We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Resource-aware integration of AUTOSAR-compliant ECUs with an empirical wcet prediction model.
- Authors
Park, I.; Chung, J.; Youn, J.; Lee, W.; Sunwoo, M.
- Abstract
This paper presents an integration method of AUTOSAR-compliant ECUs which can evaluate resource constraints in an early-stage of development. There are three types of resources for an ECU (timing, memory, and interface) which should be carefully managed for successful ECU integration. The proposed method consists of three steps: measurement, prediction, and evaluation. In the first step, a method to measure resource factors for AUTOSAR-compliant software architecture is introduced. Based on the method, a worst-case execution cycle of a runnable, memory section usages of a software component, and interface of legacy ECUs can be obtained. In the second step, the obtained factors are quantitatively predicted according to the architectural designs of the integration ECU. In the case of the timing resource, the worst-case execution time of the integration ECU can be precisely predicted by a proposed empirical model. In the last step, the resource constraints such as CPU, memory, network utilizations can be evaluated with predicted resource factors before implementation. The proposed method was applied to the integration of an in-house engine management system composed of two ECUs. The method successfully provided quantitative measures to evaluate architectural designs of three different integration scenarios.
- Subjects
SOFTWARE architecture; PREDICTION models; CONFIGURATION management software; AUTOMOTIVE electronics; WORST-case circuit analysis; REAL-time computing
- Publication
International Journal of Automotive Technology, 2016, Vol 17, Issue 4, p717
- ISSN
1229-9138
- Publication type
Article
- DOI
10.1007/s12239-016-0071-4