Use this image to train rl model

[XinyuKhan]

Thank you all for such an excellent project and such a convenient containerized desktop environment. Since I frequently work remotely and have some needs for training reinforcement learning models, and it's impossible to carry my desktop computer with me, this project has been a huge help. Below is a training image I developed based on this project; I hope it will be helpful to others.

remote_isaac_lab_docker

[XinyuKhan]

I recently discovered that the configuration of the LD_PRELOAD environment variable in this image causes .so files loaded using dlopen in the conda environment to be undetectable. I haven't found a perfect solution yet; the only workaround is to remove libdlfaker.so. If anyone knows a better solution, please share it here. I also hope that others encountering the same problem can find help here.