David P. Gluch

On a partnership between software industry and academia

This paper discusses a role for industry in software engineering education, specifically presenting a university-industry partnership between the Cardiac Rhythm Management (CRM) organization at the Guidant Corporation and Embry-Riddle Aeronautical University (ERAU). The focus of the partnership is technology transition. The partnership involves fostering students professional development, providing students experience solving realworld problems, and exploring modern directions of software engineering. The critical component of the partnership is a student-oriented research laboratory. After discussing the background and history of the project, we focus on the partnerships accomplishments. These include facilitating the transition of graduates from student to employee by developing in them extended software engineering skills and in-depth understanding of the application domain.

Pattern-Based Analysis of an Embedded Real-Time System Architecture

The emerging Society of Automotive Engineers (SAE) Architecture Analysis & Design Language (AADL) standard is an architecture modeling language for real-time, fault-tolerant, scalable, embedded, multiprocessor systems. It enables the development and predictable integration of highly evolvable systems as well as analysis of existing systems. This paper discusses the role and benefits of using the AADL in the process of analyzing an existing avionics system. We use the AADL to describe architecture patterns in the system being analyzed and to identify potentially systemic issues in the system. We discuss some of the findings related to timing, scheduling, and fault tolerance and the benefits of the use of the AADL. Additionally we highlight the benefits of working with architecture abstractions that are reflected in the AADL notation, in particular the separation of architecture design decisions from implementation decisions. Such a light-weight architecture analysis is typically followed by a full-scale AADL model of the system with required and actual timing, performance, and reliability figures, and its analysis to determine whether the requirements are met.

On Modelling, Simulating and Verifying a Decentralized Mission Control Algorithm for a Fleet of Collaborative UAVs

Abstract It can be relatively easy to correctly design a centralized mission control algorithm for a fleet of UAVs to achieve optimal mission efficiency. However, the fault intolerance of an algorithm can put the fleet at risk for missions in a noisy communication environment. This paper presents a Decentralized Mission Control (DMC) algorithm for coordinating a fleet of UAVs to accomplish a specific mission. The symbolic model of the UAV fleet configuration and the software processes are applied to the Berkley UAV systems described in [1] and [10] , which facilitate peer-to-peer communication and environmental awareness. The design goal of DMC algorithm is that the UAVs work either cooperatively to achieve the highest efficiency under normal communication modes or adaptively to guarantee the safety of the UAVs under various fault modes. The communication protocol of DMC algorithm mostly relies on message broadcasting in order to minimize the dependency on a ground station. The task assignment schema of the DCM algorithm depends on the modes of the environment variables. It is derived from the Hungary Algorithm [11] , which attains the optimal assignment for equal numbers of tasks and agents. A formal method is used to verify the safety and progress properties such as tolerance to faults and freedom from deadlocks. The real-time interactions of the system are modeled as a nexus of UPPAAL automata and are verified against expected properties specified as a list of temporal CTL queries. A C# program is used to simulate and measure the mission efficiency.

Automated code generation for safety-related applications: a case study

This paper addresses issues relating to the suitability of using automated code generation(ACG) technologies for the development of real-time, safety-critical systems. This researchexplored the characteristics of model-based software development methodologies and the automatedcode generation tools that support them. Specifically, data related to the engineeringchallenges, skills, and effort associated with ACG practices and technologies were collectedas part of a case study. Characteristics such as the generated code’s organization, size, readability,traceability to model, real-time constructs, and exception handling were identified. Inaddition, the case study involved software engineering practices that incorporate integratedanalysis and design iterations throughout a model-based development process. The researchinvestigated both the static and dynamic characteristics of the selected techniques and tools,identified characteristics of ACG tools with potential impact on safety, and considered thesemantic consistency between representations.

A study of automatic code generation for safety-critical software: preliminary report

Modern safety-critical systems (e.g., combined pacemaker/deliberator devices, distributed patient therapy delivery systems) incorporate more functionality than similar devices of the past. The development of these complex systems challenges existing quality assurance techniques; results in significantly longer development times; and demands greater staffing resources to ensure quality and timely product completion. This is an interim report on a case study of the efficacy and viability of automatic code generation (ACG) techniques applied in the development of real-time, safety-critical software-dependent systems (Whalen, 1997). The research uses model-based software engineering (MBSE) practices that incorporate integrated analysis and design iterations throughout the development process. The focus of these investigations is the application of automated code generation tools that embody various methodologies, in the development of safety critical systems. There was no attempt to embark on explicit tool comparisons or evaluations.

Embedded systems engineering with the AADL: modeling & analysis

The SAE Architecture Analysis & Design Language (AADL) is an architecture description language for real-time, fault-tolerant, scalable, embedded, modular multiprocessor systems. It enables the development of highly evolvable systems, early and quantitative analyses of a systems architecture, and evolution of an architecture model for continued analysis throughout the lifecycle. In this tutorial, we provide an overview of the AADL; demonstrate the AADLs capabilities in creating and analyzing component-based models of the task and task interaction architectures of embedded software; discuss interfacing to physical devices; highlight AADL capabilities for predictive analyses of operational characteristics such as meeting deadline, response time, and throughput requirements; and describe how the AADL can discover system integration problems early in a development effort.

Model Checking for Dependable Software-Intensive Systems

Model checking is indispensable in the development of modern digital circuitry and is emerging as a valuable instrument for software verification. Model checking has uncovered errors in a variety of software-intensive systems, including spacecraft redundancy management, aircraft collision avoidance, and weapons control systems. The approach offers the potential to help ensure behavioral properties and eliminate catastrophic errors in software systems that require high levels of dependability.