This is exactly what it is made for! Would like to discuss your approach for the kinematics and the odometry if possible.
1 Like
Sure, would love to get in touch
3 Likes
Hello everyone again,
Sorry for taking so long. The fact that this is a side project for me makes the progress a little bit slow.
I finally managed to transfer all the logic into C++ and specifically ros controllers style, and it’s now working as it did in Python but much more organized, faster, and real-time safe.
I have not implemented the needed Gtests and rostests yet as I was doing the unit tests and so on a copy project out of catkin configurations.
What is needed now? and what is the next step?
I have received a lot of requests, which was very delightful and motivating, from people who wants to contribute.
At this point, I need help from people with experience in this topic. I will add the links to two related and useful papers. Please read them carefully and check their references.
Control of an pseudo-omnidirectional, non-holonomic, mobile robot based on an ICM representation in spherical coordinates
Instantaneous Centre of Rotation Estimation of an Omnidirectional Mobile Robot
The odometry performance is not stable right now, So we will be working on this. We will also need to search if there are better presentations for the ICM, steering limitations, closest steering angle, possibility to update it into a trajectory controller…etc. But those are future developments. What’s more important now is to have stable odometry to make the first PR.
If you want to join, I have made a slack for the project, please notify me to add you. Please consider joining if you want to help in the discussed problems. The discussion there will be basically focued on solving those problems with priorities to make the first PR.
Thank you again, looking forward to seeing you there.
3 Likes
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.