Teleoperation is a fundamental feature for the development of autonomous vehicles. Not only offers it the possibility of an efficient fallback solution for emerging situations that the autonomous vehicle cannot resolv...
详细信息
ISBN:
(纸本)9798350399462
Teleoperation is a fundamental feature for the development of autonomous vehicles. Not only offers it the possibility of an efficient fallback solution for emerging situations that the autonomous vehicle cannot resolve itself but also it potentially offers new business models for transportation service providers. Fundamental to these considerations, however, is the performance of the communication network and the resulting parameters such as latency and bandwidth. These quantities have a critical impact on the ability of a human to perceive the environment, assess the situation and remotely control the vehicle safely and reliably. In this paper, a framework for simulating and evaluating the impact of network parameters on the performance of a remote human driver is presented. Using an example scenario of a shuttle bus route, the effects of latency for the teleoperation use case is evaluated. The local network and the impact on the human driver are modeled in SUMO (Simulation of Urban MObility) and OMNEST. The latency network simulation is validated against real measurement data. The car-following model Extended Intelligent Driver Model (EIDM) and its human driver model are extended so that latency effects can be quantified. The proposed approach has multiple advantages, on the one hand, it allows to evaluate the network suitability for the teleoperation use case. On the other hand, it limit the number of variables to be evaluated in e.g. user studies.
We propose a resilient voltage regulation strategy for microgrids based on a directed communication network. Initially, we employ feedback linearization techniques to address the inherent nonlinear complexity in distr...
详细信息
As the demand for high-performance and dynamically adaptable network services continues to grow, the convergence of quantum computing and network orchestration represents a promising development. This paper explores t...
详细信息
ISBN:
(纸本)9798350359107;9798350359091
As the demand for high-performance and dynamically adaptable network services continues to grow, the convergence of quantum computing and network orchestration represents a promising development. This paper explores the transformative potential of quantum computing in the orchestration of network services and introduces new paradigms that transcend classical limitations. Using a case study, we propose a workflow for an exemplary link failure process for generative AI (GenAI) and quantum computing-based service orchestration in an edge cloud (EC) network. Later, we address the challenges of integrating services based on quantum computing (using a quantum search algorithm to select the best connections) with GenAI-based topology control and management in an orchestration framework. Our work also introduces multi-solution quantum search for orchestrating services across different edge clouds, emphasizing efficiency and accuracy. Algorithmic considerations, including the impact of the oracle and hybrid approaches, are also thoroughly explored. We conclude the paper with potentials and comparisons of possible GenAI techniques and quantum search algorithms in terms of their advantages and disadvantages.
This study addresses the challenging problem of pitch angle tracking control for noncanonical nonlinear aircraft systems. Traditional control schemes for pitch angle tracking are primarily designed based on the Lyapun...
详细信息
This paper introduces a new Convolutional Neural network (ConvNet) architecture inspired by a class of partial differential equations (PDEs) called quasi-linear hyperbolic systems. With comparable performance on image...
详细信息
ISBN:
(数字)9789819991198
ISBN:
(纸本)9789819991181;9789819991198
This paper introduces a new Convolutional Neural network (ConvNet) architecture inspired by a class of partial differential equations (PDEs) called quasi-linear hyperbolic systems. With comparable performance on image classification task, it allows for the modification of the weights via a continuous group of symmetry. This is a significant shift from traditional models where the architecture and weights are essentially fixed. We wish to promote the (internal) symmetry as a new desirable property for a neural network, and to draw attention to the PDE perspective in analyzing and interpreting ConvNets in the broader Deep Learning community.
A neuro network-based predictive controller is being developed for real-time temperature control on the educational test bench «Thermal Object» using the MATLAB software package for applied programs in the d...
详细信息
SDN enables centralized control of network services by separating control and data planes, providing greater flexibility and scalability in network traffic management. However, this architecture also creates new secur...
详细信息
ISBN:
(纸本)9783031377648;9783031377655
SDN enables centralized control of network services by separating control and data planes, providing greater flexibility and scalability in network traffic management. However, this architecture also creates new security challenges, such as a single point of failure that can compromise the entire network. Furthermore, the dynamic environment of SDN makes it difficult to adapt an effective and efficient anomaly-based IDS. A high-performance Machine Learning (ML)-based IDS can meet these challenges, but careful consideration is required in algorithm selection, data preprocessing, feature engineering, and model architecture. In this article, we conducted a comparative study of several ML models on the InSDN dataset. The results showed that the DT model performs well with only 7 out of the 83 features in the InSDN dataset, reducing the training time from 10 to 0.5 s.
This article explores the high interaction attack defense mechanism based on runtime application self-protection (RASP) technology in power system network security reinforcement. With the rapid development of informat...
详细信息
Nowadays, in various domains including distributed control, telecommunication, robotics and economics to address the problems, Multi-Agent systems (MAS) is used. To solve with preprogrammed agent behavior complexity o...
详细信息
With the increasing availability of simultaneous panchromatic and hyperspectral images, object detection methods based on them have demonstrated significant application advantages. However, they still face several cha...
详细信息
暂无评论