This paper studies the resilient cooperative output regulation problem of uncertain linear multi-agent systems under denial-of-service (DoS) attacks. An integrated resilient distributed observer and a resilient distributed canonical internal model approach have been proposed. First, a hybrid sampling mechanism that mixes event-triggered and time-triggered scenarios has been adopted, with which a sampled output based resilient distributed observer is designed against DoS attacks. Second, the resilient cooperative output regulation problem is converted into an input-output stabilization problem of an augmented system composed of the plant, the reduced-order observer, and the resilient distributed canonical internal model. Then, an adaptive stabilizer is synthesized to stabilize the augmented system using a recursive iteration approach. Finally, the effectiveness of the proposed control design is verified by an example of the quarter-car active automotive suspension system.

Semantic segmentation is a critical process in remote sensing image analysis, supporting various applications. The recent development of the segment anything model (SAM), a visual foundation model designed to segment-anything, highlights the potential of foundational models in computer vision. However, SAM generates segmentation results without category labels, and predictions from semantic segmentation models for remote sensing often exhibit excessive fragmentation and imprecise boundaries. To address these limitations, we propose a strategy that integrates SAM with semantic segmentation models, replacing the imprecise boundaries of remote sensing segmentation masks with the more boundary-accurate SAM masks while retaining the original semantic information. Subsequently, a framework is designed and realized to enhance the prediction results of semantic segmentation models for remote sensing imagery by leveraging the raw outputs generated by SAM. This approach requires no additional training, modification to the semantic segmentation model, or changes to the visual foundation model, making it efficient and straightforward compared with other methods. Specifically, experimental results on two well-known datasets, LoveDA Urban and ISPRS Potsdam, demonstrate the effectiveness and broad applicability of our approach. In addition, incorporating recent visual foundation models, such as SAM-HQ and semantic SAM, further improves segmentation accuracy. As these models advance, the potential of our framework to enhance the performance of semantic segmentation for remote sensing imagery will grow. The source code for this work will be accessible athttps://github.com/qycools/SESSRS.

Network pruning is an effective technique to reduce computation costs for deep model deployment on resource-constraint devices. Searching superior sub-networks from a vast search space through Neural Architecture Search (NAS) , which conducts a one-shot supernet used as a performance estimator, is still time-consuming. In addition to searching inefficiency, such solutions also focus on FLOPs budget and suffer from an inferior ranking consistency between supernet-inherited and stand-alone performance. To solve the problems above, we propose a framework, namely DBS. Firstly, we pre-sample sub-networks with a similar budget setting as starting points, then we use a strict path-wise fair sandwich rule to train these starting points in a supernet. Second, we train Transformer-based predictors according to the performance and budget (FLOPs or latency) of starting points. After that, we freeze the parameters of predictors and apply a differentiable budgetaware search on continuous sub-networks vectors. Finally, we obtain the derived sub-networks from the optimized vectors by a decoder. We conduct comprehensive experiments on Imagenet with Resnet and Mobilenet-V2 under various FLOPs settings as well as different latency, which shows consistent improvements to the-state-of-art methods.

The cooperative robust resilient output regulation problem is investigated in this article for linear uncertain multiagent systems under denial-of-service (DoS) attacks. To tackle arbitrarily large parametric uncertainties, the canonical internal model is attached to the multiagent system, converting the cooperative robust resilient output regulation problem into the auxiliary event-triggered stabilization problem of the augmented error system. Then, employing a hybrid sampling mechanism mixing event-triggered and time-triggered scenarios, the augmented error system is stabilized by a novel sampled distributed output feedback control law. Moreover, rigorous proof shows that the Zeno behavior shall not happen. Finally, the proposed control scheme is applied to solve the geometric formation problem for a platoon of vehicles under DoS attacks.

With the advent of Neural Architecture Search (NAS), auto-designing of artificial neural networks has been made possible. Among various NAS methods, Differentiable Architecture Search (DARTS) has achieved significant progress due to its high calculating efficiency. However, it suffers from poor stability and obvious performance drop because of bi-level optimization and hard pruning. Besides, it only generates one best architecture at once. To alleviate the problems above, we design a three-stage framework with a path-wise weight sharing derivation. We first prune the supernet with differentiable methods to keep top-k operations on each edge instead of one. Then the pruned supernet is trained with our path-wise weight sharing method. At the derivation stage, the best candidate operations are selected with Evolutionary Search based on the validation accuracy of paths. Our weight sharing derivation is proved effective in improving searching stability as well as alleviating the performance drop. Furthermore, it also allows us to search for a large number of architectures with different parameter sizes at one time. Comprehensive experiments on CIFAR-10 and ImageNet show that we manage to find a group of state-of-the-art architectures (97.61% on CIFAR-10 and 76.4% on ImageNet).

This article demonstrates the realization of angle tracking and deformation suppression by developing two boundary controllers for a flexible variable-length rotary crane arm with extraneous disturbances and asymmetric input-output constraints. The dynamic model description of this kind of crane arm system is several partial differential equations integrated into few ordinary differential equations. The S-curve acceleration and deceleration scheme is utilized to adjust the elongation rate of the arm. A kind of novel observer is put forward to tackle unknown extraneous disturbances. Auxiliary systems and barrier Lyapunov functions are introduced to meet the asymmetric input-output constraints. With the help of Lyapunov's theory, the global exponential stability and uniform boundedness are analyzed. The numerical simulations are finally provided to illuminate its availability of the designed control schemes.

This article constructs two adaptive control laws to achieve deformation reduction and attitude tracking for a rotary variable-length crane arm with system parameter uncertainties and asymmetric input-output constraints. Two auxiliary systems are given to deal with the input constraints, an asymmetric-logarithm-barrier Lyapunov function is established for achieving the asymmetric output constrains, and five adaptive laws are constructed to handle system parameter uncertainties. Besides, the control design is based on a partial differential equation model, and the S-curve acceleration and deceleration method is used for regulating the arm extension speed. Both the system stability and uniform ultimate boundedness of the controlled crane arm are analyzed. Simulation results validate the effectiveness of our established control laws.

This article focuses on the vibration reducing and angle tracking problems of a flexible unmanned spacecraft system subject to input nonlinearity, asymmetric output constraint, and system parameter uncertainties. Using the backstepping technique, a boundary control scheme is designed to suppress the vibration and regulate the angle of the spacecraft. A modified asymmetric barrier Lyapunov function is utilized to ensure that the output constraint is never transgressed. Considering the system robustness, neural networks are used to handle the system parameter uncertainties and compensate for the effect of input nonlinearity. With the proposed adaptive neural network control law, the stability of the closed-loop system is proved based on the Lyapunov analysis, and numerical simulations are carried out to show the validity of the developed control scheme.

This article studies the formation tracking control problem for a group of underactuated surface vehicles with guaranteed transient properties, including connectivity maintenance, collision avoidance, and tracking performance specifications. The formation is within the leader–follower control framework, in which every follower is controlled to track its leader and maintain a desired relative distance and bearing angle with respect to its leader such that the prescribed formation geometry is achieved based on local sensing capability. The onboard sensor systems are of limited range and angle of view, thus defining a cone of detectable region for every follower. Each follower can detect its leader, if and only if the relative distance and bearing angle keep always inside the predefined detectable region such that the connectivity between the follower and its leader is maintained over time. In addition to the consideration of connectivity maintenance, no collision between the follower and its leader is also considered. A transverse function control approach is introduced to overcome the difficulties caused by the off-diagonal system matrix and underactuation. The barrier Lyapunov function and adaptive backstepping procedure are incorporated into the formation control design to achieve the boundedness of the closed-loop systems with guaranteed transient performance. Collision avoidance and connectivity maintenance between every follower and its leader are also proven mathematically. Simulation studies are performed to show the effectiveness of the proposed control design technique.

General traffic analysis based on Deep Packet Inspection (DPI) techniques at the gateways or access points cannot grasp the detailed knowledge of network applications going among internal nodes, and the statistics-based reports of routers are also lack of flow-level recognition of the traffic in the form of only five tuple. Therefore, network-wise accurate flow-awareness by packet sampling is highly desired for fine-grained quality of service guarantee, internal network management, traffic engineering, and security analysis and so on. In this paper, we propose a Spatio-Temporal Collaborative Sampling (STCS) problem based on the Software-Defined Networking (SDN) technique. The goal of STCS is to maximize the network-wise sampling accuracy of both elephant and mice flows, which considers both of the comprehensive influences of nodes and the effect on sampling accuracy imposed by the collaborative strategy among nodes in the time dimension. We present a approach to calculate the near optimal solution of STCS in two steps: 1) Top-K nodes selection by iterative comprehensive influence, and 2) spatio-temporal cosampling solution based on the local value maximization strategy. We evaluate the proposed approach by a realistic large-scale topology, and the results show that the sampling accuracy can be effectively improved by the method, especially for mice flows, and the redundant ratio of sampled packets is reduced by 34.4%.

This paper considers the cooperative output regulation problem of linear multi-agent systems with a switched exosystem. The switched exosystem can generate reference signals with different types over time which are more useful in practice. In contrast to the existing state feedback control law, in this paper, an output feedback control law is proposed.

In this paper, a novel output feedback solution based on the Q-learning algorithm using the measured data is proposed for the linear quadratic tracking (LQT) problem of unknown discrete-time systems. To tackle this technical issue, an augmented system composed of the original controlled system and the linear command generator is first constructed. Then, by using the past input, output, and reference trajectory data of the augmented system, the output feedback Q-learning scheme is able to learn the optimal tracking controller online without requiring any knowledge of the augmented system dynamics. Learning algorithms including both policy iteration (PI) and value iteration (VI) algorithms are developed to converge to the optimal solution. Finally, simulation results are provided to verify the effectiveness of the proposed scheme.

With the rapid development of various real-time services, there is urgent need for application-aware capabilities in realtime network to meet the higher demand for network's quality of service (QoS), security policy and so on. Accurate and cost-effective collection of flow information is needed. However, the contradiction between the real-time, accuracy and performance cost in the passive traffic information collection mode of old board makes it difficult to build application-aware realtime network of high-accuracy. In this paper, we propose an agile application-awareness network (A3N) for software-defined networks (SDN). A3N implements a flow-based self-adaptive sampling strategy (FSS) for the incoming traffic and combines a parallel deep packet inspection (DPI) with high-performance to perceive network traffic changes. Experimental results demonstrate the proposed A3N can gain good performance with low costs, and also provide real-time application-aware services with deep operational visibility.

For the battery energy storage system (BESS) consisting of multiple battery packages, package-level state-of-charge (SOC) balancing can provide safety redundancy in protecting battery packages from overcharging or overdischarging, and maintain the maximum power capacity of the overall BESS. In this paper, a distributed control scheme is proposed for package-level SOC balancing of a grid-connected BESS. The proposed control scheme comprises three parts. First, each battery package shares its SOC with its neighboring battery packages over the communication network. Based on the SOCs of its own and its neighbors, each battery package is equipped with an energy coordinator. Second, for each battery package, a distributed observer is employed to estimate the average desired power output, which is used to determine the local reference power output together with the energy coordinator. Third, local current controller is synthesized for each battery package to achieve local reference power output tracking. Comprehensive case studies are conducted to evaluate the effectiveness of the proposed control scheme.