tud-airlab / mppi-isaac

Model Predictive Path Integral Control using isaacgym for rollouts, gpu-accelerated
166 stars 19 forks source link

When running the benchmarks/panda_arm/run_experiment.sh file, an error message shows that some modules is missing. #49

Closed zhe-beginner closed 9 months ago

zhe-beginner commented 9 months ago

Hello, thank you very much for seeing my difficulties, I would be very grateful if you could help me. When I run the run_experiment.sh file in benchmarks, the following error message will be generated:

(mppiisaac-py3.8) ljz@ljz-computer:~/mppi-isaac/benchmarks/panda_arm$ ./run_experiment.sh pybullet build time: May 20 2022 19:44:17 Importing module 'gym_38' (/home/ljz/mppi-isaac/thirdparty/isaacgym/python/isaacgym/_bindings/linux-x86_64/gym_38.so) Setting GYM_USD_PLUG_INFO_PATH to /home/ljz/mppi-isaac/thirdparty/isaacgym/python/isaacgym/_bindings/linux-x86_64/usd/plugInfo.json PyTorch version 2.0.0+cu117 Device count 1 /home/ljz/mppi-isaac/thirdparty/isaacgym/python/isaacgym/_bindings/src/gymtorch Using /home/ljz/.cache/torch_extensions/py38_cu117 as PyTorch extensions root... Emitting ninja build file /home/ljz/.cache/torch_extensions/py38_cu117/gymtorch/build.ninja... Building extension module gymtorch... Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N) ninja: no work to do. Loading extension module gymtorch... Module forcesProMpcPlanner not found: No module named 'forcespro' Module acadosMpcPlanner not found: No module named 'acados_template' Link with name world does not has a parent. Link name is skipped. Link with name world does not has a parent. Link name is skipped. /home/ljz/.cache/pypoetry/virtualenvs/mppiisaac-DC-EcHMy-py3.8/lib/python3.8/site-packages/gym/spaces/box.py:78: UserWarning: WARN: Box bound precision lowered by casting to float32 logger.warn(f"Box bound precision lowered by casting to {self.dtype}") Link with name world does not has a parent. Link name is skipped. Link with name world does not has a parent. Link name is skipped. Traceback (most recent call last): File "/home/ljz/.cache/pypoetry/virtualenvs/mppiisaac-DC-EcHMy-py3.8/bin/runner", line 8, in sys.exit(main()) File "/home/ljz/.cache/pypoetry/virtualenvs/mppiisaac-DC-EcHMy-py3.8/lib/python3.8/site-packages/plannerbenchmark/exec/runner.py", line 201, in main myRunner.parseArguments() File "/home/ljz/.cache/pypoetry/virtualenvs/mppiisaac-DC-EcHMy-py3.8/lib/python3.8/site-packages/plannerbenchmark/exec/runner.py", line 102, in parseArguments planner = PlannerRegistry.create_planner(self._experiment, planner_file) File "/home/ljz/.cache/pypoetry/virtualenvs/mppiisaac-DC-EcHMy-py3.8/lib/python3.8/site-packages/plannerbenchmark/generic/planner_registry.py", line 32, in create_planner planner = cls.REGISTRY[name] KeyError: 'forcesprompc'

The general meaning is that the “forcespro” and “acados_template” modules are missing. But it seems that these modules cannot be installed using “pip install ***”

I would be very grateful if I could get your help.

cpezzato commented 9 months ago

Hi,

This is related to the afct that ForcesPro is a commercial software and you need a license for that (if you are a scholar you should get it for free when you request one).

If you cannot get a license then you can run the comparisons only with fabrics. To do so you can run:

runner -c setup/exp.yaml -p setup/mppi.yaml setup/fabric.yaml -n 10 --res-folder results/series --random-goal --random-obst --render

This is the same command in run_experiments.sh but withou MPC.

Also, in setup/exp.yaml you can set the duration T for your comparison, the current T=1000 is a bit lengthy for fast prototyping

I updated the README 3a98a14 in the benchmark folder to make this point clear.