Dear NVIDIA team,
I just started to use Isaac Sim for simulation and I have the following two questions:
-
I created a ground plane and wanted to have some reference lines which could help me observe the moving of my robot clearer. I’m wondering if it is possible to grid the ground plane or add a coordinate for it by changing the load_stage.py. When I launched the Isaac Sim and open the .usd file from the graphic interface, I could see the coordinate of the world frame. However, if I load the .usd file by using load_stage.py in python, the coordinate will disappear;
-
After reading the documents, I still felt a little bit confused about the kit.update(x). From my understanding, x is the step time for the simulator. The program will calculate the motion of the robot in x and then display it. If I keep the input command be the same, then [kit.update(1)] and [for i in range(10): kit.update(0.1)] should give me same results. However, they were different when I tested it. I thought I might misunderstand this function. Could you explain more about this function for me : ) ?
Thank you so much for your help!
In our 2021.2.0 release we have a new ground plane asset which should help:
The asset is available at: omniverse://localhost/Isaac/Environments/Grid/default_environment.usd
We have a new set of tutorials which might also help and this ground plane is used in a few of them
https://docs.omniverse.nvidia.com/app_isaacsim/app_isaacsim/tutorial_required_interface.html
We have split the render and physics step calls in the release as well. One thing to note is that you need to update the simulation at a rate of 1/60 (or similar) otherwise things will not be stable. If you run with an update size of (1) vs 10x(.1)
They will lead to different results, the first did one very large (and unstable simulation step) the second did 10 smaller and more stable steps.
- If you want a higher level interface to managing the simulation and time we have a few standalone samples you can run in 2021.2.0 via ./python.sh
./python.sh standalone_examples/api/omni.isaac.core/simulation_callbacks.py
There is a new SimulationContext class that provides more control over rendering/update and physics stepping Simulation Context API
- There is a second example
./python.sh standalone_examples/api/omni.isaac.core/time_stepping.py
that shows the different ways you can step simulation and rendering.