When coding odometry for an x drive, how does an arc approximation work? An x drive almost never moves in arcs. Wouldn't adding vectors each cycle be a more accurate solution?
While it looks like the robot is driving in a straight line it is realistically impossible that it would be from imperfect hardware and electronics. Therefore, we can assume the robot is driving on an arc of a circle with a very large radius and the calculations still stand very accurate. Since we are measuring these distances over very small time intervals you can imagine that circles even with radii of 6' could have a segment that satisfies the measurements we just recorded from the tracking wheels. If none of this is making sense you could also just open an image of a circle and zoom in really far on the outline to see eventually it appears as a straight line but in reality is an arc segment. Hope this helps.
@@Zach929U yeah that makes sense, just was wondering because the x drive moves so fundamentally differently than a tank drive and the code looks tailored to a tank drive.
Wait, my team is making a a tank drive and we are trying to figure out the placement, but we plan on only using 2 odometry wheels, like a differential bot. So our odometry wheels have to be on the midline right?
Hi, if your team plans to use only 2 odometry wheels that are parallel to each other, you would only be able to calculate the orientation vector. Without a perpendicular tracking wheel, you lose out on the ability to calculate the position vector (x and y coordinates). I go into more depth on how that's done in Part 3 of the video here: th-cam.com/video/qqODIdvSGac/w-d-xo.html. If your team is unable to use 3 tracking wheels, my recommendation for a 2 tracking wheel system is to position them perpendicular to each other, this way you can still calculate the position vector, albeit you lose out on orientation. To find the missing orientation, the inertial sensor is a great choice that takes up less space while providing data with similar accuracy. As for placement, if you go with the first option, your tracking wheels do not have to be on the midline as long as they follow the 3 rules outlined here: (3:12). That is, they must be parallel to each other and perpendicular to the tracking center. Hope that helped, let me know if you have any more questions.
Hey everyone, sorry for keeping you all waiting. Part 3 is out now: th-cam.com/video/qqODIdvSGac/w-d-xo.html. Enjoy!
Incredible video, keep up the good work man!
Thank you! I'm glad you enjoyed it.
You are so good at teaching! Everything was understandable and incredibly helpful!
thank you for making this! as a member of an FTC team, it’s so hard to find resources for more advanced coding help
I'm glad it helped. Good luck this season!
Amazing job with teaching !
Thanks for the kind words!
Awesome bro...
I loved the way you explained it...
Thanks, more videos to come!
Does the placement of the tracking wheels change where the tracking center is? Or is it always at the center of rotation of the robot?
WE NEED PART 4!!! Pls
Love the video. Part 3?
Part 3 is out now! th-cam.com/video/qqODIdvSGac/w-d-xo.html
Great Vid! Part 3?
Thanks! Part 3 is out now! th-cam.com/video/qqODIdvSGac/w-d-xo.html
For the distant from the tracking wheel to center point is it the center of the wheel or the closest edge of the wheel?
TYANK YOU!!!
what if delta s is outside the circle and delta left and delta right are on the circle? do we have to calculate this case ourselves?
When coding odometry for an x drive, how does an arc approximation work? An x drive almost never moves in arcs. Wouldn't adding vectors each cycle be a more accurate solution?
While it looks like the robot is driving in a straight line it is realistically impossible that it would be from imperfect hardware and electronics. Therefore, we can assume the robot is driving on an arc of a circle with a very large radius and the calculations still stand very accurate. Since we are measuring these distances over very small time intervals you can imagine that circles even with radii of 6' could have a segment that satisfies the measurements we just recorded from the tracking wheels.
If none of this is making sense you could also just open an image of a circle and zoom in really far on the outline to see eventually it appears as a straight line but in reality is an arc segment. Hope this helps.
@@Zach929U yeah that makes sense, just was wondering because the x drive moves so fundamentally differently than a tank drive and the code looks tailored to a tank drive.
@@angrybokoblin8777 In case you still are curious, a straight line is just an arc with infinite radius.
is the delta left and right just the distance that those tracking wheels had traveled?
keep up the work
haha, will keep that in mind
Amazing video! when will the next video be out?
Next video is out now! th-cam.com/video/qqODIdvSGac/w-d-xo.html
Wait, my team is making a a tank drive and we are trying to figure out the placement, but we plan on only using 2 odometry wheels, like a differential bot. So our odometry wheels have to be on the midline right?
Hi, if your team plans to use only 2 odometry wheels that are parallel to each other, you would only be able to calculate the orientation vector. Without a perpendicular tracking wheel, you lose out on the ability to calculate the position vector (x and y coordinates). I go into more depth on how that's done in Part 3 of the video here: th-cam.com/video/qqODIdvSGac/w-d-xo.html.
If your team is unable to use 3 tracking wheels, my recommendation for a 2 tracking wheel system is to position them perpendicular to each other, this way you can still calculate the position vector, albeit you lose out on orientation. To find the missing orientation, the inertial sensor is a great choice that takes up less space while providing data with similar accuracy.
As for placement, if you go with the first option, your tracking wheels do not have to be on the midline as long as they follow the 3 rules outlined here: (3:12). That is, they must be parallel to each other and perpendicular to the tracking center. Hope that helped, let me know if you have any more questions.
hey man, when is part 4 coming out, I see ur responding to people here so I thought I could ask to see if I could get a response
You are the best
When will part 3 and 4 be released?
Part 3 is out now! th-cam.com/video/qqODIdvSGac/w-d-xo.html
Part 4 will most likely be released in the coming weeks
When is part 3 coming out!
Part 3 is out now! th-cam.com/video/qqODIdvSGac/w-d-xo.html