Search

Embedded systems, e.g., self-driving systems and advanced driver-assistance systems (ADAS), require computing platforms with high computing power and low power consumption. Multi-/many-core platforms satisfy these requirements effectively. However, for hard real-time applications, multiple demands on shared resources can impede real-time performance, and memory is one resource that can impair the desired performance significantly. Therefore, it is important that memory access timing be deterministic to facilitate predictability. To realize this, the Logical Execution Time (LET) paradigm is currently attracting attention. This paper proposes a theoretical scheduling method for a model applying the LET paradigm to directed acyclic graph (DAG) nodes for a multi-/many-core platform. The proposed method considers communication timing between nodes and generates a schedule that does not cause communication contentions. In addition, the proposed method attempts to distribute tasks and reduce LET intervals to address increased execution times due to the implementation of the LET paradigm. In the evaluation, we observed that the proposed method improved the schedule length by up to 40%.

The application of cyber-physical systems (CPSs) in safety-critical application domain requires rigorous verification of their functional correctness and safety-relevant properties. We propose a practical verification process which enables to conduct safety verification of safety critical CPSs. The verification process consists of (a) a system model construction method, which generates a system model by combining software described in C and plant model code reused from model-based development, (b) a model transformation method, which transforms the plant models including differential algebraic equations (DAE) to approximate models without DAE to reduce verification complexity induced by DAE solver execution, (c) a model simplification framework, which enables the simplification of bond-graph plant models using domain-knowledge-based replacement of complex model components for further verification overhead reductions, and (d) a formal verification based on symbolic execution. We implemented the proposed methods and framework, and successfully applied the proposed verification process for safety verification of automotive brake control systems. The results of the study demonstrate that the verification detects a complex failure condition in a real-world brake control system from the generated system model and that the automated model transformations of the CPS models yield significant verification complexity reductions without impairing the ability to detect unsafe behavior. © 2017 Information Processing Society of Japan.

Driverless autonomous driving systems require cost-effective architecture satisfying design diversity and real-time performance to fulfill the fail-operational requirements that sustain system safety if a failure occurs during automated driving. However, conventional approaches cannot be applied to the systems due to design diversity. A key challenge in establishing a cost-effective multi-mode architecture is how to enhance the real-time capability of the mode switch. In this work, we propose three replication methods for fail-operational autonomous driving systems with design diversity: Input Backup Replication (IBR), Extended Primary Backup Replication (E-PBR), and Extended Leader Follower Replication (E-LFR). These methods enable accelerated recovery processing by utilizing input data and internal state backup in addition to partial hot standby. We implemented an autonomous driving prototype and found that (i) the proposed replication methods can satisfy the performance requirements for fail-operational systems, (ii) they can reduce 53.8 % of the CPU load compared with the hot standby approach in the normal mode, and (iii) the memory consumption ratio caused by the proposed methods is 0.01%. These results demonstrate that our proposed replication methods are feasible for fail-operational autonomous driving systems with design diversity.

Highly automated and autonomous driving is a major trend and vast amounts of effort and resources are presently being invested in the development of corresponding solutions. However, safety assurance is a concern, as established safety engineering standards and methodologies are not sufficient in this context. In this paper, we elaborate the fundamental safety engineering steps that are necessary to create safe vehicles of higher automation levels. Furthermore, we map these steps to the guidance presently available in existing (e.g., ISO26262) and upcoming (e.g., ISO PAS 21448) standards and point out open gaps. We then outline an approach for overcoming the identified deficiencies by integrating three different safety engineering disciplines. This includes (1) creating a safe nominal behavior specification; (2) dealing with functional insufficiencies, and (3) assuring the related performance wrt. functional safety. We exemplify our proposed methodology with a case study from industry.

The application of cyber-physical systems (CPSs) in safety-critical applications requires rigorous verification of their functional correctness and safety-relevant properties. We propose a practical verification framework which enables to fill the gaps between model-based development and the formal verification process seamlessly connecting them. The verification framework consists of (a) a model transformation method, which automatically transforms the plant models of CPSs including differential algebraic equations (DAE) to equivalent models without DAE to reduce verification complexity induced by DAE solver execution, and (b) a model simplification method, which automatically simplifies bond-graph models by replacing complex bond-graph components with simpler components for further verification overhead reductions. We successfully applied the proposed verification framework for safety verification of an automotive brake control system. The results of the study demonstrate that the automated model transformations of the CPS models yield significant verification complexity reductions without impairing the ability to detect unsafe behavior of the brake control system in a formal verification based on symbolic execution.