| 1. |
- Chen, Jinchao, et al.
(författare)
-
Global-and-Local Attention-Based Reinforcement Learning for Cooperative Behaviour Control of Multiple UAVs
- 2024
-
Ingår i: IEEE Transactions on Vehicular Technology. - Piscataway : Institute of Electrical and Electronics Engineers (IEEE). - 0018-9545 .- 1939-9359. ; 73:3, s. 4194-4206
-
Tidskriftsartikel (refereegranskat)abstract
- Due to the strong adaptability and high flexibility, unmanned aerial vehicles (UAVs) have been extensively studied and widely applied in both civil and military applications. Although UAVs can achieve significant cost reduction and performance enhancement in large-scale systems by taking full advantage of their cooperation and coordination, they result in a serious cooperative behaviour control problem. Especially in dynamic environments, the cooperative behaviour control problem which has to quickly produce a safe and effective behaviour decision for each UAV to achieve group missions, is NP-hard and difficult to settle. In this work, we design a global-and-local attention-based reinforcement learning algorithm for the cooperative behaviour control problem of UAVs. First, with the motion and coordination models, we analyze the collision avoidance, motion state update, and task execution constraints of multiple UAVs, and abstract the cooperative behaviour control problem as a multi-constraint decision-making one. Then, inspired from the human-learning process where more attention is devoted to the important parts of data, we design a multi-agent reinforcement learning algorithm with a global-and-local attention mechanism to cooperatively control the behaviours of UAVs and achieve the coordination. Simulation experiments in a multi-agent particle environment provided by OpenAI are conducted to verify the effectiveness and efficiency of the proposed approach. Compared with baselines, our approach shows significant advantages in mean reward, training time, and coordination effect. © 2023 IEEE.
|
|
| 2. |
- Lakhan, Abdullah, et al.
(författare)
-
Blockchain-Enabled Cybersecurity Efficient IIOHT Cyber-Physical System for Medical Applications
- 2022
-
Ingår i: IEEE Transactions on Network Science and Engineering. - Piscataway, NJ : IEEE. - 2327-4697 .- 2334-329X. ; 10:5, s. 2466-2479
-
Tidskriftsartikel (refereegranskat)abstract
- Cybersecurity issues such as malware, denial of service attacks, and unauthorized access to data for different applications are growing daily. The Industrial Internet of Healthcare Things (IIoHT) has recently been a new healthcare mechanism where many healthcare applications can run on hospital servers for remote medical services. For instance, cloud medical applications offer different services remotely from home. However, the existing IIoHT mechanisms can not handle critical cybersecurity issues and incur many medical care application processing and data security costs. The processing costs associated with security and deadline are the main findings of this proposed work. This work devises a cost-efficient blockchain task scheduling (CBTS) cyber-physical system (CPS) with different heuristics. All tasks are sorted, scheduled, and stored in a secure form in the IIoHT network. The performance evaluation proves that the CBTS framework outperforms the simulation results for the IIoHT application and reduces the cost by 50% of security execution and 33% of cybersecurity data validation blockchain costs compared to existing scheduling and blockchain schemes. © Copyright 2022 IEEE
|
|
| 3. |
- Lakhan, Abdullah, et al.
(författare)
-
DRLBTS : deep reinforcement learning-aware blockchain-based healthcare system
- 2023
-
Ingår i: Scientific Reports. - London : Nature Publishing Group. - 2045-2322. ; 13:1, s. 1-15
-
Tidskriftsartikel (refereegranskat)abstract
- Industrial Internet of Things (IIoT) is the new paradigm to perform different healthcare applications with different services in daily life. Healthcare applications based on IIoT paradigm are widely used to track patients health status using remote healthcare technologies. Complex biomedical sensors exploit wireless technologies, and remote services in terms of industrial workflow applications to perform different healthcare tasks, such as like heartbeat, blood pressure and others. However, existing industrial healthcare technoloiges still has to deal with many problems, such as security, task scheduling, and the cost of processing tasks in IIoT based healthcare paradigms. This paper proposes a new solution to the above-mentioned issues and presents the deep reinforcement learning-aware blockchain-based task scheduling (DRLBTS) algorithm framework with different goals. DRLBTS provides security and makespan efficient scheduling for the healthcare applications. Then, it shares secure and valid data between connected network nodes after the initial assignment and data validation. Statistical results show that DRLBTS is adaptive and meets the security, privacy, and makespan requirements of healthcare applications in the distributed network. © 2023, The Author(s).
|
|
| 4. |
- Liang, Guojun, et al.
(författare)
-
Semantics-aware Dynamic Graph Convolutional Network for Traffic Flow Forecasting
- 2023
-
Ingår i: IEEE Transactions on Vehicular Technology. - Piscataway, NJ : IEEE. - 0018-9545 .- 1939-9359. ; 72:6, s. 7796-7809
-
Tidskriftsartikel (refereegranskat)abstract
- Traffic flow forecasting is a challenging task due to its spatio-temporal nature and the stochastic features underlying complex traffic situations. Currently, Graph Convolutional Network (GCN) methods are among the most successful and promising approaches. However, most GCNs methods rely on a static graph structure, which is generally unable to extract the dynamic spatio-temporal relationships of traffic data and to interpret trip patterns or motivation behind traffic flows. In this paper, we propose a novel Semantics-aware Dynamic Graph Convolutional Network (SDGCN) for traffic flow forecasting. A sparse, state-sharing, hidden Markov model is applied to capture the patterns of traffic flows from sparse trajectory data; this way, latent states, as well as transition matrices that govern the observed trajectory, can be learned. Consequently, we can build dynamic Laplacian matrices adaptively by jointly considering the trip pattern and motivation of traffic flows. Moreover, high-order Laplacian matrices can be obtained by a newly designed forward algorithm of low time complexity. GCN is then employed to exploit spatial features, and Gated Recurrent Unit (GRU) is applied to exploit temporal features. We conduct extensive experiments on three real-world traffic datasets. Experimental results demonstrate that the prediction accuracy of SDGCN outperforms existing traffic flow forecasting methods. In addition, it provides better explanations of the generative Laplace matrices, making it suitable for traffic flow forecasting in large cities and providing insight into the causes of various phenomena such as traffic congestion. The code is publicly available at https://github.com/gorgen2020/SDGCN. © 2023 IEEE.
|
|
| 5. |
- Mumtaz, Nadia, et al.
(författare)
-
An overview of violence detection techniques : current challenges and future directions
- 2023
-
Ingår i: Artificial Intelligence Review. - Dordrecht : Springer Nature. - 0269-2821 .- 1573-7462. ; 56, s. 4641-4666
-
Tidskriftsartikel (refereegranskat)abstract
- The Big Video Data generated in today’s smart cities has raised concerns from its purposeful usage perspective, where surveillance cameras, among many others are the most prominent resources to contribute to the huge volumes of data, making its automated analysis adifcult task in terms of computation and preciseness. Violence detection (VD), broadly plunging under action and activity recognition domain, is used to analyze Big Video data for anomalous actions incurred due to humans. The VD literature is traditionally basedon manually engineered features, though advancements to deep learning based standalone models are developed for real-time VD analysis. This paper focuses on overview of deepsequence learning approaches along with localization strategies of the detected violence.This overview also dives into the initial image processing and machine learning-based VD literature and their possible advantages such as efciency against the current complex models. Furthermore,the datasets are discussed, to provide an analysis of the current models, explaining their pros and cons with future directions in VD domain derived from anin-depth analysis of the previous methods. © The Author(s), under exclusive licence to Springer Nature B.V. 2022.
|
|
| 6. |
- Neeraj, Kumar, 1991-, et al.
(författare)
-
Magnetization switching in the inertial regime
- 2022
-
Ingår i: Physical Review B. - 2469-9950 .- 2469-9969. ; 105:5
-
Tidskriftsartikel (refereegranskat)abstract
- We have numerically solved the Landau-Lifshitz-Gilbert (LLG) equation in its standard and inertial forms to study the magnetization switching dynamics in a 3d thin film ferromagnet. The dynamics is triggered by ultrashort magnetic field pulses of varying width and amplitude in the picosecond and Tesla range. We have compared the solutions of the two equations in terms of switching characteristic, speed, and energy analysis. Both equations return qualitatively similar switching dynamics, characterized by regions of slower precessional behavior and faster ballistic motion. In the case of inertial dynamics, ballistic switching is found in a 25% wider region in the parameter space given by the magnetic field amplitude and width. The energy analysis of the dynamics is qualitatively different for the standard and inertial LLG equations. In the latter case, an extra energy channel, interpreted as the kinetic energy of the system, is available. Such an extra channel is responsible for a resonant energy absorption at THz frequencies, consistent with the occurrence of spin nutation.
|
|
| 7. |
- Neeraj, Kumar, 1991-, et al.
(författare)
-
Terahertz charge and spin transport in metallic ferromagnets : The role of crystalline and magnetic order
- 2022
-
Ingår i: Applied Physics Letters. - : AIP Publishing. - 0003-6951 .- 1077-3118. ; 120:10
-
Tidskriftsartikel (refereegranskat)abstract
- We study the charge and spin dependent scattering in a set of CoFeB thin films whose crystalline order is systematically enhanced and controlled by annealing at increasingly higher temperatures. Terahertz conductivity measurements reveal that charge transport closely follows the development of the crystalline phase, with the increasing structural order leading to higher conductivity. The terahertz-induced ultrafast demagnetization, driven by spin-flip scattering mediated by the spin–orbit interaction, is measurable in the pristine amorphous sample and much reduced in the sample with the highest crystalline order. Surprisingly, the largest demagnetization is observed at intermediate annealing temperatures, where the enhancement in spin-flip probability is not associated with an increased charge scattering. We are able to correlate the demagnetization amplitude with the magnitude of the in-plane magnetic anisotropy, which we characterize independently, suggesting a magnetoresistance-like description of the phenomenon.
|
|
| 8. |
- Neeraj, Kumar, 1991-
(författare)
-
Terahertz spin dynamics in metallic thin film ferromagnets
- 2021
-
Doktorsavhandling (övrigt vetenskapligt/konstnärligt)abstract
- The past two decades have witnessed an increasing interest in understanding and controlling materials at the pico- and femtosecond time scales, the so-called ultrafast regime. Among the broad field of condensed matter physics, magnetism and magnetic materials have attracted much interest both from a fundamental and an applied perspective. The field of ultrafast magnetism is at the frontier of current physics research, with fundamental questions that are still unanswered but which have the potential of impacting the data storage technology upon which our digitalized world relies on. Ultrafast lasers in the visible range (i.e., with energies in the eV range) have been widely used to study ultrafast magnetization dynamics, but due to the relatively large photon energy, they create highly non equilibrium states which tend to mask the fundamental coupling processes leading to ultrafast demagnetization. However, the relatively recent appearance of intense coherent terahertz (THz) radiation (with photon energies in the meV range) offers a new way to understand and manipulate the magnetic order, and is receiving much attention in the research community. As a major part of this thesis, a table-top experimental setup for generating intense THz radiation has been developed for the purpose of carrying out pump-probe studies of thin ferromagnetic metallic films. The setup is capable of delivering state-of-the-art THz electric fields as large as 1 MV/cm, corresponding to 0.3 T magnetic fields which can directly couple to the magnetization to trigger ultrafast dynamics. The ultrafast magnetization dynamics is probed with the time resolved magneto-optical Kerr effect with a resolution of approximately 40 fs. Three main scientific results have been obtained with this thesis work. First, the experimental evidence, in the form of a spin nutation in the THz range, of inertial magnetization dynamics in thin film ferromagnets, which we could describe with a modified version of the textbook Landau Lifshitz-Gilbert (LLG) equation to include a realistic inertial tensor. Second, with this modified LLG equation, we performed simulations to study the role of inertia in the switching of the magnetization with picosecond magnetic field pulses. We found that inertia leads to a so-called ballistic switching which is more robust to the details of the magnetic field pulse. Third, we studied the influence of crystalline order on the charge and spin transport at terahertz rates. We found that while the charge scattering follows the degree of crystalline order in the film, the spin scattering is enhanced at intermediate crystalline phases which have not fully ordered, but where the magnetic anisotropy is largest.
|
|
| 9. |
- Zhou Hagström, Nanna, 1993-, et al.
(författare)
-
Megahertz-rate ultrafast X-ray scattering and holographic imaging at the European XFEL
- 2022
-
Ingår i: Journal of Synchrotron Radiation. - : International Union of Crystallography (IUCr). - 0909-0495 .- 1600-5775. ; 29, s. 1454-1464
-
Tidskriftsartikel (refereegranskat)abstract
- The advent of X-ray free-electron lasers (XFELs) has revolutionized fundamental science, from atomic to condensed matter physics, from chemistry to biology, giving researchers access to X-rays with unprecedented brightness, coherence and pulse duration. All XFEL facilities built until recently provided X-ray pulses at a relatively low repetition rate, with limited data statistics. Here, results from the first megahertz-repetition-rate X-ray scattering experiments at the Spectroscopy and Coherent Scattering (SCS) instrument of the European XFEL are presented. The experimental capabilities that the SCS instrument offers, resulting from the operation at megahertz repetition rates and the availability of the novel DSSC 2D imaging detector, are illustrated. Time-resolved magnetic X-ray scattering and holographic imaging experiments in solid state samples were chosen as representative, providing an ideal test-bed for operation at megahertz rates. Our results are relevant and applicable to any other non-destructive XFEL experiments in the soft X-ray range.
|
|
