One of the most significant advances in the development of computer science, information and communication technologies is represented by the cyber-physical systems (CPS). They are systems of ...collaborating computational entities which are in intensive connection with the surrounding physical world and its on-going processes, providing and using, at the same time, data-accessing and data-processing services available on the Internet. Cyber-physical production systems (CPPS), relying on the latest, and the foreseeable further developments of computer science, information and communication technologies on one hand, and of manufacturing science and technology, on the other, may lead to the 4th industrial revolution, frequently noted as Industrie 4.0. The paper underlines that there are significant roots in general – and in particular to the CIRP community – which point towards CPPS. Expectations towards research in and implementation of CPS and CPPS are outlined and some case studies are introduced. Related new R&D challenges are highlighted.
We construct a control law that manages to adaptively stabilize a class of linear 2×2 hyperbolic systems of partial differential equations (PDEs) from a single boundary sensing anti-collocated with ...the boundary where actuation takes place. We do this by introducing a series of invertible transformations that bring the system into an observer canonical form, for which adaptive control design becomes feasible. We establish pointwise boundedness of all signals in the closed loop system, and pointwise convergence of the system states to zero. The theory is demonstrated in a simulation.
This paper examines event-triggered data transmission in distributed networked control systems with packet loss and transmission delays. We propose a distributed event-triggering scheme, where a ...subsystem broadcasts its state information to its neighbors only when the subsystem's local state error exceeds a specified threshold. In this scheme, a subsystem is able to make broadcast decisions using its locally sampled data. It can also locally predict the maximal allowable number of successive data dropouts (MANSD) and the state-based deadlines for transmission delays. Moreover, the designer's selection of the local event for a subsystem only requires information on that individual subsystem. Our analysis applies to both linear and nonlinear subsystems. Designing local events for a nonlinear subsystem requires us to find a controller that ensures that subsystem to be input-to-state stable. For linear subsystems, the design problem becomes a linear matrix inequality feasibility problem. With the assumption that the number of each subsystem's successive data dropouts is less than its MANSD, we show that if the transmission delays are zero, the resulting system is finite-gain Lp stable. If the delays are bounded by given deadlines, the system is asymptotically stable. We also show that those state-based deadlines for transmission delays are always greater than a positive constant.
We propose a distributed formation control algorithm augmented with communication awareness. We consider autonomous underwater vehicles (AUVs) that are able to communicate over an acoustic link using ...a Time Division Multiple Access (TDMA) protocol, and to measure the Signal-to-Noise Ratio (SNR) of incoming messages. Based on the measured SNR and packet loss, we endow them with a distributed formation control scheme that accounts for the time-varying nature of the acoustic communication channel. This scheme allows a network of N AUVs to follow a pre-determined, twice-differentiable path while adapting their formation. The size of the formation is dynamically scaled by a formation adaptation mechanism to stabilize the estimated packet loss probability at a desired level. A distributed packet loss estimator is then built on top of the same average consensus routines used by the formation control algorithm, and thus comes with a minimal communication overhead. We test the algorithm by means of high-fidelity simulators, and verify its efficacy in making the network of agents retain formation-wide communication capabilities in a range of cases.
While several graph mining systems have been developed to run different graph mining algorithms on such large networks, they have difficulty processing Web-scale graphs owing to the significant ...communication and I/O expenses incurred when workers connect with one another and frequently read subgraphs. In this study, we provide a cost-effective, scalable, distributed method for mining real-world graphs by exploiting graph properties. Our technique drastically decreases the communication cost, the main bottleneck of distributed systems, by using alternate edge placement criteria depending on the types of vertices.
Trends in big data analytics Kambatla, Karthik; Kollias, Giorgos; Kumar, Vipin ...
Journal of parallel and distributed computing,
07/2014, Volume:
74, Issue:
7
Journal Article
Peer reviewed
One of the major applications of future generation parallel and distributed systems is in big-data analytics. Data repositories for such applications currently exceed exabytes and are rapidly ...increasing in size. Beyond their sheer magnitude, these datasets and associated applications’ considerations pose significant challenges for method and software development. Datasets are often distributed and their size and privacy considerations warrant distributed techniques. Data often resides on platforms with widely varying computational and network capabilities. Considerations of fault-tolerance, security, and access control are critical in many applications (Dean and Ghemawat, 2004; Apache hadoop). Analysis tasks often have hard deadlines, and data quality is a major concern in yet other applications. For most emerging applications, data-driven models and methods, capable of operating at scale, are as-yet unknown. Even when known methods can be scaled, validation of results is a major issue. Characteristics of hardware platforms and the software stack fundamentally impact data analytics. In this article, we provide an overview of the state-of-the-art and focus on emerging trends to highlight the hardware, software, and application landscape of big-data analytics.
•An overview of the state-of-the-art in big-data analytics.•Trends in scale and application landscape of big-data analytics.•Current and future trends in hardware that can help us in addressing the massive datasets.•Discussion of software techniques currently employed and future trends to address the applications.
Blockchain technology enables a decentralized and distributed environment with no need for a central authority. Transactions are simultaneously secure and trustworthy due to the use of cryptographic ...principles. In recent years, blockchain technology has become very trendy and penetrated different domains, mostly due to the popularity of cryptocurrencies. One field where blockchain technology has tremendous potential is healthcare, due to the need for a more patient-centric approach to healthcare systems and to connect disparate systems and increase the accuracy of electronic healthcare records (EHRs). In this systematic review, an analysis of state-of-the-art blockchain research in the field of healthcare is conducted. The aim is to reveal the potential applications of the technology and to highlight the challenges and possible directions of blockchain research in healthcare. First, background information is discussed, followed by a description of the exact methodology used in this paper. Next, an analysis of the results is given, which includes a bibliometric overview, an analysis of gathered data and its properties, and the results of a literature quality assessment. Lastly, there is a discussion of the results from the analysis. The findings indicate that blockchain technology research in healthcare is increasing and it is mostly used for data sharing, managing health records and access control. Other scenarios are very rare. Most research is aimed at presenting novel structural designs in the form of frameworks, architectures or models. Findings also show that technical details about the used blockchain elements are not given in most of the analyzed publications and that most research does not present any prototype implementation or implementation details. Often even with a prototype implementation, no details about blockchain elements are given.
We design two closely related state feedback adaptive control laws for stabilization of a class of2×2 linear hyperbolic system of partial differential equations (PDEs) with constant but uncertain ...in-domain and boundary parameters. One control law uses an identifier, while the other is based on swapping design. We establish boundedness of all signals in the closed loop system, pointwise in space and time, and convergence of the system states to zero pointwise in space. The theory is demonstrated in simulations.
In domains such as health care and finance, shortage of labeled data and computational resources is a critical issue while developing machine learning algorithms. To address the issue of labeled data ...scarcity in training and deployment of neural network-based systems, we propose a new technique to train deep neural networks over several data sources. Our method allows for deep neural networks to be trained using data from multiple entities in a distributed fashion. We evaluate our algorithm on existing datasets and show that it obtains performance which is similar to a regular neural network trained on a single machine. We further extend it to incorporate semi-supervised learning when training with few labeled samples, and analyze any security concerns that may arise. Our algorithm paves the way for distributed training of deep neural networks in data sensitive applications when raw data may not be shared directly.