roboticlemon

Definitely agree with @lucasw, it's very annoying having to use mouse input every time you launch. Was this issue ever addressed? I'd be happy to look into doing a PR.

Same thing, this is still broken...

Also very interested to see the project fully developed for its application in full human pose inference

Hi @gurbain, first of all thank you for the pull request, for some reason practical implementations of swerve drive (4WIS) controllers are almost non existent in the ROS ecosystem. One...

Hi @gurbain, Thank you for the braking fix. The suboptimality of turning problem is still present albeit less noticeable now that you have steer clipping. The problem is not that...

Far too computationally expensive to perform those kinds of computer vision operations on a raspberry pi- the GPU isn't strong enough. It would be possible to run it on a...

I too am interested in this. A holonomic motion model which supported pure translation and rotation would be a very good addition to this library. I'd be interested in contributing...

Convert your DICOM to a standard image format like jpeg or png then process your data, Otherwise you will need to submit your own PR to add this functionality.

I would filter the pointcloud using a simple cropbox first. See [this answer](https://answers.ros.org/question/186115/how-to-filter-robots-footprint-from-a-pointcloud/) if you need something more complicated.

I'm experiencing this same issue.