This paper investigates the security path-tracking control problem for the connected vehicle swarm with uncertain dynamics, specifically emphasizing its resilience against dual-channel deception attacks. To ensure the security and reliability of data transmission against such cyber threats, a novel attack compensation mechanism is proposed to eliminate the impact of deception attacks by securing data transmission against signal tampering in both sensor-controller and controller-actuator channels. We design a fuzzy state estimator to reconstruct uncompromised signals from tampered data, thereby constructing a security control strategy resilient to deception attacks. Considering practical network bandwidth constraints, a fuzzy adaptive dynamic event-triggered path-tracking control strategy is developed, which integrates the proposed attack compensation mechanism and state estimator to mitigate the impact of cyber attacks. Sufficient conditions are established to guarantee the security path-tracking performance. With the Lyapunov stability theory, it is proved that error systems are uniformly ultimately bounded, and the Zeno phenomenon can be effectively prevented. Simulation results validate the effectiveness of the theoretical approach.

Applying deep reinforcement learning (DRL) to solve quantum control problems has become a popular research direction. However, the exploration capability and reward design for the learning agent, which usually affect the DRL’s application performance, has not been sufficiently emphasized. In this article, we propose a particle-assisted DRL (PDRL) method to address the above concern by enhancing exploration capabilities and designing appropriate reward functions for efficient quantum state manipulation. In PDRL, each episode in the quantum learning process is characterized by three kinds of events, i.e., unidentifiable, identifiable, and successful events. To improve exploration, exploration particles and feedback particles are employed in the early learning phase when episodes end in identifiable and successful events, respectively. To assign rewards, three event-based reward functions are provided for the DRL’s agent, exploration particles and feedback particles, respectively. Numerical results on single-qubit, two-qubit, and many-qubit systems validate the effectiveness of PDRL. Comparative results with existing DRL methods demonstrate the superior performance of PDRL for quantum state manipulation.

Wastewater treatment yields significant societal benefits in resource recycling, economic development, and public health. Dissolved oxygen (DO) concentration during the wastewater treatment process serves as a critical indicator for assessing effluent quality. Therefore, maintaining DO within an appropriate range is essential. This study proposes a data-driven tracking controller based on an action-dependent heuristic dynamic programming (ADHDP) approach incorporating lifelong learning (LL) to achieve precise DO concentration tracking. First, the LL-ADHDP controller, acting as an auxiliary controller, is integrated with a prior-knowledge-based controller to achieve model-free tracking control. The online LL-ADHDP approach enhances the approximation accuracy of both the critic and action networks. Second, integrating the LL mechanism into these networks mitigates catastrophic forgetting and improves overall robustness. Third, the method is applied to the benchmark simulation model no. 1. Experimental results demonstrate its superior tracking performance. Finally, simulations with diverse reference trajectories confirm the good dynamic performance and effectively reduce the tracking error of the proposed LL-ADHDP method.

Video-Text Retrieval (VTR) is a core task in multi-modal understanding, drawing growing attention from both academia and industry in recent years. While numerous VTR methods have achieved success, most of them assume accurate visual-text correspondences during training, which is difficult to ensure in practice due to ubiquitous noise, known as noisy correspondences (NC). In this paper, we rethink how to mitigate the NC from the perspective of representative reference features (termed agents), and propose a novel relation-aware purified consistency (RPC) network to amend direct pairwise correlation, including representative agents construction and relation-aware ranking distribution alignment. The proposed RPC enjoys several merits. First, to learn the agents well without any correspondence supervision, we customize the agents construction according to the three characteristics of reliability, representativeness, and resilience. Second, the ranking distribution-based alignment process leverages the structural information inherent in inter-pair relationships, making it more robust compared to individual comparisons. Extensive experiments on five datasets under different settings demonstrate the efficacy and robustness of our method.

Cross-Domain Few-Shot Segmentation (CD-FSS) extends the generalization ability of Few-Shot Segmentation (FSS) beyond a single domain, enabling more practical applications. However, directly employing conventional FSS methods suffers from severe performance degradation in cross-domain settings, primarily due to feature sensitivity and support-to-query matching process sensitivity across domains. Existing methods for CD-FSS either focus on domain adaptation of features or delve into designing matching strategies for enhanced cross-domain robustness. Nonetheless, they overlook the fact that these two issues are interdependent and should be addressed jointly. In this work, we tackle these two issues within a unified framework by optimizing features in the frequency domain and enhancing the matching process in the spatial domain, working jointly to handle the deviations introduced by the domain gap. To this end, we propose a coherent Dual-Agent Optimization (DATO) framework, including a consistent mutual aggregation (CMA) and a correlation rectification strategy (CRS). In the consistent mutual aggregation module, we employ a set of agents to learn domain-invariant features across domains, and then use these features to enhance the original representations for feature adaptation. In the correlation rectification strategy, the agent-aggregated domain-invariant features serve as a bridge, transforming the support-to-query matching process into a referable feature space and reducing its domain sensitivity. Extensive experiments demonstrate the efficacy of our approach.

Category-level object pose estimation aims to determine the pose and size of novel objects in specific categories. Existing correspondence-based approaches typically adopt point-based representations to establish the correspondences between primitive observed points and normalized object coordinates. However, due to the inherent shape-dependence of canonical coordinates, these methods suffer from semantic incoherence across diverse object shapes. To resolve this issue, we innovatively leverage the sphere as a shared proxy shape of objects to learn shape-independent transformation via spherical representations. Based on this insight, we introduce a novel architecture called SpherePose, which yields precise correspondence prediction through three core designs. Firstly, We endow the point-wise feature extraction with SO(3)-invariance, which facilitates robust mapping between camera coordinate space and object coordinate space regardless of rotation transformation. Secondly, the spherical attention mechanism is designed to propagate and integrate features among spherical anchors from a comprehensive perspective, thus mitigating the interference of noise and incomplete point cloud. Lastly, a hyperbolic correspondence loss function is designed to distinguish subtle distinctions, which can promote the precision of correspondence prediction. Experimental results on CAMERA25, REAL275 and HouseCat6D benchmarks demonstrate the superior performance of our method, verifying the effectiveness of spherical representations and architectural innovations.

DETR-based methods, which use multi-layer transformer decoders to refine object queries iteratively, have shown promising performance in 3D indoor object detection. However, the scene point features in the transformer decoder remain fixed, leading to minimal contributions from later decoder layers, thereby limiting performance improvement. Recently, State Space Models (SSM) have shown efficient context modeling ability with linear complexity through iterative interactions between system states and inputs. Inspired by SSMs, we propose a new 3D object DEtection paradigm with an interactive STate space model (DEST). In the interactive SSM, we design a novel state-dependent SSM parameterization method that enables system states to effectively serve as queries in 3D indoor detection tasks. In addition, we introduce four key designs tailored to the characteristics of point cloud and SSM: The serialization and bidirectional scanning strategies enable bidirectional feature interaction among scene points within the SSM. The inter-state attention mechanism models the relationships between state points, while the gated feed-forward network enhances inter-channel correlations. To the best of our knowledge, this is the first method to model queries as system states and scene points as system inputs, which can simultaneously update scene point features and query features with linear complexity. Extensive experiments on two challenging datasets demonstrate the effectiveness of our DEST-based method. Our method improves the GroupFree baseline in terms of $\text{AP}_{50}$ on ScanNet V2 (+5.3) and SUN RGB-D (+3.2) datasets. Based on the VDETR baseline, Our method sets a new state-of-the-art on the ScanNetV2 and SUN RGB-D datasets.

3D Instance segmentation (3DIS) on point clouds is a fundamental task in the field of 3D scene understanding. Existing fully-supervised networks have achieved promising results but remain heavily reliant on point-wise annotated data. Using the instance bounding boxes as annotations for weakly-supervised learning is a feasible way to solve the label-efficiency problem. In this paper, we propose POTNet, a novel training paradigm designed to generate point-wise pseudo-labels using only bounding box annotations, which effectively considering both local and global information. We leverage prototype learning method to extract local features from the non-overlapping regions indicated by the bounding boxes as instance prototypes. We employ an optimal transport algorithm to assign points in overlapping regions to their corresponding instances based on the similarity matrix between prototypes and these points. Our approach enables the generation of point-wise pseudo-labels that fully account for local and global correlations. We demonstrate the effectiveness of our method by achieving performance comparable to state-of-the-art approaches on multiple datasets, without requiring any additional supplementary data or retraining processes.

In this paper, a Zeroing Neural Network (ZNN)-based control framework is proposed for finite-time visual servoing of robotic manipulators, without requiring camera calibration or kinematic modeling. To address the challenge of the unknown robot-camera interaction, a data-driven Jacobian estimator is introduced, enabling real-time mapping without offline training or analytical derivation. A finite-time noise-rejection ZNN (FTNRZNN) controller is developed to ensure robust and fast joint-level control under measurement noise. The continuous-time scheme is further discretized for digital implementation. Rigorous Lyapunov analysis guarantees finite-time convergence. Simulations and real-world experiments validate the effectiveness of the method in both regulation and trajectory tracking, demonstrating strong adaptability to unstructured environments.

This article focuses on a dynamic event-triggered adaptive neural networks backstepping control for a class of uncertain strict-feedback systems with communication constraints. The uncertain terms including external disturbances and unknown nonlinear functions are approximated by radial basis function neural networks, in which the weight update laws are obtained via the gradient descent algorithm, ensuring the local boundedness of the approximation error of neural networks. Then, to enhance the transmission efficiency of control signals, a dynamic event-triggered mechanism is introduced, which enables the dynamic adjustment of threshold parameters in response to the actual tracking performance. It is strictly proved via the Lyapunov stability criterion that the tracking error can converge to a desired small neighborhood of the origin, and all signals in the closed-loop system are bounded. Finally, the validity of the control strategy is demonstrated through a simulation example.Note to Practitioners— In practical network control systems, control signals are typically transmitted continuously or periodically to devices through the communication network in the form of data packets. As communication networks are usually shared by various system nodes, and resources such as communication channel bandwidth and computational capabilities are limited, improving the transmission efficiency of control signals becomes a crucial design problem for controllers in network control systems. Therefore, This study introduces a control method via event-triggered sampling, aiming to enhance sampling efficiency while ensuring the stability and reliability of the system. The proposed control method is suitable for a broad category of strict-feedback nonlinear systems with communication constraints, offering notable advantages such as low-complexity design and straightforward implementation.

In industries such as finance, healthcare, and new energy vehicles, data classification and grading standards ensure regulatory compliance and protect sensitive information. However, automating text file classification under these standards presents several challenges. Traditional machine learning and deep learning approaches require large labeled datasets, which are often scarce. Existing classification methods are typically domain-specific, limiting cross-domain adaptability. Moreover, many approaches simply categorize documents as regulatory or nonregulatory and assign security levels, but fail to map them accurately to specific rules. To address these challenges, this article proposes prompt-driven grading and classification algorithm (PGCA), a prompt learning-based method for text classification and grading. PGCA integrates a structured feature repository and SQL-inspired prompt templates to efficiently extract and match features from text documents, establishing mappings between text, and classification rules and grading standards. Furthermore, the integration of a preclassification strategy enables the filtration of irrelevant rules, thereby substantially reducing computational overhead. Experiments show that PGCA achieves classification accuracy between 95.0% and 99.0%, outperforming baselines such as TsF-KNN, Gen-DT, bt-SVM, AGCRCNN, and AC-BiLSTM by 4%–25%. Additionally, the preclassification stage cuts computational costs by 63.7% while keeping accuracy loss to within 1%.

Depression is a major mental health issue in contemporary society, with an estimated 350 million people affected globally. The number of individuals diagnosed with depression continues to rise each year. Currently, clinical practice relies entirely on self-reporting and clinical assessment, which carries the risk of subjective biases. In this article, we propose a multimodal method based on facial expression and pupil to detect depression more objectively and precisely. Our method first extracts the features of facial expressions and pupil diameter using residual networks and 1-D convolutional neural networks. Second, a cross-modal fusion model based on self-attention networks (CMF-SNs) is proposed, which utilizes cross-modal attention networks within modalities and parallel self-attention networks between different modalities to extract CMF features of facial expressions and pupil diameter, effectively complementing information between different modalities. Finally, the obtained features are fully connected to identify depression. Multiple controlled experiments show that compared to single modality, the multimodal fusion method based on self-attention networks has a higher ability to recognize depression, with the highest accuracy of 75.0%. In addition, we conducted comparative experiments under three different stimulation paradigms, and the results showed that the classification accuracy under negative and neutral stimuli was higher than that under positive stimuli, indicating a bias of depressed patients toward negative images. The experimental results demonstrate the superiority of our multimodal fusion method.

Event cameras are bio-inspired sensors with diverse advantages, including high temporal resolution and minimal power consumption. Therefore, event cameras enjoy a wide range of applications in computer vision, among which event keypoint detection plays a vital role. However, repeatable event keypoint detection remains challenging because the lack of temporal interframe interaction leads to descriptors with limited temporal consistency, which restricts the ability to perceive keypoint motion. Besides, detectors learned at single scale features are not suitable for event keypoints with significant motion speed differences in high-speed scenarios. To deal with these problems, we propose a novel Spatio-Temporal Pyramid Keypoint Detection Network (STPNet) for event cameras via a temporally consistent descriptor learning (TCL) module and a spatially diverse detector learning (SDL) module. The proposed STPNet enjoys several merits. First, the TCL module generates temporally consistent descriptors for specific keypoint motion patterns. Second, the SDL module produces spatially diverse detectors for applications in high-speed motion scenarios. Extensive experimental results on three challenging benchmarks show that our method notably outperforms state-of-the-art event keypoint detection methods. Specifically, our STPNet can outperform the best event keypoint detection method by 0.21px in reprj. error on Event-Camera, 4% in IoU on N-Caltech101, 0.13px in reprj. error on HVGA ATIS Corner and 5.94% in matching accuracy on DSEC.

This paper investigates the attack-compensated output control problem for a class of Markov jump cyber-physical systems (MJCPSs) subject to mismatched modes. A sub-domain-awareness adaptive probabilistic event-triggered mechanism (APETM) is innovatively developed to fully enhance the control performance of the networked control system. Here, sub-domain-awareness means the network’s communication ability, which may be affected by time delay and network resource. To defend against the cyber-attacks imposed on the communication channel, a predictor-based compensator is constructed to mitigate the impact of attacks on the control performance. Different from the existing works, the mismatch degree between the controller and the system, governed by a random variable obeying any discrete-time distribution over$[0, 1]$, is changing dynamically. A mismatched output feedback controller is designed based on the APETM and attack-compensator, and an augmented closed-loop system is obtained. Both time delay-and triggering threshold-dependent Lyapunov functionals are introduced to perform the asymptotical stability analysis. With a separation technique, a solving algorithm is provided which converts the non-convex analysis conditions into traceable ones. Finally, simulations are conducted to validate the proposed results through a mass-spring-damping system model.Note to Practitioners—Cyber-physical systems (CPSs) have found many applications in the fields such as smart grid, gas distribution systems, and automated manufacturing systems. The open communication network deployed in CPSs face two main challenges including the security of data transmission and the efficiency of resource utilization. Owing to the practical environment and other factors, system parameters may change abruptly, which can be well defined by a Markov model. In the resultant Markov jump system, it is typically assumed that the mode mismatch degree between the controller and the system remains the same, which is unrealistic in a time-varying network environment. Motivated by these observations, this paper investigates the secure output control of Markov jump CPSs (MJCPSs) subject to mismatched modes and unknown cyber-attacks. An adaptive probabilistic event-triggered mechanism is innovatively proposed, which compared with the existing mechanisms strikes a higher level balance between the communication burden and the control performance. And attack-compensated mismatched output controllers are developed to stabilize MJCPSs and meanwhile preserve the security level. Finally, a mass-spring-damping system is adopted to verify the application effectiveness of the proposed results. Note that communication traffic control and secure control are finely involved based on CPSs, making our proposed results more applicable to practical systems.

This article investigates the leader-following consensus problem of discrete-time nonlinear multiagent systems (MASs) over fading channels. With the consideration of the transmission among followers maybe affected by the fading networks, the nonidentical fading channels model is constructed. To reduce the transmission network burden, the dynamical event-triggered mechanism (DETM) is developed. Different from most of existing event-triggered strategies, the threshold parameter in the developed dynamical event-triggering condition is dynamically adjusted according to a dynamic rule. Based on the DETM, a distributed consensus control protocol is designed under fading channels. Then, sufficient criteria are provided to ensure that MASs can achieve the leaser-following consensus, and satisfy the$H_{\infty}$performance index in the presence of fading channels. The desired controller parameters can be derived in terms of solutions of matrix inequalities that are lightly solvable. In the end, simulation results show that the designed dynamical event transmission policy is capable of diminishing communication burden more promptly and effectively than some existing ones.

In this article, a synchronization control method is studied for coupled neural networks (CNNs) with constant time delay using sampled-data information. A distributed control protocol relying on the sampled-data information of neighboring nodes is proposed. Lyapunov functional is constructed to analyze the synchronization of CNNs with constant time delay. Using Park's integral inequality and improved free-weight matrix integral inequality, sufficient conditions are provided for CNNs to achieve synchronization with less conservatism. In addition, the maximum sampling interval is determined by transforming the sufficient conditions into an optimization problem, and an aperiodic sampling control technique is implemented to reduce the communication energy load. Finally, numerical simulations are provided to demonstrate that the proposed method is capable of achieving synchronization.

This article presents a prescribed-time tracking control method for underactuated unmanned surface vessels (USVs) using a neural reinforcement learning (RL) approach. First, the hand position approach, addressing the underactuated characteristic, is employed to convert the model of USV into the integral cascade form. Second, inheriting the advantages of prescribed performance control (PPC), the proposed controller not only stabilizes the tracking error within an asymmetric prescribed-time range, but also removes the limitation of initial conditions. Subsequently, the identifier—actor-critic architecture is introduced in the optimized backstepping design, which gives the solution of the Hamilton–Jacobi–Bellman (HJB) equation. Meanwhile, the relative threshold event-triggered mechanism is also considered to reduce the communication burden and executive frequency of actuators. Finally, employing the Lyapunov stability theory, it is proven that all signals in the closed-loop system are bounded, and the developed control scheme is demonstrated to be effective through simulation and experimental results.

In this paper, we investigate a neural adaptive formation control problem for underactuated unmanned surface vehicles (USVs). Considering the limitation of communication distance and the security of formation systems, collision-free and connectivity maintenance are guaranteed by defining a prescribed-time tuning function and proper error transformation. Furthermore, a new nonlinear first-order filter, solving the complexity problem, is designed to promote the system performance. Subsequently, neural networks (NNs) are used to approximate USVs’ dynamics and their transient performance is improved by prediction error. By blending prediction errors and neural approximation, it is guaranteed the general external disturbances and approximation errors are compensated via constructed disturbance observers (DOs), simultaneously. Meanwhile, utilizing the minimal number of learning parameters (MNLPs) methodology, the number of NNs’ learning parameters can be significantly reduced. It is rigorously proved that all signals in the closed-loop system are bounded via Lyapunov stability theorem. Finally, simulation and experimental studies are presented to verify the effectiveness and advantages of theoretical results.

The broad learning system (BLS) of intelligent vehicle in different target environments is studied in this article. First, this article provides with the target recognition image data to be trained and detected through the automated guided vehicle (AGV) mobile platform, which can grab the recognition image of different angles and backgrounds. In order to avoid the data generalization phenomenon, the dataset can be expanded by the data normalization and data enhancement. Second, the data are input into the shared convolution layer to extract the feature image and maintain the image. The parameters of image height, width, and channel number are invariable, and the new feature image is obtained by further extraction. Furthermore, the region proposal network (RPN) prefiltering algorithm based on hierarchical clustering is used to filter the objects in the candidate box to determine the region image corresponding to the feature image. Then, the feature images of different sizes input into region of interest (ROI) pooling are used to keep the size of the image in the ROI consistent. Finally, the normalized image is input into the classifier module to obtain the category of the target recognition image to be detected. Through the simulation experiments of different groups, it can be seen that the target recognition system proposed in this design can not only accurately detect the objects but also stably recognize the objects in different environments. The target recognition accuracy for the optimized system is about 95%.

Cloud computing has become an important IT infrastructure in the big data era; more and more users are motivated to outsource the storage and computation tasks to the cloud server for convenient services. However, privacy has become the biggest concern, and tasks are expected to be processed in a privacy-preserving manner. This paper proposes a secure SIFT feature extraction scheme with better integrity, accuracy and efficiency than the existing methods. SIFT includes lots of complex steps, including the construction of DoG scale space, extremum detection, extremum location adjustment, rejecting of extremum point with low contrast, eliminating of the edge response, orientation assignment, and descriptor generation. These complex steps need to be disassembled into elementary operations such as addition, multiplication, comparison for secure implementation. We adopt a serial of secret-sharing protocols for better accuracy and efficiency. In addition, we design a secure absolute value comparison protocol to support absolute value comparison operations in the secure SIFT feature extraction. The SIFT feature extraction steps are completely implemented in the ciphertext domain. And the communications between the clouds are appropriately packed to reduce the communication rounds. We carefully analyzed the accuracy and efficiency of our scheme. The experimental results show that our scheme outperforms the existing state-of-the-art.

Deep reinforcement learning (DRL) is a machine learning method based on rewards, which can be extended to solve some complex and realistic decision-making problems. Autonomous driving needs to deal with a variety of complex and changeable traffic scenarios, so the application of DRL in autonomous driving presents a broad application prospect. In this article, an end-to-end autonomous driving policy learning method based on DRL is proposed. On the basis of proximal policy optimization (PPO), we combine a curiosity-driven method called recurrent neural network (RNN) to generate an intrinsic reward signal to encounter the agent to explore its environment, which improves the efficiency of exploration. We introduce an auxiliary critic network on the original actor-critic framework and choose the lower estimate which is predicted by the dual critic network when the network update to avoid the overestimation bias. We test our method on the lane- keeping task and overtaking task in the open racing car simulator (TORCS) driving simulator and compare with other DRL methods, experimental results show that our proposed method can improve the training efficiency and control performance in driving tasks.

Visible-infrared person re-identification (VI-ReID) is challenging due to the large modality discrepancy between visible and infrared images. Existing methods mainly focus on learning modality-shared representations by embedding images from different modalities into a common feature space, in which some discriminative modality information is discarded. Different from these methods, in this paper, we propose a novel Modality-Specific Memory Network (MSMNet) to complete the missing modality information and aggregate visible and infrared modality features into a unified feature space for the VI-ReID task. The proposed model enjoys several merits. First, it can exploit the missing modality information to alleviate the modality discrepancy when only the single-modality input is provided. To the best of our knowledge, this is the first work to exploit the missing modality information completion and alleviate the modality discrepancy with the memory network. Second, to guide the learning process of the memory network, we design three effective learning strategies, including feature consistency, memory representativeness and structural alignment. By incorporating these learning strategies in a unified model, the memory network can be well learned to propagate identity-related information between modalities and boost the VI-ReID performance. Extensive experimental results on two standard benchmarks (SYSU-MM01 and RegDB) demonstrate that the proposed MSMNet performs favorably against state-of-the-art methods.

Occluded person re-identification (Re-ID) is a challenging task as persons are frequently occluded by various obstacles or other persons, especially in the crowd scenario. To address these issues, we propose a novel end-to-end Part-Aware Transformer (PAT) for occluded person Re-ID through diverse part discovery via a transformer encoder-decoder architecture, including a pixel context based transformer encoder and a part prototype based transformer decoder. The proposed PAT model enjoys several merits. First, to the best of our knowledge, this is the first work to exploit the transformer encoder-decoder architecture for occluded person Re-ID in a unified deep model. Second, to learn part prototypes well with only identity labels, we design two effective mechanisms including part diversity and part discriminability. Consequently, we can achieve diverse part discovery for occluded person Re-ID in a weakly supervised manner. Extensive experimental results on six challenging benchmarks for three tasks (occluded, partial and holistic Re-ID) demonstrate that our proposed PAT performs favor-ably against stat-of-the-art methods.