Search Results (422)

Point cloud perception is important in autonomous driving, robotics, and other security-critical 3D systems, yet learned point cloud classifiers remain vulnerable to transferable adversarial perturbations. A central difficulty in transfer-based black-box attacks is surrogate overfitting: an update that is highly effective on an accessible source model may not generalize to an unknown target architecture. We introduce SpecEOT, a source-agnostic and graph-spectral expectation-over-transformation attack. A fixed graph Fourier transform (GFT) basis is constructed from each clean point cloud. At every optimization iteration, each non-identity view independently samples a frequency band and a perturbation sign from uniform distributions; the resulting view gradients are averaged with equal weights and used to update the adversarial point cloud through projected Adam ascent. We evaluate the stochastic method over repeated seeds, extend the ablation to two source architectures, and analyze the interaction between band count and randomization strength while reporting computational cost and assessing robustness to Gaussian jitter and point dropout. SpecEOT achieves strong transferability on ModelNet40 and ShapeNet. Full article

This paper proposes a three-dimensional prescribed-time hierarchical cooperative guidance law (3-D PTHCGL) for rapid and reliable head-on encirclement in cooperative interception of high-speed targets under incomplete directed communication topology, target maneuvers, and limited terminal engagement time. The proposed method consists of a distributed estimator layer (DEL) and a local controller layer (LCL), addressing three key issues in head-on encirclement formation: information acquisition, rapid convergence, and geometric configuration. In the DEL, a distributed prescribed-time estimator (DPTE) is developed to enable followers without direct communication with the leader to estimate the leader’s states from neighborhood information within a prescribed time. In the LCL, a prescribed-time extended state observer (PTESO) and a three-dimensional prescribed-time cooperative guidance law (3-D PTCGL) are designed to estimate target-maneuver-induced disturbances and unknown states, and to guarantee prescribed-time convergence of estimation and cooperative tracking errors. Furthermore, virtual line-of-sight (LOS) angles are introduced based on a three-dimensional head-on interception kinematic model to characterize the head-on encirclement configuration, and range-to-go together with radial relative velocity are adopted instead of time-to-go to reduce sensitivity to time-estimation errors. Simulation results demonstrate the effectiveness and robustness of the proposed method in achieving prescribed-time head-on encirclement and simultaneous attack without a speed advantage. Full article

Public discussion in Australia often presents pet cats as a major cause of wildlife decline, and cat management is frequently promoted as a major conservation response. However, this focus may draw attention away from other common and preventable recorded causes of threatened species rescue. Using the NSW Wildlife Rehabilitation Data Dashboard, we present the first statewide analysis of NSW wildlife rehabilitation records to specifically examine threatened species, rather than pooling threatened and non-threatened wildlife together. This dataset represented 52,475 individuals and 158 threatened species rescued from 2013 to 2024 in NSW. Rehabilitation data represent only animals that are found, reported, and brought into care, mostly in human-frequented environments, and should not be interpreted as estimates of total wildlife mortality, predation pressure, or ecological impact beyond the rehabilitation dataset. Outcomes were often poor: 24.1% were released, 58.5% died, and 17.5% had other recorded outcomes. The largest recorded category was Unknown (21.9%), which should not be interpreted as absence of cause, but as cases where a specific cause could not be assigned at the time of reporting to the dashboard. Among identified recorded causes, the most common were Entanglement (11.6%), Weather Event (11.4%), Abandoned/Orphaned (10.1%), Unsuitable Environment (7.3%), and Motor-Vehicle Collision (5.8%). Together, Unknown and these five identified causes accounted for 68.1% of recorded threatened species rescues. By comparison, all animal attacks combined accounted for 4.4%; dog-related rescues were more than three times as frequent as cat-related rescues, with cat-related rescues comprising 0.6%. These findings show that cat attacks formed a small proportion of recorded rescues, although overall ecological impact cannot be determined from these data. Prevention efforts targeting common and preventable recorded causes may help reduce the number of threatened animals entering care. Full article

LiDAR point clouds are widely used in remote sensing perception scenarios, such as autonomous driving. However, LiDAR-based perception models remain vulnerable to adversarial perturbations, which may compromise the reliability of safety-critical 3D perception systems. Among different attack paradigms, transfer-based attacks are particularly practical because they generate adversarial examples on accessible surrogate models and apply the generated examples directly to unknown target models. Nevertheless, existing transferable attacks on point clouds often perturb regions that are discriminative for the surrogate model but insufficiently stable across different architectures, leading to limited transferability and noticeable geometric distortion. To address this problem, we propose SAGE, a Saliency And Geometry Enhanced transferable attack framework for LiDAR point cloud perception in remote sensing. Specifically, SAGE unifies point-coordinate priors with source-model gradient signals to generate a saliency map, which serves as a transferable indicator of vulnerable local structures. SAGE further leverages this map through saliency-guided perturbation allocation and explicit geometric constraints to enhance transferability while preserving point-cloud geometry. To demonstrate the effectiveness of SAGE, we evaluate SAGE on point-cloud classification benchmarks and further validate it on LiDAR-based 3D object detection using KITTI and nuScenes. Experimental results show that SAGE consistently outperforms existing transferable attack methods in attack success rate while preserving favorable geometric quality of adversarial point clouds. These findings demonstrate that SAGE offers an effective and practical framework for assessing the transfer robustness of LiDAR-based remote sensing perception systems. Full article

This paper investigates an observer-based secure control method for networked non-Lipschitz nonlinear systems subject to unknown nonlinearities, external disturbances, sensor noises, and intermittent denial-of-service (DoS) attacks. Multi-layer neural networks (MNNs) are adopted to compensate for non-smooth, non-Lipschitz terms, guaranteeing bounded approximation errors. A resilient high-gain observer fused with the MNN is developed to continuously reconstruct system states. When DoS attacks block sensor channels, the observer acts as a virtual dynamic engine to substitute for lost real-time measurements, providing uninterrupted feedback to the controller. Furthermore, to optimize communication efficiency, an observer-based static event-triggered mechanism (SETM) coupled with a hold-input strategy is integrated. Employing the Lyapunov–Krasovskii functional method, sufficient conditions are derived to prove that the closed-loop system remains uniformly ultimately bounded (UUB) under the joint effects of approximation errors, disturbances, and attacks. Simulation results on a two-link manipulator demonstrate that the proposed secure control scheme effectively counters aggressive DoS attacks while achieving a 56.8% reduction in network transmissions compared with conventional periodic sampling paradigms, striking a favorable balance between tracking accuracy and resource efficiency. Full article

This paper addresses the challenges of achieving robust coordination in discrete-time multi-robot systems subject to uncertainties and Byzantine attacks affecting both actuator and sensor channels. Such adversarial disruptions degrade system performance by corrupting control inputs and state measurements, ultimately threatening stability and consensus in networked robotic systems. To overcome these limitations, a novel discrete-time adaptive control framework is proposed that ensures reliable tracking and stability under both uncoupled and coupled robot dynamics. The approach integrates a modified graph-theoretic structure with node-dependent weighting to capture heterogeneous robot interactions, while explicitly modeling attack effects within the system dynamics. An adaptive control law is developed using a nonlinear basis function approximation to handle unknown system uncertainties, along with a dynamic weight update mechanism that compensates for adversarial disturbances in real time. For the uncoupled case, stability is established through a composite Lyapunov function incorporating logarithmic and quadratic terms, guaranteeing boundedness of all closed-loop signals and asymptotic convergence of the tracking error. This framework is further extended to systems with coupled dynamics by introducing an auxiliary estimation mechanism to reconstruct unmeasurable interactions, leading to a unified adaptive controller capable of mitigating both internal uncertainties and external attacks. Rigorous Lyapunov-based analysis demonstrates that the proposed method ensures asymptotic tracking performance despite the presence of Byzantine disturbances. Numerical simulations validate the theoretical results, showing improved resilience, accurate trajectory tracking, and enhanced robustness compared to existing approaches. Full article

The exponential growth of Indicators of Compromise (IoCs) has overwhelmed manual triage processes in Security Operations Centers (SOCs), necessitating automated solutions for large-scale log analysis. This study proposes a hybrid machine learning framework that integrates supervised and unsupervised learning to automate the classification, clustering, and contextual interpretation of Cyber Threat Intelligence (CTI). The primary contribution lies in a multi-stage feature engineering pipeline that enriches raw SIEM logs with cyclical temporal encoding and geographical metadata. In the supervised phase, a comparative evaluation of gradient boosting classifiers—XGBoost, LightGBM, and CatBoost—demonstrates that all three achieve competitive performance in categorizing known attack techniques, consistently outperforming the Random Forest baseline. The results indicate that classifier performance is dataset-dependent, and practitioners are encouraged to select the most suitable model based on their operational environment. Simultaneously, the unsupervised phase employs density-based clustering to identify emerging and previously unknown threat patterns by correlating adversarial behaviors with source attribution. By combining these two approaches, the framework ensures near-real-time feasibility and significantly enhances the scalability of automated threat extraction from distributed honeypot environments. Full article

Smart-city advanced metering infrastructure enables utility-scale remote analytics, but some forms of under-registration closely resemble lawful changes in demand and are hard to model as anomalies. We study a narrow, physically motivated event family at the single-meter level, namely multiplicative under-registration with unknown onset (a shunt-like attack), in which recorded active energy is approximately scaled by a factor $\alpha <1$ after a change-point while the daily-profile structure and spectral shape remain invariant. We formalize the problem and develop a physics-guided detector family based on weighted daily-profile regression (GLS) and its robust variant (RGLS), with quality-control filters, spectral-consistency checks, and an optional reactive-channel gate, designed to stay selective under confounders such as rooftop photovoltaics, electric-vehicle charging, and heat-pump onsets. On a device-disjoint Low Carbon London benchmark (487 households) the preferred GLS detector attains precision $0.915$, recall $0.978$, and $F_1=0.945$ at $\alpha =0.10$ while keeping the non-theft suspected rate near $1\%$; a cross-dataset check on Open Power System Data with real EV/PV/heat-pump overlays yields zero false alarms on all 72 cases, and Mendeley and WPuQ benchmarks add a second large family and a reactive-channel test. We compare against external baselines (classical change-point detection, Isolation Forest, autoencoder, LSTM, gradient boosting, and a supervised statistical pipeline) on the same protocol: generic anomaly detectors fail on this shape-preserving attack, and supervised models match the detector only in-distribution while, unlike it, failing to transfer to real lawful confounders. All metrics carry bootstrap confidence intervals, and a full reproducibility bundle accompanies the submission. Full article

Modern connected vehicles rely on the controller area network (CAN) to disseminate safety-critical in-vehicle information, including sensor-related and vehicle-state signals such as engine revolutions per minute (RPM) and gear state, among electronic control units (ECUs). Because CANs lack built-in authentication and encryption, malicious message injection and spoofing can compromise the integrity and availability of vehicular sensing and control functions. Existing deep-learning-based intrusion-detection systems (IDSs) show a clear trade-off: supervised methods perform well on known attacks but rely on costly labels, whereas unsupervised methods can identify unseen attacks but often suffer from high false-positive rates. To address these limitations, this paper proposes a semi-supervised generative adversarial network (SGAN) framework for CAN bus intrusion detection that combines image-based CAN representation with adversarial learning. Consecutive CAN messages are converted into $64\times 9$ grayscale images, and the proposed framework is trained in three phases. First, the discriminator establishes an initial decision boundary using a small labeled subset. It then refines this boundary through distribution-level likelihood objectives and generated samples. Finally, the generator is trained to produce realistic samples capable of deceiving the discriminator. The proposed method was evaluated on the Hacking and Countermeasure Research Lab (HCRL) car-hacking dataset using leave-one-class-out experiments to simulate unknown attacks and achieved an average accuracy of 99.73% and an average F1-score of 99.63% on unknown attacks. Moreover, with only 0.21 M parameters and 3.25 M floating-point operations (FLOPs), the model is well suited for resource-constrained in-vehicle platforms. These results indicate that the proposed framework can serve as a practical cybersecurity component for protecting CAN-carried data in vehicular sensing applications. Full article

Vision Transformers (ViTs) have achieved strong performance in computer vision, but their adversarial robustness remains underexplored. Existing ViT-oriented attacks mainly rely on iterative optimization, leading to high generation cost and limited transferability. Moreover, most generative attacks target a single class, making them inefficient for multi-target scenarios. To address these issues, we propose Attention-Guided Multi-Target Generative Adversarial Attack (AMGAA). AMGAA leverages ViT self-attention to guide target feature fusion and adaptively selects important source patches for perturbation generation. It jointly optimizes adversarial, attention constraint, and total variation losses to improve targeted attack success while preserving visual naturalness. Experiments on CIFAR-10 and ImageNet show that AMGAA achieves average attack success rates (ASRs) of 43.2% and 39.0% in ImageNet single-target transfer attacks and CIFAR-10 multi-target attacks, respectively. Compared with the generative attack methods evaluated in our experiments, AMGAA improves ASR by 5.7 percentage points in the multi-target setting and by 8.1 percentage points in unknown-class generalization. AMGAA also obtains a low LPIPS of 0.018, indicating good visual imperceptibility. Ablation studies confirm the effectiveness of its key components. Full article

Background: Root rot caused by Fusarium solani is a devastating disease in Angelica sinensis (danggui), leading to severe yield and quality losses. Sustainable control strategies are urgently needed. According to the plant “cry for help” theory, plants under pathogen attack may recruit beneficial microbes via root exudates. However, whether A. sinensis employs this strategy against F. solani remains unknown. This study aimed to identify potential “cry for help” metabolites and evaluate their biocontrol potential. Methods: LC-MS analysis revealed that F. solani infection significantly altered the metabolic profiles of both A. sinensis roots and rhizosphere soil. Results: Comparative analysis identified seven metabolites specifically upregulated in infected plants but not detected in the pathogen, including taurine, oxoadipic acid, quinolinic acid, 6-phosphogluconic acid, methyl cinnamate, 2-phenylethanol, and (R)-3-hydroxybutyric acid. Exogenous application of these seven metabolites revealed that taurine and methyl cinnamate significantly alleviated disease symptoms, improved plant growth (root length, biomass), and enhanced the activities of key defense enzymes (peroxidase, POD, phenylalanine ammonia-lyase, PAL, lipoxygenase, LOX, polyphenol oxidase, PPO). Furthermore, taurine and methyl cinnamate reshaped the rhizosphere microbiome. The incidence of root rot was reduced by 51.3% and 50.8%, respectively. Taurine enriched actinobacteria (e.g., Paeniglutamicibacter) and reduced the relative abundance of pathogenic Ascomycota fungi, while methyl cinnamate markedly enriched the nitrogen-fixing bacterium Azotobacter and the saprophytic fungus Schizothecium. Crucially, both treatments significantly suppressed the proliferation of F. solani in the rhizosphere. Conclusions: Our findings demonstrate for the first time that A. sinensis activates a “cry for help” response upon attack by F. solani, with taurine and methyl cinnamate preliminarily identified as key signaling metabolites that can directly or indirectly inhibit the development of A. sinensis root rot. These compounds enhance plant resistance and recruit beneficial microorganisms, offering a novel and promising ecological strategy for the green control of A. sinensis root rot. Full article

The mountain pine beetle, Dendroctonus ponderosae, is a bark beetle that can cause extensive tree mortality of its hosts in western North America. Lodgepole pine, Pinus contorta, is one of its primary and most widely distributed hosts. The insect exhibits a one-year life cycle with a dispersal flight of emerging adults, referred to as brood adults, which attack new trees in the late spring–early summer. The larvae develop through the summer and overwinter. The following spring, development resumes, followed by pupation and emergence of a new generation of brood adults. Every year, a proportion of adults survive the winter, referred to as parent adults, re-emerge and attack new trees prior to brood adult emergence. These contribute little to population dynamics, but it is unknown whether these parent adults contribute to host finding by brood adults by initiating attacks that lead to clustering of attacks by brood adults. We tested this hypothesis by sequentially marking attacked trees from early spring to the fall in 2006, 2008, and 2015 in stands in northern Colorado. The sudden increase in the number of attacks indicates brood adult emergence. We tested for clustering of parent adult- and brood adult-attacked trees and of brood adult-attacked trees around parent adult-attacked trees using Ripley’s K-function. We found evidence of clustering of parent adult-attacked trees and brood adult-attacked trees, but there is no evidence of clustering of brood adult-attacked trees around parent adult-attacked trees. Full article

Securing the Industrial Internet of Things (IIoT) is paramount, yet proprietary protocols remain vulnerable to deep-state logic flaws that traditional fuzzers often fail to reach. We propose MALF, a Multi-Agent LLM Fuzzing Framework that couples a dynamic Industrial Security Knowledge Graph (ISKG) with collaborative cognitive agents for effective, efficient, and trustworthy IIoT security testing. A self-evolving knowledge loop mitigates LLM hallucinations by grounding the generation in verifiable graph constraints; QLoRA-tuned models aligned with hexadecimal features enable low-latency mutation; and Chain-of-Thought reasoning reconstructs protocol states for intent-driven attacks. On a heterogeneous testbed spanning five industrial protocols and ten vendors, MALF achieves an average Test Case Acceptance Rate of 88.3% (peak 91.2% on Modbus/TCP) and 91.2% ISKG-defined state coverage, outperforming rule-based, RL-based, and LLM baselines. On a 15-vulnerability N-Day benchmark, MALF detects all known cases, against 60%, 47%, 40%, and 27% for NCMFuzzer, MARLFuzz, BooFuzz, and Fuzz4All, respectively. In a separate real-world campaign, MALF further identifies 14 previously unknown vulnerability candidates, of which four have been assigned CNVD identifiers (CNVD-2024-16009, CNVD-2025-22875, CNVD-2025-29811, CNVD-2026-06041) and 10 remain under vendor review. These results provide controlled-testbed evidence that knowledge-grounded AI agents can systematically expose deep-state vulnerabilities in opaque IIoT environments. Full article

The possibility of zero-day attacks on Internet of Things (IoT) networks is high, particularly in dynamic and heterogeneous IoT environments, including emerging battlefield scenarios (IoBT). Detecting these attacks requires adaptive and generalizable security mechanisms. Due to the unique and unknown signatures of these attacks, they go undetected using signature-based Intrusion Detection Systems (IDSs) on the one side. On the other side, current anomaly-based IDSs that employ traditional machine learning on statistical features struggle to adapt and generalize to unknown networks, which is the case in IoBT. Transformer-based deep learning models have shown the capability of learning complex sequential patterns. This ability can be leveraged to analyze packet payloads that encompass opcodes capable of executing malicious patterns within an IoT network. In this work, we propose a dual-stage Transformer IDS that operates on the raw payload of network packets to detect zero-day attacks. Due to the lack of IoBT datasets, we evaluate the algorithm on three comprehensive IoT traffic benchmarks—MQTT-IoT, IoT-23, and CIC-IoT-2022—which have a high number of IoT devices and various attacks. Importantly, model evaluation is performed in two cross-validation settings to address the key operational challenges associated with unseen scenarios and networks. The evaluation settings are split-at-scenario to evaluate the detection ability of zero-day attacks and split-at-dataset to evaluate the model’s generalizability to new environments. In the former, the average increase in the F1-score of the proposed algorithm over the baseline model is 44% in detecting four zero-day attacks presented in the MQTT-IoT dataset. In the latter, the average increase in the F1-score is 16% in detecting malicious attacks across the three datasets. These results show the benefit of advanced AI in securing the next generation of IoT systems in future Internet applications. Full article

This paper investigates a tracking control problem for a class of strict-feedback nonlinear systems with time delays, asymmetric output constraints, and deception attacks on the controller. First, by introducing a new error conversion technology, any nonzero and bounded initial state is converted to zero, which not only solves the overshoot/oscillation problem of the output during the constraint switching phase but also unifies the control design of constrained and unconstrained systems. Second, a barrier function with asymmetric output constraints is designed, which converts the problem of satisfying the tracking control of nonlinear systems under output constraints into one of ensuring the boundedness. In addition, radial basis function neural networks (RBFNNs) are utilized to handle both unknown uncertain terms and deception attacks simultaneously. By utilizing the new asymmetric delayed barrier function error together with an RBFNN technique, the tracking error is ultimately uniformly bounded, regardless of the presence or absence of output constraints. Finally, the superiority of the proposed strategy is verified through its simulation on an unmanned aerial vehicle (UAV) system. Full article