Originally published at: https://developer.nvidia.com/blog/building-collaborative-robotics-using-ros-and-isaac-sdk/
The Robot Operating System (ROS) offers many software libraries and tools to help build robot applications, including framework, algorithms, sensors, and robot platforms. It is extremely popular with roboticists and researchers everywhere. A continuous influx of packages and libraries from the ecosystem makes it easy to get started on robotics projects. Figure 1. Using the…
Hi Guys.
I am having hard times locating the files for execution of the command
bazel run packages/ros_bridge/apps/ros_to_superpixels
at my 2020.1 NX installation there is no file with name ros_to_superpixels.
Maybe there is a newer version of the IsaaC SDK that includes it?
Otherwise how to get it installed? Is there a specific file that needs to be created in order to execute the proposed command above? Which exactly file with which content? could you supply the missed file, please?
Thanks
Hello Andrey,
Sorry we forgot to mention: As you have suspected, ros_to_superpixels application will be part of the upcoming release (soon). In the mean time, ros_to_navigation app should be functional in earlier releases including 2020.1. Also, RosToImage codelet released by source shows how to convert ROS images to the Isaac format. Application using RosToImage will be shared with 2020.2.
Best,
Oguz
@OguzSaglam
Thank you for following up.
Maybe you could suggest quick patch for manual integration of the ros_to_superpixels application into 2020 nx release? like to copy code and paste it to files?
When is this release expected ?
2020.2 goes out in a week or so. stay tuned; you should receive an email soon.
how do we deploy the app to the jetson?
how do we get it deployed in the released release?
Hi Andrey,
Looks like there is a typo in the instructions. Please replace “ros_to_superpixel” with “ros_to_superpixels”. Thanks for pointing this out.
Best,
Oguz
1 Like
Fixed the typo!
what we can run on Host PC is
bazel run packages/ros_bridge/apps/ros_to_superpixels
it works with realsense d435.
now the concern, however is to find the command to deploy it on jetson
could you extend on this, please?
cause there is no deployment file for robot[jetson] as far as I can see
Glad it worked on your host PC Andrey! For Jetson, please see the section on the blog that starts with the following text.
Now, deploy your application on your Jetson device from the development workstation:
2 Likes
Could we have a helping hand getting zed2 to work with ros bridge, please? In a manner similar to the realsense implementation? Somehow? Thanks
ros_to_superpixels application shows how to receive images from ROS. It doesn’t assume a realsense camera. Isaac also has camera drivers including v4l2, realsense, and Zed. I believe the current implementation of the superpixels requires depth data, but you could run other perception algorithms similarly if your camera doesn’t give depth.
Hi Andrey,
ros_to_superpixels requires depth topic from ROS. ROS Zed/Zed2 publishes depth topic in 32FC1 format, which is not currently supported by Isaac ros bridge depth converter. However you can try to update Isaac ros bridge depth converter for 32FC1 or any other format.
Thanks !!