中国人民解放军海军航空大学

With the rapid development of the low-altitude economy,low-altitude targets such as unmanned aerial vehicles(UAVs)have been widely applied in fields like logistics,agricultural plant protection,and emergency rescue.These targets are characterized by low flight altitude,slow movement speed,and small size(commonly known as"low-slow-small"),which put extremely high demands on surveillance systems.As a core detection method,radar is in urgent need of enhancing its ability to accurately detect and classify low-altitude targets such as birds and UAVs in complex backgrounds.This paper makes a comparative analysis of major low-altitude detection means,systematically reviews the progress of radar detection and recognition technologies for birds and UAVs,and focuses on the applications of typical systems based on digital array pan-search radar,external radiation source radar,and meteorological radar.It expounds on the key role of radar low-altitude detection in scenarios such as airport bird strike prevention and UAV countermeasures,and discusses emerging technical directions like distributed radar and integrated communication-sensing detection.Finally,it looks ahead to future developments in waveform optimization design,intelligent information processing,and other aspects.

Objective With the rapid development of visual sensing and artificial intelligence technologies,computer vision is being increasingly applied in fields such as autonomous driving and robot navigation.Visual relative pose estimation can be broadly categorized into monocular and multi-camera systems based on the number of cameras employed.Compared to monocular configurations,multi-camera systems offer distinct advantages,including a wider field of view and richer environmental information through synchronized multi-perspective observation.These attributes render them particularly suitable for applications demanding high safety and robustness,such as autonomous driving,where they have become the predominant sensing configuration.However,existing relative pose estimation methods based on multi-camera setups typically rely on a large number of feature correspondences,resulting in high computational complexity and difficulty in meeting real-time requirements.Especially in the dominant scenario of planar vehicle motion,general methods fail to fully utilize motion constraints,leading to model redundancy and computational resource waste.This study aims to design an efficient minimal solver for multi-camera pose estimation that incorporates planar motion prior,specifically targeting autonomous driving applications on structured roads. Methods An efficient minimal solver that incorporates prior knowledge of planar motion is proposed.Built upon the generalized camera model,this method fixes the world coordinate system to a spatial point and introduces a depth-parameterized translation representation.In typical autonomous driving scenarios,vehicles predominantly operate on structured roads or near-planar surfaces.This motion characteristic provides a critical constraint for pose estimation:the six-degree-of-freedom rigid motion of the platform can be simplified to a planar motion model involving only yaw rotation about the vertical axis and translation constrained to the horizontal plane.Based on this observation,a planar motion model tailored for multi-camera systems is constructed.This formulation transforms the complex multi-camera pose estimation problem into a linear model that can be solved with only three sets of feature correspondences,significantly reducing reliance on feature quantity and computational complexity.The method is evaluated through comprehensive simulations and real-world experiments,covering performance under image noise,road slope perturbations,and vehicle vibrations. Results and Discussions The proposed method,named 3pt-Depth,is systematically compared with several baselines including 17pt-Li,6pt-Stew,and 2pt-Choi in terms of efficiency,stability,and robustness.In computational efficiency,3pt-Depth achieves an average runtime of 14.755 μs,which is over 3 times faster than 17pt-Li and significantly faster than 6pt-Stew(Tab.1).Under noise-free conditions,3pt-Depth exhibits competitive numerical stability,with translation error performance better than 17pt-Li and clearly superior to 6pt-Stew(Fig.3).Simulations are conducted using a binocular vision system with a baseline length of 1 meter by default,consistent with sensor configurations in the widely used KITTI dataset and representative of real-world automotive setups.To systematically evaluate method performance under various disturbances,we examine the impact of image noise,road slope,vehicle vibration,and calibration errors.All solvers are integrated within the RANSAC framework,and the estimate yielding the highest inlier count is used for final error evaluation.In image noise tests,3pt-Depth demonstrates the strongest tolerance to image noise across different motion patterns,with the slowest accuracy degradation as noise increases(Fig.4).It also outperforms 2pt-Choi under road slope disturbances(pitch and roll)and vehicle vibration scenarios(Fig.5-7).Further experiments show that 3pt-Depth maintains accuracy under increasing extrinsic rotation and translation calibration errors,and scales well with the number of cameras(Fig.8).Real-world validation on the KITTI dataset shows that 3pt-Depth achieves lowest rotation and translation errors across all sequences,confirming its practical effectiveness in autonomous driving environments(Tab.2-3). Conclusions This paper presents an efficient minimal solver for multi-camera pose estimation under planar motion constraints,specifically designed for autonomous driving.By leveraging the planar motion prior and a depth-parameterized translation representation,the method requires only three feature correspondences,significantly reducing computational complexity and dependency on feature matches while maintaining estimation accuracy.The method shows strong robustness against common disturbances in driving environments.Future work will focus on extending the method to adapt to non-ideal planar motions and exploring tight-coupling frameworks with other sensors like IMUs to enhance robustness in complex real-world scenarios.

Enhancing the signal waveform's power efficiency and information transmission efficiency is one of the issues in tracking,telemetry,and command(TT&C)system waveform design.It is found that the basic characteristics such as biorthogonal,optimal time-frequency energy aggregate,and spectral controllable of prolate spheroidal wave functions(PSWFs)make it be a reliable candidate for TT&C signal waveform design.In this paper,PSWF are introduced into TT&C signal waveform design,and a four-dimensional unified spread spectrum TT&C signal waveform design method based on PSWF waveform shaping is proposed.This new idea,by using high-energy aggregation PSWF as waveform shaping functions,can obtain higher power efficiency and information transmission rate,better anti-interference and anti-interception capabilities.It provides a better waveform scheme for the design of TT&C signal waveforms.For instance,the suggested approach can boost the spread spectrum gain by 3.44 dB and the information transmission rate by one when compared to conventional TT&C signals based on rectangular pulse functions under the same system error performance.

Deck operation scheduling is the key technology to improve the sortie rate of carrier-based aircraft.In view of the multi-stage and high coupling characteristics of amphibious assault ship carrier-borne helicopters operation,an integrated scheduling model and optimization algorithm were proposed,which included warehousing and deck transportation,maintenance service support,and sortie departure.First,an integrated scheduling model of carrier helicopter sortie operation was developed based on the resource and logical restrictions of each stage in order to reduce the duration of the sortie.Then,the problem is abstracted as a resource-constrained project scheduling problem with transition times,and a two-segment genetic-neighborhood search algorithm is designed to solve the problem based on a serial schedule generation scheme.Through simulation experiments conducted on the concentrated sortie cases of 12 shipboard helicopters,the effectiveness of the scheduling model and algorithm in optimizing the sortie process was verified.This significantly reduced the sortie preparation time for the helicopter group.Additionally,it was shown that the entire scheduling is more significantly impacted by decisions made during the maintenance service support phase.

With the exploration of the complex and wide-area marine environment,the effective identification of sea ice,enteromorpha,oil spill and other marine targets is of great significance for safeguarding maritime rights and interests.The classification of sea ice and other targets is very challenging due to their diverse manifestations,different physicochemical and optical properties,changes in sea surface light and sea water movement.In this regard,more information of the target could be obtained by means of the unique advantage of the combination of the atlas of airborne hyperspectral images.Three classical classification algorithms,spectral angle matching algorithm(SAM),maximum likelihood classification method(MLC)and support vector machine(SVM),were used to classify the sea ice,entera margin and oil spill targets based on the sea background.The qualitative and quantitative comparisons show that the classification results of SAM algorithm for enteromorpha are poor,the Kappa coefficient is 0.67,and that of MLC algorithm for sea ice and oil spill are poor,the boundary of classification area is fuzzy,while the SVM algorithm has good classification effect,tits overall accuracy,average accuracy and Kappa coefficient are above 0.9,which is relatively stable on the whole.