Snake-like Robot Modeling and Navigation (Project 224)

[robertogl]

Contribute to the discussion by asking and/or answering questions, commenting, or sharing your ideas for solutions to project 224

[Nachiket497]

Hello, my name is Nachiket Sawwalakhe and I am a student at VNIT Nagpur. I am interested in working and collaborating on this project. 

[SafalRaj10]

Hi, This is Safal Raj, a Mechanical Engineering student at IIT Patna. I'm intrigued and will be contributing to this project.

[grsasaki-tmw]

Hello @Nachiket497 and @SafalRaj10, we do appreciate your interest in this project! If you have any questions or points you would like to discuss, I welcome and encourage you to post here. Best Regards, Gen Sasaki

[Nachiket497]

I and @amit-420 have created a simulink model a snake and we want to run it on ground plane. How to add a ground env along with the snake model ? can you please provide some resources on this?
@grsasaki-tmw

[grsasaki-tmw]

@Nachiket497, assuming you are using Simscape Multibody as the project recommends, you can use the Infinite Plane block to define the ground plane. (https://www.mathworks.com/help/sm/ref/infiniteplane.html)

[Nachiket497]

Will check on this.

[amit-420]

@grsasaki-tmw we are trying to implement infiniteplane in snake model. As of now we have connected "R" of the block to world frame and "G" to the contact force block as mentioned in the documentation but we cannot find how to move ahead. Could you provide any example where infinte plane is used with a CAD model with multiple contact points ?

[bhavikpatel2]

Hi, I am a research scholar at IIT Guwahati and am interested in working and collaborating on this project

[bhavikpatel2]

Hi @robertogl
I am fine on Friday. I am following +5:30 GMT. I will connect with you anytime on Teams/ Google Meet/ Zoom whatever suits you.
Email id: [email protected]

[bhavikpatel2]

Hi @robertogl
It was a nice conversation with you.
Finally, I found the problem. The transformation sensor doesn't have full-angle information. It has only absolute value and therefore, the head angle controller failed. I have rectified that problem, and the robot tracks whatever reference angle is given. Finally, I implemented a head controller. You may find the final version of implemented head controller in [Robot_head_control/Snake_Robot_Head_Control.slx].

[robertogl]

Awesome! I am so glad to hear that you found and fixed the problem. It sounds like you have done with the project. Is there anything else you are planning to do before submitting the solution? Otherwise, please go ahead and complete the submission form.
Thanks!

[bhavikpatel2]

Yes, I will submit it by making readme,md file and comments within two week.
Thank you.

[bhavikpatel2]

Hi @robertogl

I have submitted the solution and updated the Readme.md file. I tried to make it as simple as possible for learners. If any upgradation is required, kindly suggest me.
Thank you for being so supportive in completing this project.

[OdlLed234]

Hi. I am a student at the University of the Witwatersrand, Johannesburg, and interested in working and collaborating on this project

[Antoine-ms]

Hi, I am Antoine Masson, a mechanical and control system engineering student at INSA Rennes, France. I am interested in working and collaboring on the project

[Antoine-ms]

I uploaded 3 versions of my anisotropic model on my repository (3_1 ; 3_2 ; 3_3). I checked the reference frames again and I tried to swap them, but I still couldn't manage to get my snake to move.

[Antoine-ms]

Hi @robertogl

I finished the design of my controller, I tested it on several scenario and everything works fine exept for one last problem :
The desired direction of the snake is computed as an angle in radian between -pi and pi, which raises the problem of the discontinuity. Do you have any idea how I can handle that ?
Thanks

[robertogl]

The error to feed back has to be calculated properly so to identify those jumps and unwrap their values when bigger than a certain threshold ('pi') by adding 2pi with the appropriate sign. For these cases our unwrap function will be useful.

[Antoine-ms]

Hi @robertogl

I am struggling to get the unwrap block to work. It seems like it has no effect on the signal. I tried to play with its parameters but I couldn't make it work. Do you think you could take a look on that ?
Thanks

[robertogl]

I am having some issues in running the model. But have you tried applying the unwrap only to the error angle?

[Alexan41]

Is the competition still on?

[robertogl]

Hi @Alexan41,

Currently the 'Sustainability and Renewable Energy' challenge is on until January. However, this project does not qualify as part of that. But if you mean this project, yes this is still active!

Cheers,
Roberto

[Alexan41]

Okay but I focus on the snake like robot project because I've already started working on it.

[Alexan41]

Please @robertogl ,

I made progress with my project and I want to share with you what I have done so far so that you can appreciate.
I don't know very weel how github works so I would need some tips to share my project with you.

Thanks in advance.

[robertogl]

Hi @Alexan41,

You can easily use this template. Just click on 'Use this template' and then 'Create your repository'. Once you have done that you just need to change the name of the repository and the owner (you will see your profile name in the drop down menu), and then just click on 'submit'. In the new repository you can literally drag and drop the files you need to share.

That's a good start to share your initial files. Let me know if you need anything else.

[RKommX]

Hi, I'm Rajdeep Paul, 2nd year aerospace engineering student at IIT Bombay. I wish to spend my semester break working on this project.