Insight最新文献

Companies in the high-tech equipment industry are continuously looking for ways to optimize their business. A notoriously difficult part of optimizing is the R&D activities, as risks and uncertainties are inherent. In our experience, creating and using a reference architecture for a product or portfolio to guide future developments is a good way to improve R&D effectiveness and efficiency. But developing a reference architecture by capturing the relevant information and establishing the structure, the models and their interrelations, the tools, and secondly, getting clarity on how to use such reference is not easy. In this article, we describe a method to ‘distill’ a reference architecture using the knowledge built-up in years of developing products and using the customer and business values to capture the key architectural decisions for future products. We explain the purpose and usage of a reference architecture and how to organize it. The experiences obtained in Thermo Fisher Scientific have proven the importance and practicality of this approach.

Network softwarization has revolutionized the architecture of cellular wireless networks. State-of-the-art container based virtual radio access networks (vRAN) provide enormous flexibility and reduced life-cycle management costs, but they also come with prohibitive energy consumption. We argue that for future AI-native wireless networks to be flexible and energy efficient, there is a need for a new abstraction in network softwarization that caters for neural network type of workloads and allows a large degree of service composability. In this paper we present the NeuroRAN architecture, which leverages stateful function as a user facing execution model, and is complemented with virtualized resources and decentralized resource management. We show that neural network based implementations of common transceiver functional blocks fit the proposed architecture, and we discuss key research challenges related to compilation and code generation, resource management, reliability and security.

Over the past five years, with funding from the US Air Force and the US Space Force, SERC researchers at the University of Southern California's Information Sciences Institute have undertaken a series of case studies that have focused on the introduction of agile and DevSecOps practices into a space-based software-only acquisition environment. These studies have identified best practices and revealed useful lessons learned. While the initial baseline DoDI 5000.02 project was entirely waterfall-based, subsequent projects have introduced agile/DevSecOps methods in progressively increasing levels, with the second project consisting of a roughly a 50/50 hybrid agile/waterfall mix and with the current project consisting of an approximately 70/30 hybrid agile/waterfall mix effort. All projects exhibit similar code complexity and size.

In 2019, the research council of the Systems Engineering Research Center (SERC), a US Defense Department sponsored university affiliated research center (UARC), developed a set of roadmaps (SERC 2019) structuring and guiding four areas of systems engineering research: digital engineering, velocity, security, and artificial intelligence (AI) and autonomy. This paper presents the development of these roadmaps and the key underlying transformation aspects.

Modern engineered systems, and learning-based systems, in particular, provide unprecedented complexity that requires advancement in our methods to achieve confidence in mission success through test and evaluation (T&E). We define learning-based systems as engineered systems that incorporate a learning algorithm (artificial intelligence) component of the overall system. A part of the unparalleled complexity is the rate at which learning-based systems change over traditional engineered systems. Where traditional systems are expected to steadily decline (change) in performance due to time (aging), learning-based systems undergo a constant change which must be better understood to achieve high confidence in mission success. To this end, we propose pairing Bayesian methods with systems theory to quantify changes in operational conditions, changes in adversarial actions, resultant changes in the learning-based system structure, and resultant confidence measures in mission success. We provide insights, in this article, into our overall goal and progress toward developing a framework for evaluation through an understanding of equivalence of testing.

Artificial intelligence (AI) and machine learning (ML) technology are becoming increasingly critical in systems: both to provide new capabilities and in the practice of systems engineering itself, especially as digital transformation improves the automation of many routine engineering tasks. The application of AI, ML, and autonomy to complex and critical systems encourage the development of new systems engineering methods, processes, and tools. This article highlights a series of workshops conducted jointly by the US Army Combat Capabilities Development Command Armaments Center (CCDC AC) Systems Engineering Directorate and the Systems Engineering Research Center (SERC). These workshops focus on the relationships between AI and systems engineering and elicit input from hundreds of stakeholders across government, industry, and academia. They also provide critical direction to the SERC's research roadmap on AI/autonomy as it looks towards the long-term outcome of “human-machine co-learning.” Though the workshops are US-centric, the lessons and insights gained are applicable globally.

The newly established Institute of Systems Engineering for Future Mobility within the German Aerospace Center opened its doors at the beginning of 2022. Emerging from the former OFFIS Division Transportation after a two-year transition phase, the new institute can draw on more than thirty years of experience in the research field of safety-critical systems. With the transition to the DLR, the institute's new research roadmap focuses on technical trustworthiness for highly automated and autonomous systems. Within this field, the institute will develop new concepts, methods, and tools to support the integration and assurance of technical trustworthiness for automated and autonomous systems during their whole lifecycle – from the early development through verification, validation, and operation to updates of the systems in the field.

Automotive software is undergoing a rapid change toward artificial intelligence and towards more and more connectedness with other systems. For both, an incremental design paradigm is desired, where the car's software is frequently updated after production but still can guarantee the highest automotive safety standards. We present a design flow and tool framework enabling a DevOps paradigm for automotive software development. DevOps means that software is developed in a continuous loop of development, deployment, usage in the field, collection of runtime data and feedback to the developers for the next design iteration. The software developers get support in defining, developing, and verifying new software functions based on the data gathered in the field by the previous software generation. The software developers can define contracts describing the time and resource assumptions on the integration environment and guarantees for other dependent software components in the system. These contracts allow a composition of software components and proof obligations to be discharged at design time through virtual integration testing and runtime through continuous monitoring of assumptions and guarantees on the software component's interfaces. An update package, consisting of the software component and its contracts, is then automatically created, transferred over the air, and deployed in the car. Monitors derived from the contracts allow for supervising the system's behavior, detecting failures at runtime, and annotating the situation to be included in a data collection, fueling the next design iteration.