RuntimeError: normal expects all elements of std >= 0.0 #101
Description
Hi unitree:
I encountered an error while training with the latest code. The log as follows. How can i resolve this?
################################################################################
Learning iteration 29782/50000
Computation: 79754 steps/s (collection: 1.118s, learning 0.114s)
Mean action noise std: 0.35
Mean value_function loss: inf
Mean surrogate loss: 0.0000
Mean entropy loss: 9.1892
Mean reward: 38.28
Mean episode length: 1000.00
Episode_Reward/track_lin_vel_xy: 0.8978
Episode_Reward/track_ang_vel_z: 0.2551
Episode_Reward/alive: 0.1500
Episode_Reward/base_linear_velocity: -0.0145
Episode_Reward/base_angular_velocity: -0.0362
Episode_Reward/joint_vel: -0.1196
Episode_Reward/joint_acc: -0.0414
Episode_Reward/action_rate: -0.3541
Episode_Reward/dof_pos_limits: -0.0009
Episode_Reward/energy: -0.0033
Episode_Reward/joint_deviation_arms: -0.1670
Episode_Reward/joint_deviation_waists: -0.0606
Episode_Reward/joint_deviation_legs: -0.0972
Episode_Reward/flat_orientation_l2: -0.0101
Episode_Reward/base_height: -0.0008
Episode_Reward/gait: 0.5482
Episode_Reward/feet_slide: -0.0191
Episode_Reward/feet_clearance: 0.9766
Episode_Reward/undesired_contacts: -0.0006
Curriculum/terrain_levels: 4.8444
Curriculum/lin_vel_cmd_levels: 1.0000
Metrics/base_velocity/error_vel_xy: 0.2912
Metrics/base_velocity/error_vel_yaw: 0.9518
Episode_Termination/time_out: 0.9988
Episode_Termination/base_height: 0.0002
Episode_Termination/bad_orientation: 0.0010
Total timesteps: 2927788032
Iteration time: 1.23s
Time elapsed: 09:50:10
ETA: 06:40:38
Error executing job with overrides: []
Traceback (most recent call last):
File "/home/tigran/.conda/envs/env_isaaclab/lib/python3.11/site-packages/isaaclab/source/isaaclab_tasks/isaaclab_tasks/utils/hydra.py", line 101, in hydra_main
func(env_cfg, agent_cfg, *args, **kwargs)
File "/home/tigran/robot_g1/unitree_rl_lab/scripts/rsl_rl/train.py", line 204, in main
runner.learn(num_learning_iterations=agent_cfg.max_iterations, init_at_random_ep_len=True)
File "/home/tigran/.conda/envs/env_isaaclab/lib/python3.11/site-packages/rsl_rl/runners/on_policy_runner.py", line 149, in learn
loss_dict = self.alg.update()
^^^^^^^^^^^^^^^^^
File "/home/tigran/.conda/envs/env_isaaclab/lib/python3.11/site-packages/rsl_rl/algorithms/ppo.py", line 249, in update
self.policy.act(obs_batch, masks=masks_batch, hidden_states=hid_states_batch[0])
File "/home/tigran/.conda/envs/env_isaaclab/lib/python3.11/site-packages/rsl_rl/modules/actor_critic.py", line 119, in act
return self.distribution.sample()
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/tigran/.conda/envs/env_isaaclab/lib/python3.11/site-packages/torch/distributions/normal.py", line 74, in sample
return torch.normal(self.loc.expand(shape), self.scale.expand(shape))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: normal expects all elements of std >= 0.0
Set the environment variable HYDRA_FULL_ERROR=1 for a complete stack trace.
2025-12-23T21:26:12Z [35,652,833ms] [Warning] [omni.physx.plugin] USD stage detach not called, holding a loose ptr to a stage!
2025-12-23T21:26:13Z [35,653,707ms] [Warning] [carb] Recursive unloadAllPlugins() detected!