Model-Driven Scheduling of Real-Time Tasks for Robotics Systems

Abstract

The correct timing is crucial for a robotics system that requires to expose a highly reliable behavior. Especially since the demand for robots, which are applied in collaborative environments, is increasing drastically, robots need to be even more reliable, and safe. In this paper, we propose a DSL to model the timing behavior of real-time sensitive componentbased robotics systems. The DSL is integrated into the CoSiMA framework, which offers the ability to model, simulate, deploy, and analyze the (timing) behavior of robotics systems on different robotic platforms. In order to show the application of the DSL, we modeled and executed an experimental system using CoSiMA, which lets the humanoid robot COMAN perform a Zero Moment Point-based walk on a straight line, to eventually compare the actual execution with the model.