Thank you so much for this lecture, actually, I was completely hopeless about final grade of the robotics basics course in my university, now I can figure things out. you're teaching this lessons really good.
Good morning Angela, is there a rule for choosing the direction of the X axis while setting up the coordination frame once the Z axis is set? I am not talking about choosing the Y Axis once the Z and X axes are set: the choice is bounded due to the righ hand rule. For example at 18:33 you set the X axis "into the page", but it could have been set upwards, downwards or out of the page. This would then imply a different Y axis. Is there a rationale for your choice? regards, Jonathan Ps thanks again for the efforts you put into these videos!
It depends on whether you want to use the Denavit-Hartenberg method to get the homogeneous transformation matrix, or not. If you do, then there are specific rules for the direction of x, also. In this video, I am assuming you will get the htm by directly finding the rotation matrix and displacement vector; in that case, there is no rule for the direction of x; it's a free choice. In my new videos (see my playlist called Robotics 1 2017), I teach this topic assuming that the Denavit-Hartenberg method WILL be used, and so I teach two additional rules for assigning the X axis (it must be perpendicular to its own and the previous Z axis, and it must intersect the previous Z axis).
Good morning Ms. Sodemann, firstly: many thanks for sharing these videos. These are *very* well done and help tremendously with easy to understand examples (if I might suggest: gathering them in playlist would help tracking the sequence on youtube). I have a quastrion on the SCARA Manipulator showed at 12:57: you are adding an a1 and an a2 link. It looks to me that this is only one link as there are no joints in between them. Am I missing anything? Thanks & Regards from Germany
Hi Jonathan, You can find all of my videos in a playlist type of order at my website: www.robogrok.com At 12:57, the a1 and a2 links are in different directions; that is why they are labeled separately. a1 is in the z0 direction, and a2 is in the x1 direction.
Hallo Angela, thanks for the reply. I understood that a link is a connection between joints and therefore between the first and the second cilindrical joint there is only 1 link. So labelling a1 or a2 (because of different directions in the kinematic model) makes the model clearer, but there is still only one joint. Is my understanding correct? Thanks again Jonathan
Hi Jonathan, I'm not completely sure I understand the question, but let me try this explanation: Kinematically speaking, we could replace a1 and a2 with one link along the diagonal, and the math will come out exactly the same. So, links a1 and a2 are kinematically one link. But, when we physically build the manipulator, the links a1 and a2 will represent different physical quantities. For example, the students in my class use the parts in their robotics kit to build a SCARA pick and place manipulator like this: th-cam.com/video/xyJDUyxwD2I/w-d-xo.html and use the equations derived here in their code to control it. In this case, the link a1 represents the height of a servo, and the link a2 represents the length of the bracket connecting the servos together. In cases like this, keeping the link lengths separate can help in the task of code-writing, since the variable 'height of the servo' and the variable 'length of the bracket' will be used multiple times in the code.
Hallo Angela, I was probably too complicated in my question. I try it again: between a1 and a2 in the video ( 12:57 ) I see no joint, which physically I understand to be only one robot "arm" (how could otherwise two arms be joined if not with a joint?). So I was wondering why you labeled them separately (a1 and a2) instead of labelling only once (e.g. "a1") as there should be only one arm. Thanks again, Jonathan
Hi Angela, I am looking now at the video 1 1 4 1of2. I understand now that I missunderstood something about the concept of link: I thought that 2 links can only be joint with a joint (cylindrical or prismatic), so if there is no joint there is only one link, i.e. the robot has one arm. It looks that this is not the case: in both this and the video 1 1 4 you label the vertical link a1 and the horizontal a2. My question is probably trivial: if there are no joint between them, how are these links physically connected? Is there another way to connect links physically rather than with joints? Thanks again Jonathan
Hi thanks for the video, you focus so much on RHR but actually on what basis you select x axis? is it really free to choose because i am not sure about it...
![[[OFFICIAL MV]] เอิ้นขวัญให้อ้าย- อ๊ะอาย สกุณาพร | ค่ายเพลงเป็นหนึ่ง](http://i.ytimg.com/vi/83x6L3X3qh0/mqdefault.jpg)
![โฮ่ง! (SUGOI) - LYKN [ OFFICIAL MV ]](http://i.ytimg.com/vi/nFp4zjAARFs/mqdefault.jpg)
Thank you so much for this lecture, actually, I was completely hopeless about final grade of the robotics basics course in my university, now I can figure things out. you're teaching this lessons really good.
Thank you Angela, you saved my life, my school prof sucks.
Good morning Angela,
is there a rule for choosing the direction of the X axis while setting up the coordination frame once the Z axis is set?
I am not talking about choosing the Y Axis once the Z and X axes are set: the choice is bounded due to the righ hand rule.
For example at 18:33 you set the X axis "into the page", but it could have been set upwards, downwards or out of the page.
This would then imply a different Y axis.
Is there a rationale for your choice?
regards,
Jonathan
Ps thanks again for the efforts you put into these videos!
It depends on whether you want to use the Denavit-Hartenberg method to get the homogeneous transformation matrix, or not. If you do, then there are specific rules for the direction of x, also. In this video, I am assuming you will get the htm by directly finding the rotation matrix and displacement vector; in that case, there is no rule for the direction of x; it's a free choice.
In my new videos (see my playlist called Robotics 1 2017), I teach this topic assuming that the Denavit-Hartenberg method WILL be used, and so I teach two additional rules for assigning the X axis (it must be perpendicular to its own and the previous Z axis, and it must intersect the previous Z axis).
Makes sense. Thanks.
Good morning Ms. Sodemann,
firstly: many thanks for sharing these videos. These are *very* well done and help tremendously with easy to understand examples (if I might suggest: gathering them in playlist would help tracking the sequence on youtube).
I have a quastrion on the SCARA Manipulator showed at 12:57: you are adding an a1 and an a2 link. It looks to me that this is only one link as there are no joints in between them. Am I missing anything?
Thanks & Regards from Germany
Hi Jonathan,
You can find all of my videos in a playlist type of order at my website: www.robogrok.com
At 12:57, the a1 and a2 links are in different directions; that is why they are labeled separately. a1 is in the z0 direction, and a2 is in the x1 direction.
Hallo Angela,
thanks for the reply.
I understood that a link is a connection between joints and therefore between the first and the second cilindrical joint there is only 1 link.
So labelling a1 or a2 (because of different directions in the kinematic model) makes the model clearer, but there is still only one joint. Is my understanding correct?
Thanks again
Jonathan
Hi Jonathan,
I'm not completely sure I understand the question, but let me try this explanation:
Kinematically speaking, we could replace a1 and a2 with one link along the diagonal, and the math will come out exactly the same. So, links a1 and a2 are kinematically one link.
But, when we physically build the manipulator, the links a1 and a2 will represent different physical quantities. For example, the students in my class use the parts in their robotics kit to build a SCARA pick and place manipulator like this: th-cam.com/video/xyJDUyxwD2I/w-d-xo.html and use the equations derived here in their code to control it. In this case, the link a1 represents the height of a servo, and the link a2 represents the length of the bracket connecting the servos together. In cases like this, keeping the link lengths separate can help in the task of code-writing, since the variable 'height of the servo' and the variable 'length of the bracket' will be used multiple times in the code.
Hallo Angela,
I was probably too complicated in my question.
I try it again: between a1 and a2 in the video ( 12:57 ) I see no joint, which physically I understand to be only one robot "arm" (how could otherwise two arms be joined if not with a joint?). So I was wondering why you labeled them separately (a1 and a2) instead of labelling only once (e.g. "a1") as there should be only one arm.
Thanks again,
Jonathan
Hi Angela,
I am looking now at the video 1 1 4 1of2. I understand now that I missunderstood something about the concept of link: I thought that 2 links can only be joint with a joint (cylindrical or prismatic), so if there is no joint there is only one link, i.e. the robot has one arm.
It looks that this is not the case: in both this and the video 1 1 4 you label the vertical link a1 and the horizontal a2.
My question is probably trivial: if there are no joint between them, how are these links physically connected? Is there another way to connect links physically rather than with joints?
Thanks again
Jonathan
In the first example: x1 is perpendicular to z1 but its not perpendicular to z0 ? , is that correct?
Hi thanks for the video, you focus so much on RHR but actually on what basis you select x axis? is it really free to choose because i am not sure about it...
Karolina En start from reading DH coordinate video