Home // ADAPTIVE 2020, The Twelfth International Conference on Adaptive and Self-Adaptive Systems and Applications // View article

Adaptation of Schedules and Scheduling Parameters in Cybernetic Systems

Authors:
Uwe Baumgarten
Andreas Heimrath
Joachim Froeschl
Jan Ahlbrecht

Keywords: Software Cybernetics; Microkernel; Adaptive Scheduling

Abstract:
Abstract—In the automotive domain, components are increasingly based on software, that is being integrated through fewer and more powerful connected computing systems. With the main focus being the control of hardware with actuators and the monitoring of sensors, such systems are inherently complex. Additionally, their functions have both static and dynamic dependencies. Functions appear in a huge number with a large variety of dependencies. For decades, cybernetic approaches are being used to handle complexity of dynamic systems in a natural manner. It has shown effective in different scientific domains like economics and management, governance, social sciences, and technical systems. Based upon the experience with a flexible energy and power management in the automotive industry (fEPM) we transfer those findings to software systems of sophisticated cars integrating more and more functions and components. We have identified Stafford Beer’s Viable System Model (VSM) as a promising approach. We follow this approach step by step, starting with a basic VSM-based system/component. Using the principle of recursivity, we build horizontal and vertical combinations of systems resulting in powerful computational nodes with dynamic load depending on the individual feedback control systems and the control loops of all layers. Having the whole software stack in mind, we started to investigate modern microkernels (like seL4, Fiasco.OC and other) with their components. The scheduling, beside others, is the most promising component, where a VSM structure can be established. Here, we introduce an adaptive scheduler for the scheduling of an automotive workload, in which a machine learning process could exist. This is in line with other approaches in the field of autonomic or organic computing. A simplified energy and power management serves as an example for illustrating the applicability and usefulness of Beer’s VSM in technical software systems.

Pages: 105 to 111

Copyright: Copyright (c) IARIA, 2020

Publication date: April 26, 2020

Published in: conference

ISSN: 2308-4146

ISBN: 978-1-61208-781-8

Location: Nice, France

Dates: from October 25, 2020 to October 29, 2020