Machine learning (ML) has gained significant adoption in Android malware detection to address the escalating threats posed by the rapid proliferation of malware attacks. However, recent studies have revealed the inherent vulnerabilities of ML-based detection systems to evasion attacks. While efforts have been made to address this critical issue, many of the existing defensive methods encounter challenges such as lower effectiveness or reduced generalization capabilities. In this paper, we introduce MalPurifier, a novel adversarial purification framework specifically engineered for Android malware detection. Specifically, MalPurifier integrates three key innovations: a diversified adversarial perturbation mechanism for robustness and generalizability, a protective noise injection strategy for benign data integrity, and a Denoising AutoEncoder (DAE) with a dual-objective loss for accurate purification and classification. Extensive experiments on two large-scale datasets demonstrate that MalPurifier significantly outperforms state-of-the-art defenses. It robustly defends against a comprehensive set of 37 perturbation-based evasion attacks, consistently achieving robust accuracies above 90.91%. As a lightweight, model-agnostic, and plug-and-play module, MalPurifier offers a practical and effective solution to bolster the security of ML-based Android malware detectors.

Machine learning (ML) models for Android malware detection face escalating threats from adversarial query attacks, which iteratively perturb malware samples guided solely by model outputs. Existing stateful defenses primarily detect such attacks by analyzing query similarities or distribution anomalies. However, these methods are less effective in the malware domain, where discrete feature spaces and strict functional constraints combine with their reliance on only a single similarity or distribution metric, making it difficult to capture the varied patterns of query-based attacks. In this paper, we introduce MalShield, a stateful defense model that monitors the query stream through an adaptive sliding window, which adjusts its span according to the incoming query rate to preserve relevant context and enable real-time computation of multiple anomaly indicators. We design multiple complementary anomaly indicators, such as similarity decay and feature growth, to capture both temporal patterns and deviations in feature space, enabling more sensitive detection of query attacks. The K-of-N voting rule alerts only when enough indicators agree, cutting false positives and adapting to varied query attacks, thus providing robust, interpretable defense without changing the classifier. Experiments on Androzoo and Drebin datasets, against eight black-box query attacks, show MalShield achieves more than 96% reduction in evasion rates, surpassing existing stateful defenses and at best outperforms existing defense models by 100%.

Edge computing (EC) and container technology have been widely used to increase the flexibility of computing resources and meet the real-time requirements for delay-sensitive applications. However, it has been shown that edge clouds suffer from distributed denial-of-service (DDoS) attacks, especially low-rate DDoS (LDDoS) attacks, which can be stealthily crafted to evade detection. Unfortunately, the existing techniques cannot provide effective protection, and the amplifying resource consumption and service delay incurred by defense greatly diminish the efficiency of the security system. To tackle these problems, this paper exploits Moving Target Defense (MTD) techniques and deep reinforcement learning (DRL) for mitigating the impact of LDDoS attacks in a resource-efficient way by effectively partially invalidating, avoiding, and tolerating malicious traffic that improves the Web services’ security and quality with lower overhead. We first design several lightweight MTD mechanisms by utilizing the built-in functionalities of container-based applications. To further optimize resource utilization, we formulate the interaction between attacks and MTD deployment as a Markov decision process (MDP), and adopt a deep Q-network (DQN) algorithm to achieve the best trade-off between effectiveness and overhead. The simulations prove the effectiveness of the proposed approach in LDDoS mitigation, with a significant improvement of up to 31.7% in security and 26.95% in service quality when compared with other practical strategies, and the experimental results also demonstrate that our method exhibits the lowest response time per request of 276.66 ms and the lowest webpage load time of 1.413 s with only 2.44% additional memory usage in comparison with previous works in the high workload scenario.

Machine learning (ML) has been widely adopted for Android malware detection to deal with serious threats brought by explosive malware attacks. However, it has been recently proven that ML-based detection systems exhibit inherent vulnerabilities to evasion attacks, which inject adversarial perturbations into a malicious app to hide its malicious behaviors and evade detection. To date, researchers have not found effective solutions for this critical problem. Although there are some similar works in the image classification field, most of those ideas cannot be borrowed due to the significant differences between images and Android apps. In this paper, we exploit Moving Target Defense (MTD) to continually change the attack surface of the protected detector and create uncertainty on the attacker side. We thus propose a novel Android malware detection framework named MTDroid, which fully leverages a seamless blend of dynamicity, diversity, and heterogeneity to mitigate the impact of evasion attacks. To this end, we develop a dynamic model pool to decrease the exposure time of a single classifier, by building and rebuilding multiple heterogeneous models with distinct data. We then generate diversified variant models to provide defensive measures against various attacks, and further improve robustness through ensemble learning. Specifically, we propose a two-stage selection algorithm to optimize the ensemble learning process, and design a hybrid update strategy to refresh the framework dynamically. The experimental results show that MTDroid significantly enhances the robustness against a wide range of attacks and outperforms the state-of-the-art methods upon three popular practical datasets.

This article presents an event-triggered data-driven load frequency control (LFC) method for multiarea interconnected power systems via model-free adaptive control, where the dynamic model of the power system is assumed to be unknown completely. By introducing the dynamic linearization technique for the unknown power system, an equivalent data relationship model between the area-control-error (ACE) data and the input signal is established. Then, a data-driven LFC scheme is developed only relying on the input and output data of the power system. Meanwhile, an event-triggered strategy is also proposed in the design of data-driven LFC such that the communication and computation burden of the system can be reduced. Whether the current instant is the transmission instant is determined by judging the proposed triggering condition at each sampling instant. It is showed that the presented event-triggered data-driven LFC method is independent to any model information of the power system and does not need to measure any state signals. Simulation tests are carried out to verify the effectiveness of the presented control method.

Deep learning has great potential to reshape manufacturing industries. In this article, we present DeepWelding, a novel framework that applies deep learning techniques to improve gas tungsten arc welding process monitoring and penetration detection using multisource sensing images. The framework is capable of analyzing multiple types of optical sensing images synchronously and consists of three deep learning enhanced consecutive phases: image preprocessing, image selection, and weld penetration classification. Specifically, we adopted generative adversarial networks (pix2pix) for image denoising and classic convolutional neural networks (AlexNet) for image selection. Both pix2pix and AlexNet delivered satisfactory performance. However, five individual neural networks with heterogeneous architectures demonstrated inconsistent generalization capabilities in the classification phase when holding out multisource images generated with specific experimental settings. Therefore, two ensemble methods combining multiple neural networks are designed to improve the model performance on unseen data collected from different experimental settings. We have also found that the quality of model prediction is heavily influenced by the data stream collection environment. We think these findings are beneficial for the broad intelligent welding community.

Aquatic animal movement results from a complex balance between muscular actuation, swimmer's inertia, damping, and stiffness; as well as, the effects from the fluid environment. Most aquatic animals utilize undulatory propulsion methods during swimming. Propulsion mode transition involves a variation of these parameters, and to better investigate the variation of these parameters during propulsion mode switching, and provide guidance for swimming robot design, we studied propulsion mechanism of undulation locomotion by combining biological investigation, mathematical simulation and experimental validation. A modular robot platform, with assembling function, was built based on the obtained biological features to realize the corresponding propulsion methods. Then a modular dynamic modeling method was proposed to simulate robot locomotion using a CPG based algorithm and a PD control method, further revealing the underlying mechanism for undulatory locomotion. Finally, experiments were conducted using the robotic platform to validate the found conclusions as well as enhance the propulsion mechanism of undulatory motion, providing a generic guidance for swimming robot design.

The whirligig beetle, claimed to be one of the most energy-efficient swimmers in the animal kingdom, has evolved a series of propulsion strategies that may serve as a source of inspiration for the design of propulsion mechanisms for energy-efficient surface swimming. In this paper, we introduce a robot platform that was developed to test an energy-efficient propulsion mechanism inspired by the whirligig beetle. A propulsor-body-fluid interaction dynamics model is proposed, and based on this model, the propulsor flexural rigidity and beating patterns are optimized in order to achieve energy-efficient linear swimming and turning. The optimization results indicate that a propulsor with decreasing flexural rigidity enhances vortex shedding and improves thrust generation. It has also been found that an alternating asymmetrical beating sequence and optimal beating frequency of 0.71 Hz improves propulsion efficiency for linear swimming of the robot. The alternating beating of the outboard propulsors and the unfolded inboard propulsors working as brakes results in efficient turning with a smaller turning radius. Both simulation and experimental studies were conducted, and the results illustrate that decreasing flexural rigidity along the propulsor length, an oscillating body motion, and an S-shaped trajectory are critical for energy-efficient propulsion of the robot.

The bio-inspired materials and bio-inspired robotics laboratory at the University of Tennessee developed a swimming robot with a rigid body and flexible propellers inspired from the whirligig beetle. We improved the propulsion efficiency by identifying the optimal structure for the propellers and beating patterns to minimize the energy consumption for this swimming robot. In this video, we present the design, simulation, and experiment of the robot platform.

Whirligig beetle, claimed in the literature to be one of the highest measured for a thrust-generating apparatus within the animal kingdom, has evolved a series of propulsion strategies that may serve as a source of inspiration for designing highly efficient propulsive systems. First, a robotic platform was developed to test an energy-efficient propulsion mechanism inspired by the whirligig beetle. Second, a mathematical model for the robot was proposed to account for the fluid dynamics generated by the robotic swimming. Third, an optimal problem was formulated and solved for the propulsor and beating pattern design. The results indicated that soft middle, stiff end propulsor, and alternating, asymmetrical beating pattern will improve the propulsion efficiency for a swimming robot with four propulsors. Finally, simulation and experiments were conducted to further analyze the effect of beating pattern to the robotic propulsion efficiency. It was found that the oscillated body movement and S-shaped trajectory introduced by the optimal beating pattern would improve the propulsion efficiency for the designed robot.