Skip to content

fegonzalez7/rob_unal_clase5

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 

Repository files navigation

The phantom revenge

Fight results:

Open on your own risks

Screenshot-from-2022-04-13-16-09-08.png

Robot wins!

Screenshot-from-2022-04-13-16-09-16.png


Forward kinematics

This repo won't teach you how to get the DH parameters, to do that I higly recommend the 3rd chaptor of Spong. Nevertheless, I will let some pictures that are useful.

Phantom X Mashmellows

measures


Lazyness alert: Before unfold the picture, try yourself. It's the best way to learn.

Phantom X StdDH Frames

frames


The future (?) - PX 100

In the fortunate case you are working with the PX100 things maybe a little bit different. I found mr. Chitiva's repo quite good.

Lazyness alert no. 2: Use the repo as a guide, don't steal his code, try first.

Requirements

The all mighty RTB for python

We have used the Matlab RTB, but you know Matlab, so lets install the python version.

Open a terminal and run:

sudo apt install python3-pip
pip3 install roboticstoolbox-python
pip3 install sympy

Check the following sources:

Definitely Peter Corke is so pro, I printed a poster of him 😆.

Python RTB usage

Let's create a python script:

import roboticstoolbox as rtb
from spatialmath import SE3

robot = rtb.DHRobot([rtb.RevoluteDH(a=1),rtb.RevoluteDH(a=1)],name="2R")
print(robot)
robot.plot([0.75,0.26])
input()

Screenshot-from-2022-04-03-00-57-57.png

Run the px_rtb.py script and check if you obtain the a real PhantomX model.

Activity: Create a notebook and follow the same pipeline presented in px_rtb.py. What does happen with the figures? 😬

Moveit

Moveit! is a sort of sourcery that allows you to plan your robot's motion. It relies mainly on the OMPL, which is a full library for motion planning.

In the case you do not have moveit installed:

sudo apt install ros-noetic-moveit

Note: By the time I wrote this, the last stable version of moveit was 1.1.9.

Because our main focus is on robotics we will have a short taste of what Moveit is capable of, so first clone px_text_moveit on your catkin_ws and compile it.

cd ~/catkin_ws/src
git clone https://github.com/felipeg17/px_test_moveit.git
catkin build px_test_moveit

Remember to keep px_robot updated, so give it a pull. Go to the px_robot folder and type:

git pull origin master

I added a couple of interesting new files, the most important px_moveit_v2.py, it should be located on ../scripts.

Source everything and run:

roslaunch px_robot px_moveit.launch 

If everything goes well, you shall see something like this: Screenshot-from-2022-04-13-15-41-56.png

Screenshot-from-2022-04-13-15-42-09.png

Then go to the px_robot/scripts and run:

python px_moveit_v2.py 

Screencast-from-13-04-22-15-52-02.gif

Note: Do not dare bitching about the quality. It was for speed purposes.

If you want to learn more about moveit y recommend the following resources:

Notice: Also Realpe has a lot of experience on moveit, go and check out his repos. Realpe and Fajardo have been working with me, if you want to have a portfolio like them, consider joining to my research team, it is a limited offer, once I complete the team, people will beg for being part of this project.

Screenshot-from-2022-04-13-16-30-51.png


<Lab guide> : https://drive.google.com/file/d/1RGBmvqbng3HATgVkgL6y5xWPqr2GDuzn/view?usp=sharing



<Recorded class:> : https://drive.google.com/file/d/12lFjGgYykycmhP98ZCwVEyhBpqWARUnh/view?usp=sharing


Acknowledgments:

  • Jose Manuel Fajardo
  • Sebastian Realpe
  • Cristian Chitiva (Well crafted repo for the PX100)
  • William Sierra (Thanks for the pics)
  • Brian Vasquez (Thanks for the pics)

This repo has taken a lot of effort, so consider to leave a star, follow me, and if you feel generous I have Paypal (just kidding).

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published