Libraries for real-time Matplotlib animation
Closed this issue · 6 comments
When executing trajectory_test.py with custom_sim_gui = True in:
data_driven_mpc/ros_gp_mpc/config/configuration_parameters.py
Lines 34 to 36 in c3198cb
The following error arises:
Traceback (most recent call last):
File "src/experiments/trajectory_test.py", line 194, in
main(input_arguments)
File "src/experiments/trajectory_test.py", line 78, in main
world_rad=world_radius, full_traj=reference_traj)
File "/home/user/catkin_ws/src/data_driven_mpc/ros_gp_mpc/src/utils/visualization.py", line 377, in initialize_drone_plotter
mng.resize(*mng.window.maxsize())
AttributeError: 'MainWindow' object has no attribute 'maxsize'
I guess it is related to the missing libraries required to show the animation. It would be great to have a guide on how to install them (or at least a list of them)
Hi @JulioLP . Thanks for trying out our code.
May I ask you to just comment out this line in particular? And the line above aswell.
These two lines will just maximize the GUI window to full screen, so they are definitely non-essential. If removing them solves your problem we might use this as a temporary workaround.
Thanks for the fast response! I have commented the line and now there is a different error:
Traceback (most recent call last):
File "src/experiments/trajectory_test.py", line 194, in
main(input_arguments)
File "src/experiments/trajectory_test.py", line 119, in main
targets_reached=None, pred_traj=x_pred, x_pred_cov=None)
File "/home/user/catkin_ws/src/data_driven_mpc/ros_gp_mpc/src/utils/visualization.py", line 478, in draw_drone_simulation
projected_traj_artists[1].set_3d_properties(x_trajectory[trajectory_start_pt:, 2])
File "/home/user/temp/virtualenv_acados/gp_mpc_venv/lib/python3.6/site-packages/mpl_toolkits/mplot3d/art3d.py", line 143, in set_3d_properties
zs = np.broadcast_to(zs, xs.shape)
File "<array_function internals>", line 6, in broadcast_to
File "/home/user/temp/virtualenv_acados/gp_mpc_venv/lib/python3.6/site-packages/numpy/lib/stride_tricks.py", line 180, in broadcast_to
return _broadcast_to(array, shape, subok=subok, readonly=True)
File "/home/user/temp/virtualenv_acados/gp_mpc_venv/lib/python3.6/site-packages/numpy/lib/stride_tricks.py", line 118, in _broadcast_to
raise ValueError('cannot broadcast a non-scalar to a scalar array')
ValueError: cannot broadcast a non-scalar to a scalar array
Just for clarification, I have only modified the lines that you commented, and the "custom_sim_gui = True" that is needed.
Hi @JulioLP . I could reproduce your issue. It is caused by an incompatibility of matplotlib. Make sure that you have matplotlib==3.2.2 installed. You can check the version by activating your virtualenv and pasting the following in a python shell:
import matplotlib as mpl
mpl.__version__
Now with the new matplotlib version it runs, but the plot window is rendered black, and the following message arises:
Running simulation...
0%| | 0/1000 [00:00<?, ?it/s]/home/julio/temprpg/virtualenv_acados/gp_mpc_venv/lib/python3.6/site-packages/mpl_toolkits/mplot3d/proj3d.py:189: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
return np.array([xs, ys, zs, np.ones_like(xs) ])
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:20<00:00, 48.86it/s]
Now with the new
matplotlibversion it runs, but the plot window is rendered black, and the following message arises:Running simulation...
0%| | 0/1000 [00:00<?, ?it/s]/home/julio/temprpg/virtualenv_acados/gp_mpc_venv/lib/python3.6/site-packages/mpl_toolkits/mplot3d/proj3d.py:189: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray
return np.array([xs, ys, zs, np.ones_like(xs) ])
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1000/1000 [00:20<00:00, 48.86it/s]
I have the same problem. Have you solved it？
I never solved it since it was fairly easy to use Gazebo/Rviz as visualizers instead. I recommend you to go that way!