. But going straight on an x drive has the same free speed as going straight on a tank drive.
3 Likes
False, driving straight on an X drive is root 2 faster than driving diagonal on an X drive and straight on a tank drive. Imagine a line coming out of the center of the wheel parallel to the direction the wheel is facing and imagine a line coming out parallel to the direction of robot travel. Now construct a right triangle out of those lines. The length ratio between the line parallel to the wheel and the line parallel to the robot direction is the amount of extra speed you get relative to a tank drive. Picture coming soon
Edit:
7 Likes
I mean it also depends on your definition of “straight” for an x drive. The direction that goes sqrt 2 faster is where the wheels are pointing inwards at 45 degrees toward the path the robot takes.
Another way to visualize this:
Each pair of parallel wheels is a set. If you were to write a program that had the “right” set spin forward for one rotation, then the “left” set spin forward for one rotation, where would the robot be at the end of the program?
With 4" wheels, the first motion would be ~12.6" forward and to the right, the second motion would be ~12.6" forward and to the left, and the combined motion would put the robot ~17.7" directly in front of its starting position.
The same distance is traveled when the two sets spin simultaneously.
4 Likes
And another way:
If you play first person shooter games and run forward while strafing, you move faster than either running forward or strafing.
2 Likes
The problem I found with thinking about it this way is that it only works for wheels angled at 45 degrees. If the wheels were angled at 30 degrees then the drive would move faster than wheels at 45 degrees when you use this logic. and if the wheels were angled at 60 degrees then they would move slower than with then angled at 45 degrees.
But in reality (when drive inefficiencies are not included) the farther you angle the wheels the faster the drive should go.
Look at it this way: if the wheels are angled at 45 degrees and they are both trying to go at full speed then how far must the robot move in order to allow both wheels to move at this speed.
(Ik I haven’t illustrated my point very well, but this is the only way I can think of explaining it through text)
2 Likes
Point taken.
I’ll point out there is no good reason to build a 4-wheel holonomic base at anything other than 45 degrees. You want the wheels to all use the same angle relative to both strafing directions, which can only be 45.
For the specific case we are discussing, your point (in my opinion) only adds confusion.
2 Likes
it would be much easier to explain with the wheels right in front of you in a visual format, otherwise I would just have to use vectors on a sticky note which is much harder to get the general point across with.
For the perspective of straight I chose my analysis was correct @Gameoa
You simply rotated the perspective to be centric of the wheel, which is fine, but when comparing it to a tank drive, comparing them at the same perspective is better.
I was comparing it to the perspective of the whole robot chassis, not the wheel specifically