Home // International Journal On Advances in Systems and Measurements, volume 12, numbers 3 and 4, 2019 // View article

HW/SW Co-Design Approach to Optimize Embedded Systems on Reliability

Authors:
Andreas Strasser
Philipp Stelzer
Christian Steger
Norbert Druml

Keywords: Safety critical systems; Aging of circuits and systems; Safety Validation HW/SW; Failure-in-Time Analysis; Algorithm Safety Evaluation

Abstract:
Autonomous driving is disruptively changing the automotive industry. The importance of safety, reliability, and fault-tolerance is steadily increasing through the complexity and autonomy of self-driving cars. In the past, developers relied on the driver as a fail-safe backup to transfer the control and the responsibility to him in case of unexpected faults. In fully autonomous vehicles this backup solution will be not available anymore. This requires novel safety concepts and methodologies such as an optimization of high reliability of the systems. For optimization it is necessary to quantify different algorithm solutions from a safety point of view because this enables the possibility of comparing different solutions. In this publication, we are analyzing the consequences of different hardware and software algorithm implementations on component reliability. For this purpose we have designed two novel algorithm safety validation methodologies that allow the quantification of algorithms from a safety point of view and applied them to two independent use cases to evaluate the effects on component reliability. Both methodologies are used for optimizing the reliability of safety-critical automotive embedded systems for autonomous driving during Hardware/Software Co-Design.

Pages: 158 to 168

Copyright: Copyright (c) to authors, 2019. Used with permission.

Publication date: December 30, 2019

Published in: journal

ISSN: 1942-261x