-
Notifications
You must be signed in to change notification settings - Fork 1.8k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Vision - systems failure - "overload" #2310
Comments
Hi @RodrigoCodeBernardo May I first ask the following questions, please.
If you are not using the RealSense ROS wrapper, are you using a RealSense camera as a general camera with the mentioned AR applications? |
Hi, I use a Intel Realsense 415
|
The age of the configuration in this project is one where I would strongly recommend upgrading to a more recent configuration if possible. The librealsense SDK, RealSense ROS wrapper and camera firmware driver all date back to January 2018, at the start of the launch of the RealSense 400 Series cameras. Since then, a vast amount of bug fixes, enhancements and new features have been added to librealsense, the ROS wrapper and firmware driver in the subsequent years. The age of the configuration makes it difficult to diagnose, but I would start by asking whether you have installed support for an RGBD type of launch using the instructions at #2092 (comment) (selecting the instruction for the particular ROS1 branch that you are using) as support for an RGBD launch with rs_rgbd.launch is not bundled with the RealSense ROS wrapper by default. |
This is my rs_rgbd.launch
Rodrigo Bernardo |
The framerate of of the topics I am using:
|
Seeing that kind of difference between the color and depth FPS, I would recommend trying to disable an RGB option called auto_exposure_priority (a separate option from auto_exposure) to see whether it makes a positive difference. If auto_exposure is enabled and auto_exposure_priority is disabled then it forces the streams to try to maintain a constant FPS rate instead of being allowed to vary. You could test quickly whether the problem is related to auto_exposure_priority by opening the dynamic reconfigure interface window after launch, going to the rgb_camera category of options and then finding and disabling the auto_exposure_priority setting. The section of the ROS wrapper documentation at the link below explains how to open the dynamic reconfigure interface after launch has completed by using the instruction |
I am using two machines, on my master I have no difference between the topics, I only have on my slave. This difference might be associated with the internet connection, which is over Wifi. I will change this parameter to see if it improves. |
Hi @RodrigoCodeBernardo Do you require further assistance with this case, please? Thanks! |
Case closed due to no further comments received. |
Hi,
I am starting to work with a vision for object detection. I started by testing using ar_tags detection, using the VISP library for that. So far, everything works perfectly. When I add this node to the remaining nodes of my system, I start to have communication errors; for example, the communication with the robotic arm controller is lost. My system is overloaded with too much information, and it becomes "heavy" subscribing to my RGBD camera topics.
At first, I thought the problem was in the code I wrote, but I tested it with the ar_tag ([http://wiki.ros.org/ar_track_alvar]) code, and it happened the same. I have also changed the frequency of the publishers and subscribers of the vision system node, but the problem persists.
My problem arises when I subscribe to the camera topics. It only works fine if I only use the camera.
Has anyone ever had a similar problem? My system consists of a mobile base and a robotic arm, and I use several nos. However, it only happens with the vision.
Thank you in advance for your help and attention.
The text was updated successfully, but these errors were encountered: