Combining real-time of kalman filter or extended with robustness particle filter, proposes a variable method for target movement linear motion and non-linear motion. The recommended density function is generated by when the do motion; makes latest observation information to integrate into in sequential importance sampling, increase reintegration likelihood prior distribution, improve accuracy algorithm. results simulation show that can effectively reduce tracking error without sacrificing...

This paper proposes a novel point cloud multi-scale completion algorithm guided by image rotation attention mechanisms to address the challenges of severe information loss and suboptimal fusion effects in multi-modal feature extraction integration during shape completion. The proposed network employs an encoder–decoder structure, integrating Rotating Channel Attention (RCA) module for enhanced method clouds improve both local global information. also utilizes multi-level self-attention...

Loop closure detection is an important component of visual simultaneous localization and mapping (SLAM). However, most existing loop methods are vulnerable to complex environments use limited information from images. As higher-level image multi-information fusion can improve the robustness place recognition, a semantic–visual–geometric information-based algorithm (SVG-Loop) proposed in this paper. In detail, reduce interference dynamic features, semantic bag-of-words model was firstly...

Hyperspectral compressed sensing (HCS) is a new imaging method that effectively reduces the power consumption of data acquisition. In this article, we present novel HCS algorithm by incorporating spatial-spectral hybrid sensing, followed reconstruction based on spectral unmixing. At sampling stage, measurements are acquired compressive scheme to preserve necessary information for following unmixing, where spatial mainly retains endmember information, and abundance information. Due...

A major challenge for underwater acoustic target classification relates to significant performance decrease in complex environment. Recent researches have shown that the auditory feature extracted from Gammatone filter has remarkable ability on robust speaker identification. If this can be simulated, accuracy of will improved significantly noisy Aiming at purpose, a novel implementation filter-based is proposed and applied paper. Support Vector Machine (SVM) used as classifier our...

Hyperspectral compressed sensing (HCS) based on spectral unmixing technique has shown great reconstruction performance. In particular, the linear mixed model (LMM) been widely used in HCS reconstruction. However, due to complexity of environmental conditions, instrumental configurations, and material nonlinear mixing effects, LMM cannot accurately represent hyperspectral images, which limits improvement quality. this article, first, by introducing variability, mixing, residuals, a multitype...

The performance of underwater acoustic target recognition system decreases significantly mainly due to the complex conditions. Nevertheless, human auditory is capable segregating different targets in noisy Several auditory-based features have been applied recognition, but only represent amplitude signal. However, phase represented by instantaneous frequency (IF) might also contain some information distinguish targets. In this paper, a novel feature extraction method based on Gammatone...

In order to enhance the performance of existing beamformer-based direction-of-arrival (DOA) estimation methods, a novel DOA method based on sparse and low-rank decomposition sample covariance matrix is proposed. First, divided into desired signal interference as well that noise. The first component shown have property meanwhile, second can be approximately regarded matrix. Second, convex optimization problem decompose constructed, which efficiently solved using interior point or augmented...

Nowadays, much attention in electronic warfare is paid to the Specific Emitter Identification (SEI), which seems be one of most difficult subjects field. The principal function SEI recognizing different radar devices same type. In cases, classical methods identification could not enough realize task. main unique characteristics a modern emitter are hidden structures time-domain waveforms, especially pulse envelope and instantaneous frequency, used as fingerprint features this paper. order...

An efficient lossy compression algorithm for multispectral images based on block Karhunen–Loève transform (KLT) is proposed. First, a two-dimensional discrete wavelet performed each band of to remove the spatial correlation. Subsequently, partitioned into non-overlapping blocks same size, and coefficients in domain are treated as single object. Blocks that co-located spectral orientation affected by an adaptive their Finally, embedded coding with optimized truncation all principal components...

This paper focuses on the problem of lossy compression for hyperspectral images and presents an efficient algorithm based distributed source coding. The proposed employs a block-based quantizer followed by lossless coding, which is implemented through use multilevel coset codes. First, bitrate allocation to assign rational each block. Subsequently, multilinear regression model employed construct side information block, optimal quantization step size block obtained under assigned while...

To address wind power curtailment by the mechanism of "ordering heat", integrated community energy system (ICES) is firstly established based on a major breakthrough in large-scale centralized supply; then, day-ahead scheduling optimization model for ICES built quantum-behaved particle swarm algorithm, and rigid coupling between heating network grid improved adjusting CHP output; finally, influences storage devices interactions output are analyzed typical case, accommodation capability...

Existing 3D single object trackers (SOTs) of a point cloud all apply downscaling when extracting features from points. This operation leads to loss spatial and structural information, degrading tracking performance sparsely distributed small-scale objects. To address this problem, structure aware SOT is proposed. Specifically, the backbone network combined with auxiliary learn point-wise representations. During training stage, subsidiary used perform additional tasks supervisions, which...

The direction-of-arrival (DOA) estimation for the long-distance underwater acoustic sources usually suffers from low levels of signal-to-noise ratio (SNR) and insufficient snapshot data. Therefore, a novel signal self-nulling based DOA algorithm is proposed. First, assume that angular sector, where desired located, known. Then, beampattern minimum variance distortionless response beamformer in this sector circularly mapped onto unit circle. Finally, pseudo spatial spectrum constructed by...

The vector hydrophone can collect the acoustic pressure and particle velocity information of same point in sound field synchronously independently. It has four channels, which output projection components along x, y z axes Cartesian coordinate system observed respectively. a coherent source are completely correlated far-field region, there is no phase difference between them. However, on practical measurement occasions, random exists signals different channels. Three sources analyzed....

Robust adaptive beamforming (RAB) has became a popular research topic, with various RAB techniques being proposed in the past decades. However, because sample covariance matrix rather than interference-plus-noise co-variance is used to calculate weight vector, performance of previously developed not optimal. In this paper, novel algorithm, which uses reconstructed matrix, developed. First, and desired signal are by matched spectrum processing. Then, vector directly obtained using...

The acoustic vector sensor (AVS) array is a powerful tool for underwater target’s direction-of-arrival (DOA) estimation without any bearing ambiguities. However, traditional DOA algorithms generally suffer from low signal-to-noise ratio (SNR) as well snapshot deficiency. By exploiting the properties of minimum variance distortionless response (MVDR) beamformer, new method basing on matched steering searching proposed in this article. Firstly, attain rough estimate desired using algorithms....

Compressed sensing is one of the key technologies to reduce volume hyperspectral image for real-time storage and transmission. Reconstruction based on spectral unmixing show tremendous potential in compressed compared with other conventional algorithms that directly reconstruct images. In this article, a joint spatial-spectral scheme proposed. scheme, data are collected by hybrid sampling. As reconstruction, an objective function developed introducing fidelity constraints spatial...

To deal with the requirement of high-performance controller single leg foot force for improving flexibility complex environment quadruped robot, inaccuracies dynamical model and control changing through load changes, analysis hydraulic system model, actuated is built controlling. Combined characteristic control, a feed-forward paralleled Active Disturbance Rejection Controller (ADRC) foot-end-force support designed, which have not only ADRC but also fast responding as feed-forward. Proved by...

Specific Emitter Identification (SEI) is one of most difficult tasks in modern electronic warfare. The principle function the SEI utilizing fingerprint features to identify different emitters same type by a precise measurement, which plays an important role recognition process. Under premise that time-domain waveforms such as instantaneous amplitude and frequency are specified features, frechet distance can achieve preferable identification accuracy, compared Hausdorff Dynamic Time Warping...

This paper introduces a modified estimation of signal parameters via rotational invariance techniques (ESPRIT) algorithm for multisource joint direction arrival (DOA) and frequency using single vector hydrophone. Based on TLS-ESPRIT, the proposed may jointly estimate two-dimensional DOAs frequencies up to four uncorrelated narrow-band sources under most circumstances, can pairs all automatically. Compared with previous ESPRIT-based algorithms, adopts formulas computing directional cosines,...

Due to the instability of real underwater environment, auditory processing methods are proved be effective measures for acoustic target recognition. As MFCC and GFCC might show inevitable failure with noise disturbance, in this paper we apply a recent novel feature extraction called Power Normalized Cepstral Coefficients (PNCC) on targets Additionally, propose an enhanced PNCC by adding Subband mean IF feature. Subsequent experiments that has stronger performance terms anti-noise...

With the rapid development of electric vehicles and microgrid technology at home abroad, connection between them has become increasingly close, so studying charging discharging strategies is great significance to optimal operation microgrids. A energy scheduling method considering vehicle proposed in this paper. First, based on driving characteristics strategy implemented by vehicles, load model under both disordered ordered established. Then, operating costs pollution control costs, a...