Simulation Environment

[liaodisen]

Hello Everyone,

I am new to robotics, and I am using MacOS and a remote machine with A6000 GPU available. I can run the script python scripts/joint_monkey_smpl.py, and I used XQuartz to load the GUI window on my local machine, though it is lagging. However, when I run python scripts/vis/vis_motion_mj.py, I had error:

libGL error: No matching fbConfigs or visuals found libGL error: failed to load driver: swrast /home/d7liao/anaconda3/envs/isaac/lib/python3.8/site-packages/glfw/init.py:914: GLFWError: (65543) b'GLX: Failed to create context: BadValue (integer parameter out of range for operation)' warnings.warn(message, GLFWError) ERROR: could not create window

Does anyone have better suggestion on how to run mujoco using remote machine and how to render isaac gym more fluently? Thank you!

Edit: I think this is because of XQuartz https://github.com/XQuartz/XQuartz/issues/144, does anyone know how to get over it?

[Anonnyyy]

请问你用的gym preview4吗

[ZhengyiLuo]

I have had better luck with VNC than XQuartz; there is always some issue when I tried to use xservers from command line when GPU rendering is involved.

[ZhengyiLuo]

请问你用的gym preview4吗

Yes, preview 4.

[Anonnyyy]

Thank you very much! : ) However i met some problems when running: python scripts/vis/vis_motion_mj.py it shows:

** Current motion keys ** Sampling motion: tensor([0]) standing

---

0it [00:00, ?it/s]pbar 0it [00:00, ?it/s] 0it [00:00, ?it/s] Traceback (most recent call last): File "scripts/vis/vis_motion_mj.py", line 105, in motion_lib.load_motions(skeleton_trees=[sk_tree] num_motions, gender_betas=[torch.zeros(17)] num_motions, limb_weights=[np.zeros(10)] num_motions, random_sample=False, start_idx=curr_start) File "/home/anony/omnigrasep/PHC/phc/utils/motion_lib_base.py", line 249, in load_motions res_acc.update(self.load_motion_with_skeleton(jobs[0], None, 0)) File "/home/anony/omnigrasep/PHC/phc/utils/motion_lib_smpl.py", line 116, in load_motion_with_skeleton for f in pbar: File "/home/anony/anaconda3/envs/rlgpu/lib/python3.8/site-packages/tqdm/std.py", line 1181, in iter for obj in iterable: TypeError: 'int' object is not iterable

看起来像是motion import失败了， motion_file = "sample_data/amass_isaac_standing_upright_slim.pkl" 导入的motion list是空的。请问您有遇到这样的问题吗？

[ZhengyiLuo]

Thank you very much! : ) However i met some problems when running: python scripts/vis/vis_motion_mj.py it shows:

** Current motion keys ** Sampling motion: tensor([0]) standing

0it [00:00, ?it/s]pbar 0it [00:00, ?it/s] 0it [00:00, ?it/s] Traceback (most recent call last): File "scripts/vis/vis_motion_mj.py", line 105, in motion_lib.load_motions(skeleton_trees=[sk_tree] num_motions, gender_betas=[torch.zeros(17)] num_motions, limb_weights=[np.zeros(10)] num_motions, random_sample=False, start_idx=curr_start) File "/home/anony/omnigrasep/PHC/phc/utils/motion_lib_base.py", line 249, in load_motions res_acc.update(self.load_motion_with_skeleton(jobs[0], None, 0)) File "/home/anony/omnigrasep/PHC/phc/utils/motion_lib_smpl.py", line 116, in load_motion_with_skeleton for f in pbar: File "/home/anony/anaconda3/envs/rlgpu/lib/python3.8/site-packages/tqdm/std.py", line 1181, in iter for obj in iterable: TypeError: 'int' object is not iterable

看起来像是motion import失败了， motion_file = "sample_data/amass_isaac_standing_upright_slim.pkl" 导入的motion list是空的。请问您有遇到这样的问题吗？

I just tried sample_data/amass_isaac_standing_upright_slim.pkl and it does contain the motion. Please pull the most recent code.