stepjam / RLBench

A large-scale benchmark and learning environment.

Home Page:https://sites.google.com/corp/view/rlbench

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Error with dataset_generator

diegomaureira opened this issue · comments

Hello, I'm trying to get demos of the eight tasks used for testing the ARM Algorithm (https://arxiv.org/abs/2105.14829) but with a different arm, in this case I changed the 'robot_setup' in the environment from panda to jaco:

rlbench_env = Environment(
action_mode=MoveArmThenGripper(JointVelocity(), Discrete()),
robot_setup='jaco',
obs_config=obs_config,
headless=True)
rlbench_env.launch()

I started with take lid off saucepan task, but I have this error.

python3 dataset_generator.py --save_path=/home/diego/demosRL --tasks=take_lid_off_saucepan --image_size=128,128 --renderer=opengl --episodes_per_task=10 --variations=1 --processes=1

Warning: Ignoring XDG_SESSION_TYPE=wayland on Gnome. Use QT_QPA_PLATFORM=wayland to run on Wayland anyway.
Process Process-2:
Traceback (most recent call last):
File "/home/diego/anaconda3/lib/python3.9/site-packages/rlbench/task_environment.py", line 78, in reset
desc = self._scene.init_episode(
File "/home/diego/anaconda3/lib/python3.9/site-packages/rlbench/backend/scene.py", line 141, in init_episode
raise e
File "/home/diego/anaconda3/lib/python3.9/site-packages/rlbench/backend/scene.py", line 133, in init_episode
raise BoundaryError()
rlbench.backend.exceptions.BoundaryError

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "/home/diego/anaconda3/lib/python3.9/multiprocessing/process.py", line 315, in _bootstrap
self.run()
File "/home/diego/anaconda3/lib/python3.9/multiprocessing/process.py", line 108, in run
self._target(*self._args, **self._kwargs)
File "/home/diego/Documentos/RLBench/tools/dataset_generator.py", line 248, in run
obs, descriptions = task_env.reset()
File "/home/diego/anaconda3/lib/python3.9/site-packages/rlbench/task_environment.py", line 82, in reset
raise TaskEnvironmentError(
rlbench.backend.exceptions.TaskEnvironmentError: Could not place the task take_lid_off_saucepan in the scene. This should not happen, please raise an issues on this task.
Data collection done!

I have the same error with the other 7 tasks. I tried with sawyer, panda and ur5 and those arms work fine, but jaco and mico don't work. Anyone has an idea about what is happening here? thanks in advance.