Camera images are always delayed 2 rendering frames

[slowturtle0909]

I am working on a standalone simulation and I found that the camera images (including depth images and point clouds) are always delayed 2 rendering frames. For example, if rendering frequency is 10Hz, the rendered image is 0.2s delayed image. If the camera is moving in 1m/s, the image captured from 0.2m behind the actual camera position is obtained. Since the images are delayed, timestamps of them are not matched with their actual rendered time. This problem is making inaccurate and biased results when I use the simulation for visual odometry or navigation. How can I resolve this problem?

The posts in the link below seems to be talking about the result of the same issue.

ROS propper way to get camera pointcloud and jointState at the same time

Hi, I’ve got scene with robotic arm and camera mounted on end-effector in Isaac sim. I’m using standalone application and ROS clock to provide simulation time. I want to get exact joints position at which camera acquired an image. (and than do FK to obtaion exact camera position so I can transform camera pointcloud). Currently I have ros node which subscribes to camera and jointstate and subscribers callback just save message to memory. In case service for getting image with jointstate is call…

Camera point cloud shake when rotate

Ask a question, I found that the camera point cloud data output by isaacsim2.1 through the ros plug-in has a large shaking error when the robot rotates. What is the reason?I Established tf tree using posetree plugin , put the lidar and camera point clouds in the world coordinate system in rviz at the same time for display, finally lidar point cloud is stable, and the camera point cloud robot is stable when it moves in a straight line,but when it rotates, It keeps shaking; the video below is …

[ahaidu]

Manual world.render() calls might act as a temporary workaround for this: Problem with images I get from cameras - #3 by alempereur