Message flow analysis with complex causal links for distributed ROS 2 systems

Distributed robotic systems rely heavily on the publish–subscribe communication paradigm and middleware frameworks that support it, such as Robot Operating System (ROS), to efficiently implement modular computation graphs. The ROS 2 executor, a high-level task scheduler which handles messages, is performance bottleneck. We extend ros2_tracing, framework with instrumentation tools for real-time tracing of 2, analysis visualization flow messages across distributed systems. Our method detects one-to-many many-to-many causal links between input output including indirect through simple user-level annotations. validate our both synthetic real systems, demonstrate its low runtime overhead. Moreover, underlying intermediate execution representation database can be further leveraged extract additional metrics results. This provide valuable timing scheduling information study improve executor well optimize any system. source code available at: github.com/christophebedard/ros2-message-flow-analysis.