[Question] How to Align Joint Responses Between Isaac Sim and Gazebo for Reinforcement Learning Deployment #5

Hi @evronix ,
First of all, thank you for your excellent project! I'm currently using your code framework to train reinforcement learning models in Isaac Sim and planning to validate the deployment in Gazebo. However, I'm encountering a critical issue: the models trained in Isaac Sim perform very poorly when deployed in Gazebo. After initial investigation, I've identified that the root cause is inconsistent joint responses between the two simulators.
Could you provide guidance on how to align the joint behaviors between these two platforms?
Thank you in advance for your help! Looking forward to your reply.