Compared with generic object detection， camouflaged object detection （COD） is more difficult due to its high similarity to the background. Most COD methods used convolutional neural network （CNN） in the past. In contrast， we note the great potential of Transformer in vision tasks， so we use Swin Transformer as the backbone to generate more robust multi-level features. Considering the complex characteristics between the camouflaged object and the background， we propose an edge-aware module （EAM） to generate a fine edge prior. In addition， the feature aggregation module （FAM） is used to fuse multi-level features， and the internal connections between edge points are investigated by feeding edge feature pixels into a graph convolutional network （GCN） to purify uncertain edge points， then guiding multi-level feature fusion operation. We reduce indistinguishable noise by cascading several FAMs and get prediction maps with sharp edges. Experimental results show that the proposed model can run at the real-time speed of 46 frames per second （FPS） on a single NVIDIA 2080Ti GPU. Comparing with 12 state-of-the-art COD methods on four public datasets， the proposed method outperforms other approaches.
In on-orbit servicing missions， the servicing spacecraft typically docks with the target spacecraft to form a combined spacecraft to carry out subsequent tasks， and the cooperative attitude control between the two spacecraft is one of the challenging techniques. To address the issues of inertia uncertainty， communication delays， and external disturbances in combined spacecraft attitude control， a prescribed-time attitude cooperative control method based on disturbance observer is proposed. Firstly， a distributed control model for combined spacecraft is established， and a prescribed performance function with directly adjustable convergence time is designed to characterize both the transient and steady-state behaviors of the system. Secondly， integrating the backstepping method， an attitude controller based on prescribed-time performance is developed to achieve fast and accurate attitude maneuvering of the combined spacecraft. Thirdly， a disturbance observer is introduced to estimate the composite disturbance caused by inertia uncertainty and external perturbations. Meanwhile， based on consensus theory， an attitude cooperative controller is designed to compensate for the control torque mismatch caused by communication delays. The system’s stability is then proved using Lyapunov method. Finally， numerical simulations demonstrate that the proposed controller achieves accurate cooperative attitude control of combined spacecraft within the prescribed time under the above disturbances.
The performance of a particle detection system is primarily determined by its first stage， which mainly includes a particle radiation detector and a charge-sensitive preamplifier （CSA）. The connection between the detector and the front-end readout circuit is typically AC-coupled， which usually has a coupling capacitor in the range of 0.01—0.1 μF to isolate the input of the CSA from the detector. The application-specific integrated circuit （ASIC） readout electronics for nuclear applications 3（RENA-3） is a multi-channel mixed-signal integrated circuit （IC） offering high energy resolution and high time resolution， developed for the readout of particle radiation detectors. RENA-3 is employed as the front-end readout circuit for an interplanetary energetic particle telescope designed to detect 20─1 000 keV electrons， 25─12 000 keV protons， and ions like alpha particles. However， the maximum input charge of RENA-3 is 54 fC， which limits its maximum detection range to about 1 200 keV and is insufficient for the proton channels. While other ASICs could be adopted， available alternatives do not simultaneously meet the requirements for both high precision and wide measurement range. This paper proposes the use of a small coupling capacitance to extend the range for proton detection by attenuating the charge signal before it reaches the CSA input. The coupling capacitance is on the order of picofarads， much smaller than the detector junction capacitance. In this configuration， stray capacitance and coupling capacitance become significant. A circuit theoretical model is proposed to calculate the transfer function from input charge to output voltage， analyzing the influence of coupling capacitance and stray capacitance on the particle radiation detection system. Experiments using discrete components with known parameters are conducted to validate the theoretical model. Finally， the model is applied to the RENA-3 system for optimization to meet the instrument’s specifications， and certain RENA-3 parameters are derived.
Low-altitude traffic management requires timely inference of both unmanned aerial vehicle （UAV） platform type and likely mission intent. Declared flight plans can provide intent information for cooperative operations， but non-cooperative， abnormal or partially observed UAVs require inference from observable evidence. The existing surveillance methods mainly detect or track UAVs， classify platform type， or predict trajectories. They therefore remain limited when different missions share similar short-term motion and appearance alone lacks temporal context. This study introduces MCFNet， a multimodal framework that combines an external-view RGB image with an 8 s trajectory history for mission-level UAV intent recognition. Its bidirectional cross-attention fusion links local visual regions with trajectory time steps before classification， allowing cues such as payload state or spray effects to be interpreted with the relevant maneuver segments. Modality dropout reduces single-modality shortcuts， and a consistency loss penalizes invalid type-intent pairs. On 7 290 simulated samples covering three UAV types and 15 intents， MCFNet achieves 89.63%±0.99% intent accuracy， 89.41%±1.00% type‑intent joint accuracy and 91.33%±1.05% macro‑F1， outperforming the strongest late-fusion baseline by 3.04% and achieving a slightly higher mean intent accuracy than the multimodal bottleneck Transformer （MBT） token-level fusion baseline. The results provide an observation-based route to earlier warning and differentiated decision support for low-altitude traffic management.
With the rapid development of aviation industry， the density of terminal-airspace traffic has increased significantly， resulting in higher potential conflict among aircraft. To address this issue， this study proposes a potential conflict identification method for terminal airspace based on three-dimensional （3D） trajectory tube structures. The approach first clusters arrival trajectories to identify major traffic flows and constructs corresponding 3D trajectory tubes. It then introduces traffic-flow parameters into a probabilistic model， which is combined with three-dimensional airspace gridding to identify and locate potential conflict regions. The results demonstrate that the proposed method can effectively detect potential conflict areas within a 15 min time window and analyze the interactions among traffic flows within these regions. This research provides a valuable framework for improving the safety and efficiency of terminal airspace operations， offering both theoretical and practical significance for airspace optimization and decision-support enhancement.
