NNV: The Neural Network Verification Tool for Deep Neural Networks and Learning-Enabled Cyber-Physical Systems
Ontology highlight
ABSTRACT:
SUBMITTER: Lahiri S
PROVIDER: S-EPMC7363192 | biostudies-literature | 2020 Jun
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:In the context of Industry 4.0, the demand for the mass production of highly customized products will lead to complex products and an increasing demand for production system flexibility. Simply implementing lean production-based human-centered production or high automation to improve system flexibility is insufficient. Currently, lean automation (Jidoka) that utilizes cyber-physical systems (CPS) is considered a cost-efficient and effective approach for improving system flexibility under shrinking global economic conditions. Therefore, a smart lean automation engine enabled by CPS technologies (SLAE-CPS), which is based on an analysis of Jidoka functions and the smart capacity of CPS technologies, is proposed in this study to provide an integrated and standardized approach to design and implement a CPS-based smart Jidoka system. A set of comprehensive architecture and standardized key technologies should be presented to achieve the above-mentioned goal. Therefore, a distributed architecture that joins service-oriented architecture, agent, function block (FB), cloud, and Internet of things is proposed to support the flexible configuration, deployment, and performance of SLAE-CPS. Then, several standardized key techniques are proposed under this architecture. The first one is for converting heterogeneous physical data into uniform services for subsequent abnormality analysis and detection. The second one is a set of Jidoka scene rules, which is abstracted based on the analysis of the operator, machine, material, quality, and other factors in different time dimensions. These Jidoka rules can support executive FBs in performing different Jidoka functions. Finally, supported by the integrated and standardized approach of our proposed engine, a case study is conducted to verify the current research results. The proposed SLAE-CPS can serve as an important reference value for combining the benefits of innovative technology and proper methodology.
| S-EPMC5539867 | biostudies-other
Project description:In the near future, value streams associated with Industry 4.0 will be formed by interconnected cyber-physical elements forming complex networks that generate huge amounts of data in real time. The success or failure of industry leaders interested in the continuous improvement of lean management systems in this context is determined by their ability to recognize behavioral patterns in these big data structured within non-Euclidean domains, such as these dynamic sociotechnical complex networks. We assume that artificial intelligence in general and deep learning in particular may be able to help find useful patterns of behavior in 4.0 industrial environments in the lean management of cyber-physical systems. However, although these technologies have meant a paradigm shift in the resolution of complex problems in the past, the traditional methods of deep learning, focused on image or video analysis, both with regular structures, are not able to help in this specific field. This is why this work focuses on proposing geometric deep lean learning, a mathematical methodology that describes deep-lean-learning operations such as convolution and pooling on cyber-physical Industry 4.0 graphs. Geometric deep lean learning is expected to positively support sustainable organizational growth because customers and suppliers ought to be able to reach new levels of transparency and traceability on the quality and efficiency of processes that generate new business for both, hence generating new products, services, and cooperation opportunities in a cyber-physical environment.
| S-EPMC7038400 | biostudies-literature
Project description:Cyber-physical systems embed software into the physical world. They appear in a wide range of applications such as smart grids, robotics, and intelligent manufacturing. Cyber-physical systems have proved resistant to modeling due to their intrinsic complexity arising from the combination of physical and cyber components and the interaction between them. This study proposes a general framework for discovering cyber-physical systems directly from data. The framework involves the identification of physical systems as well as the inference of transition logics. It has been applied successfully to a number of real-world examples. The novel framework seeks to understand the underlying mechanism of cyber-physical systems as well as make predictions concerning their state trajectories based on the discovered models. Such information has been proven essential for the assessment of the performance of cyber-physical systems; it can potentially help debug in the implementation procedure and guide the redesign to achieve the required performance.
| S-EPMC6814766 | biostudies-literature
Project description:Deep neural networks (DNNs) represent promising approaches to molecular machine learning (ML). However, their applicability remains limited to single-component materials and a general DNN model capable of handling various multicomponent molecular systems with composition data is still elusive, while current ML approaches for multicomponent molecular systems are still molecular descriptor-based. Here, a general DNN architecture extending existing molecular DNN models to multicomponent systems called MEIA is proposed. Case studies showed that the MEIA architecture could extend two exiting molecular DNN models to multicomponent systems with the same procedure, and that the obtained models that could learn both the molecular structure and composition information with equal or better accuracies compared to a well-used molecular descriptor-based model in the best model for each case study. Furthermore, the case studies also showed that, for ML tasks where the molecular structure information plays a minor role, the performance improvements by DNN models were small; while for ML tasks where the molecular structure information plays a major role, the performance improvements by DNN models were large, and DNN models showed notable predictive accuracies for an extremely sparse dataset, which cannot be modeled without the molecular structure information. The enhanced predictive ability of DNN models for sparse datasets of multicomponent systems will extend the applicability of ML in the multicomponent material design. Furthermore, the general capability of MEIA to extend DNN models to multicomponent systems will provide new opportunities to utilize the progress of actively developed single-component DNNs for the modeling of multicomponent systems.
| S-EPMC7450624 | biostudies-literature
Project description:Typically, Cyber-Physical Systems (CPS) involve various interconnected systems, which can monitor and manipulate real objects and processes. They are closely related to Internet of Things (IoT) systems, except that CPS focuses on the interaction between physical, networking and computation processes. Their integration with IoT led to a new CPS aspect, the Internet of Cyber-Physical Things (IoCPT). The fast and significant evolution of CPS affects various aspects in people's way of life and enables a wider range of services and applications including e-Health, smart homes, e-Commerce, etc. However, interconnecting the cyber and physical worlds gives rise to new dangerous security challenges. Consequently, CPS security has attracted the attention of both researchers and industries. This paper surveys the main aspects of CPS and the corresponding applications, technologies, and standards. Moreover, CPS security vulnerabilities, threats and attacks are reviewed, while the key issues and challenges are identified. Additionally, the existing security measures are presented and analyzed while identifying their main limitations. Finally, several suggestions and recommendations are proposed benefiting from the lessons learned throughout this comprehensive review.
| S-EPMC7340599 | biostudies-literature
Project description:While emerging technology for self-driving automation in vehicles progresses rapidly, the transition to an era of roads full of fully connected and automated vehicles (CAVs) may take longer than expected. Until then, it is inevitable that CAVs should coexist and interact with drivers of non-autonomous vehicles (NAVs) in urban roads. During this period of transition, it is critical to provide road safety with the mixed vehicular traffic and uncertainty caused by human drivers. To investigate the issues caused by the coexistence and interaction with humans, we propose to build a component-based and interactive intelligent transportation cyber-physical systems (ITCPS) framework. Our design of the interactive ITCPS framework aims to provide a standardized structure for users by defining core components. The framework is specified by behavior models and interfaces for the desired ITCPS components and is implemented as a form of human and hardware-in-the-loop system. We developed an intersection crossing assistance service and an automatic emergency braking service as an example of practical applications using the framework. To evaluate the framework, we tested its performance to show how effectively it operates while supporting real-time processing. The results indicate that it satisfies the timing requirements of vehicle-to-vehicle (V2V) communication and the limited processing time required for performing the functions of behavior models, even though the traffic volume reaches the road capacity. A case study using statistical analysis is conducted to assess the practical value of the developed experimental environment. The results of the case study validate the reliability among the specified variables for the experiments involving human drivers. It has shown that V2V communication support has positive effects on road safety, including intersection safety, braking events, and perception-reaction time (PRT) of the drivers. Furthermore, V2V communication support and PRT are identified as the important indicators affecting road safety at an un-signalized intersection. The proposed interactive framework is expected to contribute in constructing a comprehensive environment for the urban ITCPS and providing experimental support for the analysis of human behavior in the coexistence environment.
| S-EPMC6982715 | biostudies-literature
Project description:Cooperative Cyber-Physical Systems (Co-CPSs) can be enabled using wireless communication technologies, which in principle should address reliability and safety challenges. Safety for Co-CPS enabled by wireless communication technologies is a crucial aspect and requires new dedicated design approaches. In this paper, we provide an overview of five Co-CPS use cases, as introduced in our SafeCOP EU project, and analyze their safety design requirements. Next, we provide a comprehensive analysis of the main existing wireless communication technologies giving details about the protocols developed within particular standardization bodies. We also investigate to what extent they address the non-functional requirements in terms of safety, security and real time, in the different application domains of each use case. Finally, we discuss general recommendations about the use of different wireless communication technologies showing their potentials in the selected real-world use cases. The discussion is provided under consideration in the 5G standardization process within 3GPP, whose current efforts are inline to current gaps in wireless communications protocols for Co-CPSs including many future use cases.
| S-EPMC6264001 | biostudies-other
Project description:Machine learning algorithms have been available since the 1990s, but it is much more recently that they have come into use also in the physical sciences. While these algorithms have already proven to be useful in uncovering new properties of materials and in simplifying experimental protocols, their usage in liquid crystals research is still limited. This is surprising because optical imaging techniques are often applied in this line of research, and it is precisely with images that machine learning algorithms have achieved major breakthroughs in recent years. Here we use convolutional neural networks to probe several properties of liquid crystals directly from their optical images and without using manual feature engineering. By optimizing simple architectures, we find that convolutional neural networks can predict physical properties of liquid crystals with exceptional accuracy. We show that these deep neural networks identify liquid crystal phases and predict the order parameter of simulated nematic liquid crystals almost perfectly. We also show that convolutional neural networks identify the pitch length of simulated samples of cholesteric liquid crystals and the sample temperature of an experimental liquid crystal with very high precision.
| S-EPMC7203147 | biostudies-literature
Project description:Obtaining accurate properties of many-body interacting quantum matter is a long-standing challenge in theoretical physics and chemistry, rooting into the complexity of the many-body wave-function. Classical representations of many-body states constitute a key tool for both analytical and numerical approaches to interacting quantum problems. Here, we introduce a technique to construct classical representations of many-body quantum systems based on artificial neural networks. Our constructions are based on the deep Boltzmann machine architecture, in which two layers of hidden neurons mediate quantum correlations. The approach reproduces the exact imaginary-time evolution for many-body lattice Hamiltonians, is completely deterministic, and yields networks with a polynomially-scaling number of neurons. We provide examples where physical properties of spin Hamiltonians can be efficiently obtained. Also, we show how systematic improvements upon existing restricted Boltzmann machines ansatze can be obtained. Our method is an alternative to the standard path integral and opens new routes in representing quantum many-body states.
| S-EPMC6294148 | biostudies-literature
Project description:Sensors in Cyber-Physical Systems (CPS) are typically used to collect various aspects of the region of interest and transmit the data towards upstream nodes for further processing. However, data collection in CPS is often unreliable due to severe resource constraints (e.g., bandwidth and energy), environmental impacts (e.g., equipment faults and noises), and security concerns. Besides, detecting an event through the aggregation in CPS can be intricate and untrustworthy if the sensor's data is not validated during data acquisition, before transmission, and before aggregation. This paper introduces In-network Generalized Trustworthy Data Collection (IGTDC) framework for event detection in CPS. This framework facilitates reliable data for aggregation at the edge of CPS. The main idea of IGTDC is to enable a sensor's module to examine locally whether the event's acquired data is trustworthy before transmitting towards the upstream nodes. It further validates whether the received data can be trusted or not before data aggregation at the sink node. Additionally, IGTDC helps to identify faulty sensors. For reliable event detection, we use collaborative IoT tactics, gate-level modeling with Verilog User Defined Primitive (UDP), and Programmable Logic Device (PLD) to ensure that the event's acquired data is reliable before transmitting towards the upstream nodes. We employ Gray code in gate-level modeling. It helps to ensure that the received data is reliable. Gray code also helps to distinguish a faulty sensor. Through simulation and extensive performance analysis, we demonstrate that the collected data in the IGTDC framework is reliable and can be used in the majority of CPS applications.
| S-EPMC8114821 | biostudies-literature