TSDSIR: multi-service concurrent message data flow scheduling in robotic systems

Abstract

Aiming at the characteristics of multi-service concurrency and multi-modal message data mixing faced in the robot cloud-edge-end collaboration scenarios, a multi-message dynamic scheduling method oriented to time-sensitive attributes was proposed based on the analysis of the message distribution mechanism in the robot operating system 2 (ROS2). Firstly, scheduling parameters, including message priority and transmission times, were defined based on multi-topic concurrency and real-time network state in ROS2, enabling real-time queue state characterization. Then, a hybrid time/priority scheduling model was designed with specific switching rules and boundary conditions. Finally, the dynamic scheduling method, considering time sensitivity, was integrated into ROS2's data distribution service layer. Simulation experiments and practical applications verify that the proposed method can improve the efficiency of concurrent transmission of multiple message streams in ROS2, and at the same time guarantee the real-time reliable transmission of high-priority message data with time-sensitive attributes.

Keywords

• ROS2;data distribution service;time-sensitive attribute;hybrid switching;multi-message flow scheduling