We have attempted to write a white paper with the goal of explaining how to do fair and unbiased Performance Testing in ROS 2 based on the performance testing framework that we have built at Apex.AI and the experience that we have gathered in the past 1.5 years.
In the paper, we first explain the factors that influence the performance of ROS 2 and the importance they have in performance evaluation of the system. Next, we present a standardized test setup and discuss a working implementation of an end-to-end performance testing system that will guarantee an unbiased, and reproducible evaluation.
Please find the link to our blog post on the paper here: https://www.apex.ai/post/performance-testing-in-ros-2
Any possible feedback is very welcome. We would love to know if you find information in this paper useful, anything which we can improve and other suggestions.
Thank you !