Software Impacts最新文献

WayWise is an innovative C++ and Qt-based rapid prototyping library designed to advance the development and analysis of connected, autonomous vehicles (CAVs) and Unmanned Arial Systems (UASs). It was deployed on model-sized cars and trucks as well as full-sized mobile machinery, tractors and UASs. It is actively being used in several European research projects. Developed by the RISE Dependable Transport Systems unit, the library facilitates exploration into safety and cybersecurity aspects inherent to various emerging vehicular applications within road traffic and offroad applications. This non-production library emphasizes rapid prototyping, leveraging commercial off-the-shelf hardware and the different protocols for vehicle-control communication, mainly focusing on MAVLINK. The utility of WayWise in rapidly evaluating complex vehicular behaviors is demonstrated through various research projects, thus contributing to the field of autonomous vehicular technology.

Advancements in artificial intelligence in the cybersecurity domain introduce significant security challenges. A critical concern is the exposure of deep learning techniques to adversarial attacks. Adversary users intentionally attempt to mislead the techniques by infiltrating adversarial samples to mislead the prediction of security devices. The study presents extensive experimentation of defense methods using Python-based open-source code with two benchmark datasets, and the outcomes are demonstrated using evaluation metrics. This code library can be easily utilized and reproduced for cybersecurity research on countering adversarial attacks. Exploring strategies for protecting against adversarial attacks is significant in enhancing the resilience of deep learning techniques.

Vendor lock-in has become a major concern in cloud computing. The term vendor lock-in describes situations where the subscriber cannot move data or services to another cloud vendor. This is due to heavy data volumes, high network bandwidth costs, dependencies, or unacceptable downtime. The proposed vendor lock-in dissolution practice migrates the database effectively in noticeably less time, regardless of database size and with a nominal network bandwidth requirement. Through this new practice, databases can be migrated to very remote regions, even across continents. A real-time implementation of the proposed method presented in this paper.

We provide numerical software based on the MATLAB programming language to study the Bessel-like beams generated by special instruments such as DMD. The calculations are based on the scalar Fresnel–Kirchhoff integration within the scope of Fourier Optics. This analysis is particularly important because the addition of higher-order Bessel terms may produce additional unexpected experimental results in some applications. We emphasize the seldom-mentioned imaging characteristic on the lens, where the central point is shifted, and provide numerical software to understand the expression of the Bessel-like function obtained from important theoretical derivation. It also benefits to verify and explain the experimental results.

Supervised training of machine learning models heavily relies on accurate annotations. However, data annotation, such as in the case of time-series signals, poses a labor-intensive challenge. Here, we present a new annotation software, Annotation of Time-series Events (ANNOTE), to handle longitudinal, time-series signals as in highly complex physiological events. ANNOTE offers flexibility and adaptability to streamline the annotation process through an intuitive user interface, effectively meeting diverse annotation needs. Users can annotate regions of interest with precision down to a single data point. ANNOTE presents a useful tool to support researchers in handling time-series biomedical data for downstream machine-learning analyses.

SSFCM-FWCW (Feature-Weight and Cluster-Weight based Semi-Supervised Fuzzy C-Means) is a soft clustering method. It incorporates supplementary label information to enhance the clustering quality. An adaptive local feature weighting technique is utilized to weight features based on their significance within specific clusters. Additionally, an adaptive weighting technique is applied to diminish the sensitivity to the initial center selection, effectively distinguishing between the effects of various clusters. The conjunction of label information and adaptive weighting results in an optimal fuzzy c-means clustering with an insight into the importance of individual features and clusters. An open-source Matlab implementation of SSFCM-FWCW is available.

In this paper we present a Basic Encoding Rules library implemented in three programming languages (C, C++ and C#) that offers encoding and decoding of data types in order to be used in serialization and deserialization processes in communication protocols. The implementation is consistent with the currently active standards and it offers a great degree of scalability. BERLib is also highly documented and significant examples are provided for all the programming languages used. Our work qualifies as an ideal solution for providing data encoding and decoding in communication protocol design especially for Wireless Sensor Networks and Internet of Things.

We present a new library designed to simplify the analysis of Euler characteristics. This program addresses the difficulties involved in generating 3D test objects and the complexities of extracting Octo-Voxel patterns. The library uses a novel method to rapidly generate data and extract descriptors by using effective multiprocessing. Furthermore, we have developed a method for extracting discrete CHUNKS from an image, allowing for separate multiprocessing assessment. This method accelerates the process of combination extraction and offers researchers a quick and effective way to explore Euler characteristics in a variety of applications.

This paper aims to introduce the ‘deforce’ framework, an open-source Python library constituted on top of Numpy, Scikit-Learn, PyTorch, and Mealpy. This framework provides hybrid models that combine derivative-free techniques with Cascade Forward Neural Networks (CFNNs). By inheriting from scikit-learn’s estimator, deforce’s models ensure easy integration into existing machine learning pipelines. It also has many advantages, including a simple installation process, a user-friendly interface, and adaptability to various user requirements. For researchers and practitioners looking to improve CFNN performance with minimal implementation effort, deforce offers a useful and approachable option.

This article introduces SRcdFuzzy, a MATLAB-based software designed to simulate two controllers: the Adaptive Fuzzy Iterative Learning Controller (AF-ILC) and the Adaptive Fuzzy Repetitive Generalized Predictive Controller (AFR-GPC). These controllers, as proposed in a companion paper [7], aim to minimize cyclical disturbances with small frequency range variations commonly encountered in industrial process control loops. They utilize fuzzy logic to estimate the disturbance cycle, employing rules based on past values of the integral of the absolute error between the setpoint and output. The estimated disturbance cycle period is then communicated to the regulatory controllers, guiding specific control actions to mitigate oscillations in the process output. Furthermore, each controller includes a dedicated interface, offering testing options for various scenarios, including disturbances with different frequencies.