Shanlin Sun

Infrared Small Target Detection (ISTD) is crucial for applications such as traffic monitoring and maritime surveillance. However, it remains highly challenging due to weak target signals and the absence of rich texture information, often resulting in low detection accuracy. Existing deep learning-based ISTD methods typically struggle to balance the trade-off between modeling long-range dependencies and avoiding feature oversmoothing. To this end, we propose RSFNet, a retention-based network with spatial–frequency joint enhancement for ISTD. RSFNet introduces a bidirectional 2D decay–retention attention mechanism into the vision Transformer (ViT) framework, which effectively suppresses background noise while capturing long-range dependencies. In addition, we design a Spatial–Frequency Joint Enhancement Module (SFE) to facilitate the transfer of salient target features from the encoder to the decoder. SFE integrates spatial and frequency domain features to facilitate global–local information interaction. Extensive experiments conducted on multiple publicly available ISTD datasets demonstrate that RSFNet significantly outperforms state-of-the-art (SOTA) methods in both detection accuracy and training efficiency, with a nearly 32% reduction in training time.

This paper investigates the output consensus problem in heterogeneous multi-agent systems. To address the challenges of traditional analytical methods in handling unknown dynamics and disturbances, a control framework is proposed that combines known model structures with a data-driven adaptive mechanism. The framework uses a distributed internal model to compensate for system heterogeneity and incorporates an event-triggered mechanism to reduce communication burden. To improve transient tracking performance, a reinforcement learning strategy based on centralized training and decentralized execution is introduced to adaptively optimize local feedback gains. Simulation results show that the proposed method effectively bounds closed-loop signals, achieves relatively fast convergence, and demonstrates some robustness and communication efficiency under process noise.

To address the kinematic constraints and real-time requirements of autonomous underwater vehicle (AUV), this paper proposes a novel path planning algorithm: Directional Cone and Goal-Biased Dynamic Artificial Potential Field RRT* (DCGB-DAPF-RRT*). The algorithm integrates four key techniques-Directional Cone sampling, Goal-Biased sampling, Adaptive Step Length, and a Dynamic Artificial Potential Field-to improve sampling efficiency and path quality over conventional RRT-based methods. In addition, redundant node pruning and cubic non-uniform B-spline interpolation are employed to enhance trajectory smoothness and ensure kinematic feasibility. Experimental and simulation results demonstrate that the proposed algorithm reduces path planning time by 50.0–73.6%, decreases the number of nodes by 71.0–77.8%, and lowers the number of iterations by 61.0–85.0%, while achieving a 100% success rate in complex environments. The final path length is reduced to 1766.1 m, and the maximum turning angle is limited to 11.35°, fully satisfying the motion constraints of AUV.

An increasing number of scholars are researching underwater acoustic sensor networks (UASNs), including the physical layer, the protocols of the routing layer, the MAC layer, and the cross-layer. In UASNs, the ultimate goal is to transmit data from the seabed to the surface, and a well-performed routing protocol can effectively achieve this goal. However, the nodes in the network are prone to drift, and the topology is easily changed because of the movement caused by ocean currents, resulting in a routing void. The data cannot be effectively aggregated to the sink terminal on the surface. Thus, it is extremely important to determine how to find an alternative node as a relay node after node drift and how to rebuild a reliable transmission path. Although many relay routing protocols have been proposed to avoid routing voids, few of them consider the relay node selection between the outage probability and the ocean current model. Therefore, we propose an ocean current motion model based routing (OCMR) protocol to avoid the routing void in UASNs. We predicted underwater node movement based on the ocean current motion model and designed a protection radius to construct a limited search coverage based on the optimal outage probability; then, the node with the best fitness value within the protection radius was selected as the alternative relay node using an improved WOA. In OCMR, the problem of the routing void caused by ocean current motion is effectively suppressed. The simulation results show that, compared with VBF, HH-VBF, and QELAR, the proposed OCMR platform performs well in terms of the PDR (packet delivery ratio), average end-to-end delay, and average energy consumption.

Guilin is situated in the southern part of China with abundant rainfall. There are 137 reservoirs, which are widely used for irrigation, flood control, water supply and power generation. However, there has been a lack of systematic and full-coverage remote sensing monitoring of reservoir dams for a long time. According to the latest public literature, high-resolution unmanned aerial vehicle (UAV) remote sensing has not been used to detect changes on the reservoir dams of Guilin. In this paper, an intelligent segmentation change detection method is proposed to complete the detection of dam change based on multitemporal high-resolution UAV remote sensing data. Firstly, an enhanced GrabCut that fuses the linear spectral clustering (LSC) superpixel mapping matrix and the Sobel edge operator is proposed to extract the features of reservoir dams. The edge operator is introduced into GrabCut to redefine the new energy function’s smooth item, which makes the segmentation results of enhanced GrabCut more robust and accurate. Then, through image registration, the multitemporal dam extraction results are unified to the same coordinate system to complete the difference operation, and finally the dam change results are obtained. The experimental results of two representative reservoir dams in Guilin show that the proposed method can achieve a very high accuracy of change detection, which is an important reference for related research.

Safety during physical human–robot interaction is the most basic requirement for robots. Collision detection without additional sensors is an economically feasible way to ensure it. In contrast, current collision detection approaches have an unavoidable trade-off between sensitivity to collisions, signal smoothness, and immunity to measurement noise. In this paper, we present a novel sliding mode momentum observer (NSOMO) for detecting collisions between robots and humans, including dynamic and quasistatic collisions. The collision detection method starts with a dynamic model of the robot and derives a generalized momentum-based state equation. Then a new reaching law is devised, based on which NSOMO is constructed by fusing momentum, achieving higher bandwidth and noise immunity of observation. Finally, a time-varying dynamic threshold (TVDT) model is designed to distinguish between collision signals and the estimated lumped disturbance. Its coefficients are obtained through offline data recognition. The TVDT with NSOMO enables fast and reliable collision detection and allows collision position assessment. Simulation experiments and hardware tests of the 7-DOF collaborative robot are implemented to illustrate this proposed method’s effectiveness.

Black carbon (BC) aerosols have a considerable impact on humans because they not only cause environmental pollution and reduce visibility but also harm human health. During the heating season in northern China, a large amount of coal is burned for heating, producing a large amount of BC. There are few studies on BC properties during the heating season. In this paper, BC is measured optically, so it is referred to as equivalent black carbon (EBC). This paper investigated EBC properties in depth during the heating and nonheating seasons of a typical urban environment in China with two years of EBC measurements. The results show that: (1) EBC aerosol concentrations during the heating season were significantly higher than those during the nonheating season. (2) The main sources of EBC aerosols throughout the year are liquid sources. During the heating season, solid sources (coal and biomass combustion) are dominant. (3) The proportion of brown carbon (BrC) produced by biomass energy during the heating season is greater than that during the nonheating season. (4) The resulting backward trajectory indicates that a large portion of the high EBC aerosol concentration sources originate from northern and northwestern China. Our results reveal that the characteristics and sources of EBC in the urban environment of northern China vary widely, suggesting that different measures should be taken to reduce BC aerosol concentrations during heating and nonheating seasons.

Since mid-March 2020, global COVID-19 pandemic has experienced an exponential growth in process from sporadic to sudden outbreaks. This paper selects the 8-day surge data of daily cases, death and recovery rates (March 19-26, 2020) from 18 countries with severe pandemic situation to discuss the impact of 9 factors of both socioeconomic and natural on the pathogen outbreak. Moreover, the paper also elaborates analysis and comparison of relatively slow 4-week (February 1-29, 2020) data of China’s surge cases to determine the relationship between social and natural factors and on the spread of pandemic, which provides an effective reference for delaying and controlling the pandemic development.