@techreport{PfallerMottok, type = {Working Paper}, author = {Pfaller, M. and Mottok, J{\"u}rgen}, title = {Developing the e-learning class 'practical training in c'}, language = {en} } @inproceedings{DeNiklasRooneyetal., author = {De, Sangita and Niklas, Michael and Rooney, Brian and Mottok, J{\"u}rgen and Brada, Premek}, title = {Towards Translation of Semantics of Automotive Interface Description Models from Franca to AUTOSAR Frameworks}, series = {2019 International Conference on Applied Electronics (AE), 10-11 Sept. 2019, Pilsen, Czech Republic}, booktitle = {2019 International Conference on Applied Electronics (AE), 10-11 Sept. 2019, Pilsen, Czech Republic}, publisher = {University of West Bohemia}, doi = {10.23919/AE.2019.8867018}, pages = {1 -- 6}, abstract = {The automotive industry is eventually evolving into a complex network of services. The heterogeneous and distributed nature of automotive software systems demands flexible software components which can operate in different environments. Because of heterogeneous automotive development environments, the domain experts, must cope with too many diversities, adaption layers, and incompatibilities to design applications for the current generation of autonomous driving vehicles. In this context, interface adaptation is a promising approach to achieve flexibility without directly changing the respective components. AUTOSAR, which is the de-facto standard for describing automotive system architecture and is a hugely comprehensive standard allowing designers full control from abstract system description to bare metal level deployment. However, the vehicle subsystems have still evolved to include multifarious high-level domains not covered by AUTOSAR e.g. Infotainment, Telematics etc. Therefore, it seems beneficial to bridge the semantic gaps between AUTOSAR applications and other standards of automotive application domains. The goal of this paper is to investigate interface semantic mapping and achieve transparent integration of domain-specific applications using the translation of semantics among the AUTOSAR platform software component models and other software components models of open source development platforms e.g. GENIVI. A key goal of such a modelling approach is the reuse of existing interface description languages and respective code generators. This will enhance future interoperability and decrease in incompatibility among these platforms.}, language = {en} } @inproceedings{RudolphVogetMottok, author = {Rudolph, Alexander and Voget, Stefan and Mottok, J{\"u}rgen}, title = {A consistent safety case argumentation for artificial intelligence in safety related automotive systems - An Evaluation of a New Conceptual Functional Safety Approach}, series = {Embedded Real-Time Software and Systems Conference (ERST²), Jan 2018, Toulouse/France}, booktitle = {Embedded Real-Time Software and Systems Conference (ERST²), Jan 2018, Toulouse/France}, abstract = {Regarding the actual automotive safety norms the use of artificial intelligence (AI) in safety critical environments like autonomous driving is not possible. This paper introduces a new conceptual safety modelling approach and a safety argumentation to certify AI algorithms in a safety related context. Therefore, a model of an AI-system is presented first. Afterwards, methods and safety argumentation are applied to the model, whereas it is limited to a specific subset of AI-systems, i.e. off-board learning deterministic neural networks in this case. Other cases are left over for future research. The result is a consistent safety analysis approach that applies state of the art safety argumentations from other domains to the automotive domain. This will enforce the adaptation of the functional safety norm ISO26262 to enable general AI methods in safety critical systems in future.}, language = {en} } @inproceedings{MuchaMottokDeubzer, author = {Mucha, Matthias and Mottok, J{\"u}rgen and Deubzer, Michael}, title = {Probabilistic worst case response time estimation for multi-core real-time systems}, series = {2015 4th Mediterranean Conference on Embedded Computing (MECO), 14-18 June 2015, Budva, Montenegro}, booktitle = {2015 4th Mediterranean Conference on Embedded Computing (MECO), 14-18 June 2015, Budva, Montenegro}, pages = {31 -- 36}, abstract = {Traditional methods rely on Static Timing Analysis techniques to compute the Worst Case Response Time for tasks in real-time systems. Multi-Core real-time systems are faced up with concurrent task executions, semaphore accesses, and task migrations where it may be difficult to obtain the worst case upper bound. A new three staged probabilistic estimation concept is presented. Worst Case Response Times are estimated for tasksets which consist of tasks with multiple time bases. The concept involves data generation with sample classification and sample size equalization, model fit and Worst Case Response Time estimation on the basis of extreme value distribution models. A Generalized Pareto Distribution model fit method which includes threshold detection and parameter estimation is also presented. Sample classification in combination with the new Generalized Pareto Distribution model fit method allows to estimate Worst Case Response Times with low pessimism ranges compared to estimation methods that uses the Generalized Pareto or the Gumbel max distribution without sample classification.}, language = {en} }