Volume 59, Issue 5 - Multicore technology in the mobility domains / Guest Editor: Jürgen Becker

The crucial foundation in developing embedded safety-critical real-time systems based on multicore based platforms is the capability to combined different methodologies and approaches with a sound formal description of the system architecture. With such a system of methodologies and system architectures it is possible to develop a system satisfying all requirements and even being able to optimize the overall system towards domain specific goals. This article discusses activities in ARAMiS, which have been performed on system architecture level, including improvements of modeling existing architectures and design techniques concerning reliability, independence, segregation and determinism for deployment strategies.

Using multicore processors in safety-critical systems is a challenge as well as an opportunity. The real parallelism, which may affect synchronization and determinism, leads to a safety-challenge, because new possible interferences might arise. Additionally, redundant software execution is possible within multicore systems. In complex multicore architectures one of the most important challenges is to know the system behavior and the recognition of any variations from the normal system behavior has to be guaranteed. For those cases it is necessary to monitor several states of the system, configurations, timing, etc. To monitor such a complex system a lot of information from the inside of the system needs to be evaluated without affecting the rest of the MPSoC.

Virtualization plays an important role for embedded systems where hardware support can prove beneficial, but these systems also pose a challenge due to power, resource constraints; reliability, safety, real-time requirements; diversity of devices, and operating systems. Therefore a trade-off between flexibility, determinism and performance exists in the embedded application domain. As virtualization in software always incurs overhead due to context switching, interrupt handling, etc. the aim is to minimize the overhead and make execution more deterministic using hardware support.

In the mobility domains functions are often realized by software running on Electronic Control Units (ECU)s, which are interconnected in a network of ECU. Nowadays there is a trend to move to ECUs with multicore processors which imposes a major challenge: An increasing degree of parallelism. This additional parallelism when migrating existing software that has been developed for singlecore systems, to a multicore system, has to be used in an efficient way. Furthermore, an increase in parallelism can cause concurrent executions of software functions, which are not possible when being executed on a singlecore ECU. This makes reasoning about the software behaviour much more difficult and also poses problems during certification activities, when sufficient independence between functions needs to be ensured. Therefore tool support is absolutely necessary to address the challenges. In this article we describe methods and tools that have been developed and/or extended during the ARAMiS project to address particular challenges arising during the development of multicore systems. In general these challenges result from the increase in parallelism available in multicore architectures.

Since especially the automotive domain increasingly utilizes multi- and many-core systems, appropriate models, analyses, and tooling are required to address challenges that were nearly non existent so far. APP4MC is an open source Eclipse platform that provides AUTOSAR compliant common data models namely AMALTHEA, basic parallelization features, visualizations, and the great possibility to add any existing tooling. For example, Eclipse Capra can be added to provide comprehensive traceability throughout the development processes but any proprietary, commercial, open-source, or prototypical implementations can be integrated. The platform enables the creation and management of complex tool chains including performance simulation and validation. The entire community benefits from reduced hardware costs, faster time to market, higher quality systems, and rapid adoption. APP4MC is not retricted to the automotive domain and utilizable in robotics or generic real-time systems as well.

The human gaze provides paramount cues for communication and interaction. Following this insight, gaze-based interfaces have been proposed for human-computer interaction (HCI) since the early 90s, with some believing that such interfaces will revolutionize the way we interact with our devices. Since then gaze-based HCI in stationary scenarios (e. g., desktop computing) has been rapidly maturing, and the production costs of mainstream eye trackers have been steadily decreasing. In consequence, a variety of new applications with the ambitious goal to apply eye tracking to dynamic, real-world HCI tasks and scenarios have emerged. This article gives an overview of the research conducted by the Perception Engineering Group at the University of Tübingen.