Connect Reliably to ROS 1 and 2 Stacks with High-Rate Data and Large Messages

[eshu]

We’ve developed a new Foxglove bridge connection that dramatically improves streaming performance, provides custom message and publishing support, and offers reliable live connections to your ROS 1 and ROS 2 stacks.

Announcing the New Foxglove Bridge for Live ROS Data - Foxglove

We discovered that Rosbridge can struggle with streaming high-frequency topics and large messages like images and point clouds. We designed Foxglove bridge to tackle these performances issues and to provide a dramatically streamlined experience for teams debugging their robots.

Streaming Intel RealSense images (1280x720 raw rgb8 images) from a NVIDIA Jetson TX2 robot. We found that Rosbridge was unable to keep up with reading, serializing, and sending all streamed data, dropping ~36% of messages.

Install the foxglove_bridge ROS package and open a Foxglove WebSocket connection in Foxglove Studio to get started!

To learn more about our new foxglove_bridge, check out our docs and the package’s GitHub repo.

As always, join our Slack community to ask questions, or reach out to us directly with any feedback.

[Timple]

Great work!

Random thought:
Given that this bridge is available for ROS1 and ROS2, would it be possible to connect two bridges?
So one could replace the ros1_bridge by something like this:

ROS1 → foxglove_bridge → foxglove_bridge → ROS2

[haudren]

Sounds very nice I am wondering how you achieved this speedup. Is it mostly due to a more efficient protocol? Or due to to the multithreading?

I also wonder if you were running rosbridge in bson mode in this benchmark?