Title | Performance Study of the Robot Operating System 2 with QoS and Cyber Security Settings |
Publication Type | Conference Paper |
Year of Publication | 2020 |
Authors | Fernandez, J., Allen, B., Thulasiraman, P., Bingham, B. |
Conference Name | 2020 IEEE International Systems Conference (SysCon) |
Date Published | Sept. 2020 |
Publisher | IEEE |
ISBN Number | 978-1-7281-5365-0 |
Keywords | composability, History, middleware, policy-based governance, pubcrawl, quality of service, reliability, resilience, Resiliency, robotic systems, robots, ROS, security, Throughput, unmanned systems |
Abstract | Throughout the Department of Defense, there are ongoing efforts to increase cybersecurity and improve data transfer in unmanned robotic systems (UxS). This paper explores the performance of the Robot Operating System (ROS) 2, which is built with the Data Distribution Service (DDS) standard as a middleware. Based on how quality of service (QoS) parameters are defined in the robotic middleware interface, it is possible to implement strict delivery requirements to different nodes on a dynamic nodal network with multiple unmanned systems connected. Through this research, different scenarios with varying QoS settings were implemented and compared to baseline values to help illustrate the impact of latency and throughput on data flow. DDS security settings were also enabled to help understand the cost of overhead and performance when secured data is compared to plaintext baseline values. Our experiments were performed using a basic ROS 2 network consisting of two nodes (one publisher and one subscriber). Our experiments showed a measurable latency and throughput change between different QoS profiles and security settings. We analyze the trends and tradeoffs associated with varying QoS and security settings. This paper provides performance data points that can be used to help future researchers and developers make informative choices when using ROS 2 for UxS. |
URL | https://ieeexplore.ieee.org/document/9275872/ |
DOI | 10.1109/SysCon47679.2020.9275872 |
Citation Key | fernandez_performance_2020 |