Thank you a lot sir! Videos you upload about "robotics content" helps me big time in my Ms. We say "God bless you" in Turkish "Allah senden razı olsun." So, Allah senden razı olsun, Mr. Muhammad.
Waypoint matrix is defined as n x p matrix, where 'n' is the dimensions and 'p' is the number of way points. If you are generating workspace trajectory then 'n' = 3 (x, y, z), whereas if you are generating joint space trajectory then 'n' = DOF. For example a workspace trajectory for a 4 DOF robot having 4 waypoints would be [1, 3, 2, 3; 2, 4, 3, 1; 2, 3, 4 1], a 3x4 matrix, whereas Joint space trajectory for the same robot would be [20, 40, 30, 50; 40, 30, 50, 60; 30, 20, 10, 50; 30, 40, 50, 60] a 4x4 matrix.
Dear Razi Sahab, Aoa A great effort indeed. I have really learnt a lot. However, there are certain issues which need to be resolved. Even though I followed the steps explained exactly, I am not able to obtain smooth x and y profiles on the end-effector. In fact, the robot undergoes jerks during its motion. If you so agree, I can send you the simulink files. Thank you
Hello, first thing I want to say is, amazing lectures. I have been trying to learn simulink for while and this is the best course I have found on it. Thank you Also, In this video and in the last one too, I having some problem in my simulation. The first three lines of the square are traced fine but in the fourth one it just shoots up. So the fourth line isn't traced. I compared the scope graphs and they all look similar to yours. let me know how I can fix it.
how do you trace the path followed by end effector? is there any way to create the trajectory without assigning the coordinate (x,y,z)?Also how did you create the waypoints? Can you please make a video containing all the abovementioned questions? i really appreciate your time and I am learning a lot from you. is there anything I can do to help you? I am a big fan of ROBOTs
You can record the position of the end-effector by using a Forward Kinematics (Get Transform) block from the Robotics Systems Toolbox (See previous videos in the playlist). No there is no way of generating a trajectory without specifying x,y,z coordinates or joint coordinates that will depend on the number of joints. Trajectory generation works by plotting a suitable path between the provided waypoints. These waypoints can be x,y,z coordinates or joint angles for joint space trajectories. I created the waypoints by simply using the fact that where my robot needs to go, those are the waypoints.
Thanks sir for your matlab series about robotics! I really love it and I would like to study more about it. Could you please suggest me a book in which I can find more about this topic (manipulators on matlab/simulink/simscape) to add to the matlab documentation? I hope you can help me :)
I wanted to adapt this method with something other than a square, so I used a much more complex model where the waypoints were black pixels of an image, and I converted each pixel into an XY coordinate (with Z=0). With this method, I have a matrix of over 20,000 elements. It seems to work, but it takes several hours to compile. Do you have any ideas to make Sinscape compile faster?
Remove everything from the simscape model except the trajectory generation block. Save the generated trajectory into a workspace variable. Use another simulink file to drive the robot on the saved trajectory.
Hi Thank you for the video, may Allah always bless you. I want to try this simulation, but why in my utilities there is only the coordinate transformation conversion box. do you know the problem? how to solve it?
Hello! I'm enjoying the video lecture. However, as I follow your approach, the runtime becomes too long. In your videos, the results appear instantly. Is there something I'm doing wrong?
Whenever you use SimScape Visualization the results will take time, however, if you only use RST blocks, the computations will be done almost instantaneously. I've sometimes cropped the part of the video where the simulation is running.
This is the waypoint matrix that I used wp = [0.35,0,0.1; 0.25, 0.01, 0.1; 0.25, 0.11, 0.1; 0.15, 0.11, 0.1; 0.15, 0.01, 0.1; 0.25, 0.01, 0.1]'; There is a transpose at the end of the matrix!!!
thank you for this video. I wanted to make a simple control on a dynamic model of an autonomous car (model: bicycle) using a PID controller and imposing a reference trajectory but I don't know how. I got the transfer function but I don't know how I impose a reference trajectory. Also. is the PID controller and transfer function are sufficient in block schema or not?
It depends on what your PID controller is doing. If its a joint space controller then you should provide the joint space trajectory at the reference input of the PID controller. If you can share the block diagram of your system then I might be able to suggest something more concretely. You can share it to learningorbis@gmail.com
Hi, sir. Thanks for your video. I have already see the way of setting the waypoint, but still I didn't figure out where can I set the wp file. Is it inside the simulink or should I renew a script?
You must be entering wrong velocity boundary conditions. Consult the video or the help of Trajectory Generator block to know the dimensions of the velocity boundary condition matrix.
saving waypoints in workspace is very confusing. I tried to use "to workspace block" with coordinate tranformation conversion (forward kinematic) but it isn't working. Can you please help me with this? I will be grateful
Can you share the error yo uare getting, because saving things in workspace is quite straightforward. Just connect the 'To Worksapce Block' and with default values of the block all data connected to this block will be saved in the workspace in 'Simulation Output' type variable named as 'Out'. Furthermore, multiple data streams will be present in the 'Out' variable in form of a timeseries variable.
It can be done by defining the waypoint matrix such that the robot starts from one corner of the square and move to the next corner horizontally. After that it should move downwards a bit, depending on the painting nozzle spread, and then move again. Repeat the process until the square is covered.
These time points are for Polynomial Trajectory Block. In this case the time points are times when the corresponding waypoint should be achieved. Make sure you are configuring the other things as I've done in the video. For example, Boundary Velocity Conditions should have zeros(3,n) size, where n is the number of waypoints. Anyhow this is the waypoint matrix that I used: wp = [0.35,0,0.1; 0.25, 0.01, 0.1; 0.25, 0.11, 0.1; 0.15, 0.11, 0.1; 0.15, 0.01, 0.1; 0.25, 0.01, 0.1]'; There is a transpose at the end of the matrix!!!
I have a very simple robot which makes transition only in x-direction. i need a simulink model. the problem i have is, how i can get the real world vibration into the model. I try to reconstruct the model in relative motion to get the position x.
wp is the waypoint matrix that contains 3 rows and 'n' columns, where 'n' the number of waypoints you want your robot to go through. Every row contains joint angle values for one of the joints, whereas if you are designing trajectory in workspace then first row is for x-axis, second for y-axis, and third for z-axis.
when i am moving the end effector from negative plane to positive plane i am getting the undershoot. It is happening when i am moving from positive plane to negative plane why?
If you are using the inverse kinematics then it sometimes causes problems at boundaries. If you can share the graphs at learningorbis@gmail.com then I might be able to reply more concretely.
In the same way as you have generated positional trajectory. You should have a 3x3 initial rotation matrix and a 3x3 final required rotation matrix. Use trajectory generator to generate a 9-dimensional trajectory where each dimension corresponds to one of the element of a 3x3 rotation matrix.
This is the waypoint matrix that I used wp = [0.35,0,0.1; 0.25, 0.01, 0.1; 0.25, 0.11, 0.1; 0.15, 0.11, 0.1; 0.15, 0.01, 0.1; 0.25, 0.01, 0.1]'; There is a transpose at the end of the matrix!!!
Thank you a lot sir! Videos you upload about "robotics content" helps me big time in my Ms. We say "God bless you" in Turkish "Allah senden razı olsun." So, Allah senden razı olsun, Mr. Muhammad.
You are very welcome. JazakAllah
This videos are really helping develop my 5DOF manipulator for my bachelor thesis. Thank you!
Glad to know this 🙂
Did you control your orientation of the end effector?
Hello, sir. I am so excited about the tutorial you made , but would you please help me on how to generate the trajectory in MATLAB 2018?
Hi , im so excited about the toutorial you made , would you please explain how to define the wp .m file ?
Do we need write y based on X or something ?
Waypoint matrix is defined as n x p matrix, where 'n' is the dimensions and 'p' is the number of way points. If you are generating workspace trajectory then 'n' = 3 (x, y, z), whereas if you are generating joint space trajectory then 'n' = DOF. For example a workspace trajectory for a 4 DOF robot having 4 waypoints would be [1, 3, 2, 3; 2, 4, 3, 1; 2, 3, 4 1], a 3x4 matrix, whereas Joint space trajectory for the same robot would be [20, 40, 30, 50; 40, 30, 50, 60; 30, 20, 10, 50; 30, 40, 50, 60] a 4x4 matrix.
Thank you so much for your clear explanation.Helped a lot.
Glad it helped!
Dear Razi Sahab,
Aoa
A great effort indeed. I have really learnt a lot.
However, there are certain issues which need to be resolved. Even though I followed the steps explained exactly, I am not able to obtain smooth x and y profiles on the end-effector. In fact, the robot undergoes jerks during its motion. If you so agree, I can send you the simulink files. Thank you
Yes you may share, and I'll try to have a look.
Hello, first thing I want to say is, amazing lectures. I have been trying to learn simulink for while and this is the best course I have found on it. Thank you
Also, In this video and in the last one too, I having some problem in my simulation. The first three lines of the square are traced fine but in the fourth one it just shoots up. So the fourth line isn't traced. I compared the scope graphs and they all look similar to yours. let me know how I can fix it.
Thank you.
Make sure to have the last waypoint in the trajectory same as the first one.
how do you trace the path followed by end effector? is there any way to create the trajectory without assigning the coordinate (x,y,z)?Also how did you create the waypoints? Can you please make a video containing all the abovementioned questions?
i really appreciate your time and I am learning a lot from you. is there anything I can do to help you?
I am a big fan of ROBOTs
You can record the position of the end-effector by using a Forward Kinematics (Get Transform) block from the Robotics Systems Toolbox (See previous videos in the playlist).
No there is no way of generating a trajectory without specifying x,y,z coordinates or joint coordinates that will depend on the number of joints. Trajectory generation works by plotting a suitable path between the provided waypoints.
These waypoints can be x,y,z coordinates or joint angles for joint space trajectories. I created the waypoints by simply using the fact that where my robot needs to go, those are the waypoints.
Thanks sir for your matlab series about robotics! I really love it and I would like to study more about it. Could you please suggest me a book in which I can find more about this topic (manipulators on matlab/simulink/simscape) to add to the matlab documentation? I hope you can help me :)
Sorry there is no book that I know of. You have to learn things in bits and pieces from here and there.
I wanted to adapt this method with something other than a square, so I used a much more complex model where the waypoints were black pixels of an image, and I converted each pixel into an XY coordinate (with Z=0). With this method, I have a matrix of over 20,000 elements. It seems to work, but it takes several hours to compile. Do you have any ideas to make Sinscape compile faster?
Remove everything from the simscape model except the trajectory generation block. Save the generated trajectory into a workspace variable. Use another simulink file to drive the robot on the saved trajectory.
Hi Thank you for the video, may Allah always bless you. I want to try this simulation, but why in my utilities there is only the coordinate transformation conversion box. do you know the problem? how to solve it?
You might have older version of MATLAB. Try 2019 or latest version.
Hello! I'm enjoying the video lecture. However, as I follow your approach, the runtime becomes too long. In your videos, the results appear instantly. Is there something I'm doing wrong?
Whenever you use SimScape Visualization the results will take time, however, if you only use RST blocks, the computations will be done almost instantaneously. I've sometimes cropped the part of the video where the simulation is running.
thank you sir but why you said that i'm not concerned with the acceleration and you only trace the velocities graphs ?
Just for the video I'm not looking at accelerations. But if you want you can, it depends on your requirement.
Nice ，Tq Prof.
Welcome!!
how can we know the value of "wp"? I can,t find the number at the video even I watch the part -5:30
This is the waypoint matrix that I used
wp = [0.35,0,0.1; 0.25, 0.01, 0.1; 0.25, 0.11, 0.1; 0.15, 0.11, 0.1; 0.15, 0.01, 0.1; 0.25, 0.01, 0.1]';
There is a transpose at the end of the matrix!!!
Also my end-effector is not able to reach the last waypoint/square corner. Instead it directly jumps to the home configuration/initial point
There must be a logical explanation for this. Send me your files and I'll give it a look.
sir can yo u tell wp variable for this configuration beacuse i dont figure out kinldy help in this regard sir
Follow this new video it provides everything: th-cam.com/video/f1LUAo_YzlM/w-d-xo.html
thank you for this video. I wanted to make a simple control on a dynamic model of an autonomous car (model: bicycle) using a PID controller and imposing a reference trajectory but I don't know how. I got the transfer function but I don't know how I impose a reference trajectory. Also. is the PID controller and transfer function are sufficient in block schema or not?
It depends on what your PID controller is doing. If its a joint space controller then you should provide the joint space trajectory at the reference input of the PID controller. If you can share the block diagram of your system then I might be able to suggest something more concretely. You can share it to learningorbis@gmail.com
Hi, sir. Thanks for your video. I have already see the way of setting the waypoint, but still I didn't figure out where can I set the wp file. Is it inside the simulink or should I renew a script?
wp is just a variable in MATLAB workspace. You can directly initiate wp variable in Command Window or in an M-file.
Could you please show what is in the wp array??
Kindly listen to the portion around 05:30 time
Hello, Thank you for the videos they are really helpful. Is there anywhere we can download the robot that you've done?
You can email me at learningorbis@gmail.com. I'll share the files with you.
Additionally you may download initial files from www.mathworks.com/matlabcentral/fileexchange/125875-how-to-design-robotic-manipulators-in-matlab-2021
@@LearningOrbis Thank you! ☺
I put WP points but i get massage "the dimensions of way points input and velocity boundary condition name-value pair must match" how can i solve this
You must be entering wrong velocity boundary conditions. Consult the video or the help of Trajectory Generator block to know the dimensions of the velocity boundary condition matrix.
saving waypoints in workspace is very confusing. I tried to use "to workspace block" with coordinate tranformation conversion (forward kinematic) but it isn't working. Can you please help me with this? I will be grateful
Can you share the error yo uare getting, because saving things in workspace is quite straightforward. Just connect the 'To Worksapce Block' and with default values of the block all data connected to this block will be saved in the workspace in 'Simulation Output' type variable named as 'Out'. Furthermore, multiple data streams will be present in the 'Out' variable in form of a timeseries variable.
@@LearningOrbis Thank you so much, problem resolved
Could you please make a video on how to generate trajectory for a robot arm to paint within a square??
It can be done by defining the waypoint matrix such that the robot starts from one corner of the square and move to the next corner horizontally. After that it should move downwards a bit, depending on the painting nozzle spread, and then move again. Repeat the process until the square is covered.
Thanks!
How does your Time points even work?
[0, 0.5, 1.5 2.5 3.5 4.5] is invalid. When i use this: tp = [0; 0.5; 1.5; 2.5; 3.5; 4.5];
These time points are for Polynomial Trajectory Block. In this case the time points are times when the corresponding waypoint should be achieved. Make sure you are configuring the other things as I've done in the video. For example, Boundary Velocity Conditions should have zeros(3,n) size, where n is the number of waypoints.
Anyhow this is the waypoint matrix that I used:
wp = [0.35,0,0.1; 0.25, 0.01, 0.1; 0.25, 0.11, 0.1; 0.15, 0.11, 0.1; 0.15, 0.01, 0.1; 0.25, 0.01, 0.1]';
There is a transpose at the end of the matrix!!!
@@LearningOrbis sry Had to write it. I figured Out Yesterday. But thanks a Lot. IT was a very good Help For my diploma.
I have a very simple robot which makes transition only in x-direction. i need a simulink model. the problem i have is, how i can get the real world vibration into the model. I try to reconstruct the model in relative motion to get the position x.
At @5:34 what is the content of the variable wp
wp is the waypoint matrix that contains 3 rows and 'n' columns, where 'n' the number of waypoints you want your robot to go through. Every row contains joint angle values for one of the joints, whereas if you are designing trajectory in workspace then first row is for x-axis, second for y-axis, and third for z-axis.
how to set waupoints in polynomial trajectory?specially set wp variable?
Will upload soon
when i am moving the end effector from negative plane to positive plane i am getting the undershoot. It is happening when i am moving from positive plane to negative plane why?
If you are using the inverse kinematics then it sometimes causes problems at boundaries. If you can share the graphs at learningorbis@gmail.com then I might be able to reply more concretely.
@@LearningOrbis i had sent the mail along with the graphs.
How could I generate trajectory for orientation of the end effector? I don't want to change the coordinates just the orientation of the end effector.
In the same way as you have generated positional trajectory. You should have a 3x3 initial rotation matrix and a 3x3 final required rotation matrix. Use trajectory generator to generate a 9-dimensional trajectory where each dimension corresponds to one of the element of a 3x3 rotation matrix.
@@LearningOrbis okay thanks. I will give it a try.
Hey, I hope you are doing well. Can you please share me the link where you made a video in Solidwork and Matlab?
I'm afraid I haven't made a video on that. But will try to make one in near future.
what are the values of the waypoint variable ?
Look somewhere in the previous comments, i've shared the waypoints.
what are waypoints @aakashsingh6771
sir, wp ke values kya hai ?
wp waypoints hain [x, y, z] values hain jidher robot jana chahiye
do you share the files? thanks for videos
Contact me at learningorbis@gmail.com
@@LearningOrbis do u share this file ?
Sir can you please kindly share the waypoints
This is the waypoint matrix that I used
wp = [0.35,0,0.1; 0.25, 0.01, 0.1; 0.25, 0.11, 0.1; 0.15, 0.11, 0.1; 0.15, 0.01, 0.1; 0.25, 0.01, 0.1]';
There is a transpose at the end of the matrix!!!
@@LearningOrbis Thank you sir
@@LearningOrbis I was wondering how to switch rows and columns, now i remember. Thanks!
wp?????????? i didn't get it
wp is a waypoint matrix. It contains all the way points you want your robot to go through.