Render screen is out of sync

[Qingliang-Li]

Render screen is out of sync

[qgallouedec]

I can't reproduce the issue. Here is the screen recording of OpenGL GUI (with the correct fps):

Screen.Recording.2023-10-12.at.12.10.43.mov

And the gif generated (also with the correct fps):

[Qingliang-Li]

I can't reproduce the issue. Here is the screen recording of OpenGL GUI (with the correct fps):

Screen.Recording.2023-10-12.at.12.10.43.mov
And the gif generated (also with the correct fps):

If you add a few more frames to the GIF, you will notice that the robot arm stops moving in the later frames. In this video, you can observe that the robotic arm remains stationary during the second, third, and fourth seconds, although it should be in motion.

[Qingliang-Li]

If you add a few more frames to the GIF, you will notice that the robot arm stops moving in the later frames. In this video, you can observe that the robotic arm remains stationary during the second, third, and fourth seconds, although it should be in motion.

[qgallouedec]

It moves slowly (because the desired position you've defined evolves slowly), but it's not stationary.

[qgallouedec]

If you change a bit your desired positions, here is what you get:

import gymnasium as gym
import panda_gym
import matplotlib.pyplot as plt
from matplotlib import animation
import numpy as np


def display_frames_as_gif(frames):
    patch = plt.imshow(frames[0])
    plt.axis("off")

    def animate(i):
        patch.set_data(frames[i])

    anim = animation.FuncAnimation(plt.gcf(), animate, frames=len(frames), interval=5)
    anim.save("panda.gif", writer="pillow", fps=25)


env = gym.make("PandaReach-v3", render_mode="rgb_array", renderer="OpenGL")
observation, info = env.reset()

z = observation["observation"][2]

theta = np.linspace(0, 20 * np.pi, 1000)
r = 0.1
desired_positions = np.stack([r * np.cos(theta), r * np.sin(theta), z * np.ones(1000)], axis=1)

frames = []
for i in range(200):
    frames.append(env.render())
    current_position = observation["observation"][0:3]
    desired_position = desired_positions[i]
    action = 5.0 * (desired_position - current_position)
    observation, reward, terminated, truncated, info = env.step(action)

env.close()
display_frames_as_gif(frames)

[Qingliang-Li]

I get it. Thank you very much，Gallouedec.