Search Results

Model-Based Design with production code generation has been extensively utilized throughout the automotive software engineering community because of its ability to address complexity, productivity, and quality challenges. With new applications such as lane departure warning or electromechanical steering, engineers have begun to consider Model-Based Design to develop embedded software for applications that need to comply with safety standards such as IEC 61508. For in-vehicle applications, IEC 61508 is often considered state-of-the-art or generally accepted rules of technology (GART) for development of high-integrity software [6, 11]. In order to demonstrate standards compliance, the objectives and recommendations outlined in IEC 61508-3 [8] must be mapped onto processes and tools for Model-Based Design. This paper discusses a verification and validation workflow for developing in-vehicle software components which need to comply with IEC 61508-3 using Model-Based Design.

Methods for controlling execution order in traditional text-based languages such as C and Fortran are well established. The transition to graphical programs has revealed some of the hidden issues inherent in any scheduling routine, specifically data dependency and data protection (in multirate systems). Graphical programming languages provided built-in diagnostics that allow users to analyze the data dependencies to develop optimal schedules from a data propagation perspective. This paper examines one heuristic that can be used to develop an optimal schedule for an arbitrary model.

A key step in Model-Based Design is the deployment of an algorithm as machine code onto a target processor in the production vehicle. Modern software tools automatically generate the algorithmic source code from models. Given the many combinatorial possibilities for realizing a given algorithm within the modeling environment, the generated C source code will be a function of a realization. This dependency is an important consideration because the quality and clarity of the source code impacts the amount of verification and analysis that must be done for production software development. Other factors involved in generating the machine code from the source code, such as compiler optimization and microprocessor architecture, also contribute to this optimization. Organizations that proactively data mine and gather these optimizations into a set of best practices stand to benefit from reduced development times and lower costs.

The steady growth in the number of electronic control units on the average vehicle and the complexity of the algorithms that reside on these controllers has resulted in one of the most significant initiatives in the automotive industry in years. AUTOSAR - the Automotive Open System Architecture - has united more than 100 companies, automobile manufacturers, suppliers and tool vendors to develop a standard architecture for electronic control units. By the end of 2006 Version 2.1 was released, and now OEMs as well as suppliers have started to develop and integrate AUTOSAR-compliant functionality and components into vehicles. This paper will focus on the approach and challenges faced by engineers developing AUTOSAR-compliant production code using Model-Based Design.

Current trends in the automotive, aerospace and other industries are resulting in the development of exceptionally large system-level models of control law or physical system behaviors. This is especially true in the propulsion systems algorithm and software development areas. The MathWorks has been actively engaged with a number of engineering groups faced with industry demands of increased software feature content with higher complexity and a shorter turn-around time. This paper discusses a roadblock encountered on a typical propulsion system software simulation due to its sheer size. It also discusses challenges, results, and lessons learned in solving the above problem from both a tool technology and engineering perspective.