Gergely, I. A. (2023). Resource management of automotive engine control units [Diploma Thesis, Technische Universität Wien]. reposiTUm. https://doi.org/10.34726/hss.2024.107301
Embedded systems play a crucial role in the contemporary automotive industry, where advancedvehicles integrate numerous electronic control units responsible for diverse functionalities, fromABS to sophisticated driver assistance systems.This master thesis specifically focuses on the engine control unit within the powertrain. Withincreasingly stringent regulations on emission values and a growing functional environment,Robert Bosch AG anticipates that the resources of an ECU will reach their limits within thenext 3-5 years. The current development workflow relies on resource optimizations through atrial-and-error approach, resulting in moderate success with high manual effort. In response, wepropose a methodology that leverages an existing measure based on task splitting from the earlystages of ECU development. Our goal is to reduce the manual effort required by design engineersand shorten the development cycle for ECUs.The first section of this thesis delves into presenting multi-core parallelization and load bal-ancing techniques, laying the technical foundation for the feasibility of our approach. We thenprovide the technical background of the targeted system and introduce our proposed approach,accompanied by an analysis of the current resource management process to identify potentialoptimization steps.Additionally, we present a conceptual implementation using Matlab and Python to show-case the integration of our approach into the automotive development toolchain. Finally, we as-sess and validate our approach through implementation on a real project as a proof-of-concept,thereby evaluating the feasibility of our assumptions and demonstrating the effectiveness of ourproposed methodology.
en
Embedded systems play a crucial role in the contemporary automotive industry, where advancedvehicles integrate numerous electronic control units responsible for diverse functionalities, fromABS to sophisticated driver assistance systems.This master thesis specifically focuses on the engine control unit within the powertrain. Withincreasingly stringent regulations on emission values and a growing functional environment,Robert Bosch AG anticipates that the resources of an ECU will reach their limits within thenext 3-5 years. The current development workflow relies on resource optimizations through atrial-and-error approach, resulting in moderate success with high manual effort. In response, wepropose a methodology that leverages an existing measure based on task splitting from the earlystages of ECU development. Our goal is to reduce the manual effort required by design engineersand shorten the development cycle for ECUs.The first section of this thesis delves into presenting multi-core parallelization and load bal-ancing techniques, laying the technical foundation for the feasibility of our approach. We thenprovide the technical background of the targeted system and introduce our proposed approach,accompanied by an analysis of the current resource management process to identify potentialoptimization steps.Additionally, we present a conceptual implementation using Matlab and Python to show-case the integration of our approach into the automotive development toolchain. Finally, we as-sess and validate our approach through implementation on a real project as a proof-of-concept,thereby evaluating the feasibility of our assumptions and demonstrating the effectiveness of ourproposed methodology.
de
Additional information:
Abweichender Titel nach Übersetzung der Verfasserin/des Verfassers
