Guoqing Yang

SmartOSEK: a real-time operating system for automotive electronics

This paper puts forward SmartOSEK, a dependable platform for automobile electronics, which consists of an operating system compliant with OSEK/VDX specifications and an integrated development environment (IDE) that consists of many convenient tools, such as visual designer, system generator, time analyst, scheduling analyst, and running tracer. In the operating system, SmartOSEK OS and SmartOSEK COM are presented. In the IDE, we apply the graphic design, automatic code generation and time analysis to help developers devote their mind to the modeling of the applications. A design example of automated transmission system based on SmartOSEK platform is given, and good results are achieved.

Model-based design and verification of automotive electronics compliant with OSEK/VDX

Model-based approaches are gradually applied in embedded system design with Unified Modeling Language (UML) and its profiles, but in terms of automotive electronics domain, few developers adopt UML to design system models because of inadequate tools that support the domain-specific modeling. This paper puts forward a model-based approach for automobile electronics software design and verification with a dependable platform compliant with OSEK/VDX standard. In addition, a case study is presented to demonstrate the application of the approach. The contribution of the approach is threefold. First, the approach applies the theory of model-based design with OSEK/VDX standard in automotive electronics domain. Second, the approach solves the transformation between UML models and OSEK/VDX models through an efficient method. Third, the approach simulates the system models and provides the designer with the results to optimize the design at design-level.

An improved method of task context switching in OSEK operating system

Improving real-time is one of the most essential problems in studies of real-time operating system (RTOS). The time of task context switching is an important factor that affects the real-time of RTOS. This paper concentrates on the improvement of task context switching of one kind of RTOS-OSEK operating system. According to different task states, we apply different context switching strategies to reduce the average time of context switching. This method has been successfully implemented and evaluated in our OSEK compatible operating system-SmartOSEK OS.

SVE: embedded agent based smart vehicle environment

Smart space is one of the important applications of pervasive computing. We put the concept of smart space into the field of intelligent transportation system, and bring forward a framework of smart vehicle environment based on embedded agent. In this paper we discuss the structure and characteristics of SVE and an example of SVE is given. Its trait is utilizing embedded hardware and context-aware computing technology to unite all of the electronic equipments into a smart space, thus we can provide active, attentive and comprehensive service including auto drive for the users. A key innovation of the paper is the proposal of embedded agent. We have brought into further discussion its application mode.

SmartOSEK Based Design and Verification for In-vehicle Network System: A Model-Based Approach

Electronic control units (ECUs) are widely used to improve the comfort and reliability of vehicles. Due to the increasing degree of distribution and interaction of ECUs, many approaches are adopted to design the in-vehicle networks to ensure the stability and reliability of the holistic network, but none of them support the synchronous development of in-vehicle networks and software in ECUs. This paper put forward a model-based approach to design the in-vehicle networks and the software in ECUs with presenting a dependable platform compliant with OSEK/VDX specifications. The paper presents a tool-chain that covers the entire system development life-cycle including system modeling, model transformation, model analysis, network simulation, code generation, document generation and runtime instrumentation. In addition, a case study is presented to demonstrate the application of the approach. The contribution of the approach is threefold. First, the approach applies the theory of model-based design with OSEK/VDX standard in automotive electronics domain, and implements the transformation between SysML models and OSEK/VDX models through an efficient method. Second, the approach presents a method of software co-design among distributed ECUs. Third, the approach bring forward a simulation model for the system models with in-vehicle networks and provides the designer with the simulation results to optimize the design at the early time of the development

Schedulability analysis for fault‐tolerant group‐based preemptive scheduling

The group‐based preemptive scheduling provides a flexible mechanism to define the preemptive relations between tasks. However, this scheduling scheme together with a resource access synchronization protocol and the requirements of fault tolerance makes the predication of a real‐time system’s behaviors more difficult than traditional scheduling scheme. The major contribution of this work is an algorithm to calculate the worstcase response time for tasks under the group‐based preemptive scheduling. This algorithm supports both the fault‐free and the primary‐alternative fault‐tolerance scheduling mechanism. Moreover, a method to calculate the minimum allowed time between two consecutive faults is also introduced. The algorithm has been implemented in a time analysis tool and integrated into a system development platform, which is compatible with the OSEK/VDX operating system standard, to verify the temporal property in the early system design step.

Fault tolerant scheduling for fixed-priority tasks with preemption threshold

Fixed-priority with preemption threshold (FPPT) is an important form of real-time scheduling algorithm, which fills the gap between fixed-priority preemptive (FPP) and fixed-priority non-preemptive (FPNP). However, this scheduling scheme together with the requirements of fault tolerance makes the prediction of a real-time systems behaviors more difficult than traditional scheduling scheme. The major contribution of this work is twofold. First, we present an appropriate schedulability analysis, based on response time analysis, for supporting fault-tolerant FPPT scheduling in hard real-time systems. The error-recovery techniques are considered to be used to carry out fault tolerance. Second, we propose the optimal priority assignment algorithm which can be used, together with the schedulability analysis, to improve system fault resilience.

SmartC: A Component-Based Hierarchical Modeling Language for Automotive Electronics

This paper introduces SmartC, a language designed for programming automotive electronics embedded systems such as engine control systems. SmartC is a hierarchical modeling language and implements the SmartOSEK operating system model. The SmartC models are classified into four levels, namely module level, task level, subtask level and component level. In the SmartC models, control-flow oriented models and data-flow oriented models are integrated in the hybrid SmartC models. At the task level, the model is constructed based on the control flow, whereas the component level model is constructed based on the data flow. In SmartC programs, all inter-task communication, task triggering mechanisms, and access to guarded global variables, are automatically generated by the SmartC generator which generates the C code from the SmartC code. Having well-structured concurrency mechanisms, SmartC greatly reduces the risk of concurrency errors, such as deadlock and race conditions. The SmartC language is implemented on the automated manual transmission (AMT) control system and is compatible with the OSEK/VDX specifications. We use a continuous time (CT) model as an example to illustrate the effectiveness of the language

ASOS-IV: A Smart Sensor Network Software Platform for Intelligent Vehicles

Intelligent vehicle applications, which smart sensor network technology can be used widely, have several rigorous requirements, such as real-time constraint, low power-consumption, resource limitation, self-adaptation and robust operation. In order to facilitate the development of smart sensor network applications in intelligent vehicle systems, a smart sensor network software platform named ASOS-IV is developed in this paper, which consists of an reliable real-time OS kernel for smart sensor nodes and an energy-efficient and reliable routing protocol for wireless communications between sensor nodes. Detailed simulations and analysis results indicate that ASOS-IV is a reliable real-time software platform for intelligent vehicle applications, which can improve energy efficiency, balance energy consumption of all sensor nodes and enhance system fault tolerance.© 2005 ASME