【作者】 刘亚雷；

【导师】 顾晓辉；

【作者基本信息】 南京理工大学 ， 火炮、自动武器及弹药工程， 2013， 博士

【摘要】 随着战争环境的日益复杂,传统的探测系统(如雷达、红外等)受到越来越多的威胁,被动声跟踪技术以其隐蔽性强、不易受到攻击、适用性广等优点越来越受到人们的重视。本文以智能反坦克子弹药(Brainpower Antitank Submunition,简称BAT)为应用背景,开展了三维运动声阵列对典型二维声目标跟踪理论研究,全文研究的主要内容包括：三维运动声阵列跟踪系统动态模型研究、跟踪系统最佳观测布局研究、观测信号预处理技术研究、跟踪滤波算法研究以及三维运动声阵列对双点声源角跟踪指向性能研究,具体归纳如下：(1)三维运动声阵列跟踪系统动态模型研究。阐述了战场典型二维声目标声信号产生机理及特性,分析了二维声目标的声源特性,探讨了声信号在大气中的反射、折射、透射、散射、声信号的衰减以及声信号传播的多普勒效应,认清了三维运动声阵列跟踪环境的物理现象,得到了声信号以空气为介质的传播模型,结合学位论文研究的典型二维目标特性,给出了三维运动声阵列跟踪系统动态模型的基本假设。在笛卡尔坐标系及修正极坐标系下,分别建立了运动声阵列跟踪系统的状态模型及观测模型,对模型参数进行了分析,设计了一种包含数字压力传感器电路等硬件的弹载高度测量与记录装置。(2)三维运动声阵列跟踪测量系统最佳布局研究。对由平面四元声阵列组成的跟踪测量系统的阵元布局进行了研究,提出了一种度量四元三维运动声阵列跟踪观测系统测量精度的指标准则,即PDOPF (Position Dilution of Precision Function).以二维目标的位置几何精度衰减因子函数最优为目标,对平面四元声阵列跟踪测量系统布局的位置坐标进行了解算,分析了布局精度,得到了三维运动声阵列跟踪测量系统的理论最佳布局及工程最佳布局；通过静态半实物仿真试验,进行了验证。(3)三维运动声阵列观测信号预处理技术研究。对战场环境下的干扰信号进行了分析,在阵列多传感器观测信号预处理方法中,提出了正交小波多尺度观测信号预处理算法,并通过“静态”及“动态”半实物仿真试验进行了验证研究；而在单通道观测信号预处理方法中,基于EMD理论,分析了IMF频谱特性,结合本文研究的典型声目标声信号特性,对观测信号进行了预处理,通过对单通道信号分析验证了该算法的有效性。此外,提出了一种针对信号几何窗口的变量——“当前”平均改变能量(Current Average Change Energy, CACE),利用该变量推导了基于“当前”平均改变能量的机动检测算法,将“当前”机动改变能量调制到CACE上,得到了“当前”平均改变能量机动准则。最后设计了一种基于Matlab的声信号预处理软件。(4)三维运动声阵列跟踪滤波算法研究。根据运动声阵列跟踪系统的动态模型,分别从高斯线性、高斯非线性、非高斯非线性三个方面研究了三维运动声阵列对二维声目标的跟踪滤波算法。(A)基于线性、高斯系统假设下的跟踪滤波算法。阐述了传统的线性系统滤波状态估计算法,即Kalman滤波算法,基于Kalman滤波算法提出了多尺度贯序式Kalman滤波的运动声阵列跟踪算法(Multi-scale Sequential based on Kalman Filter, MSBKF),Matlab仿真分析了该算法的跟踪性能,针对跟踪滤波与预测实时性问题,提出了运动阵列的基于当前平均改变能量的机动检测与变维自适应Kalman滤波算法(Current Average Change Energy Maneuvering Detection and Variable Dimension Adaptive Kalman Filtering, CACEMD-VDAKF),通过算法仿真,验证了CACEMD-VDAKF算法的有效性。(B)基于非线性、高斯系统假设下的跟踪滤波算法。阐述了传统的非线性系统滤波算法,即扩展Kalman滤波(EKF),分析了EKF滤波的偏差,提出了基于无迹粒子滤波的自适应交互多模型运动声阵列跟踪算法(Adaptive Interacting Multiple Model Unscented Particle Filter based on measured residual, AIMMUPF-MR),通过算法仿真,验证了AIMMUPF-MR算法在跟踪精度、稳定性及实时性上的有效性。(C)基于非线性、非高斯系统假设下的跟踪滤波算法。针对非线性、非高斯跟踪系统的状态滤波与预测问题,基于粒子滤波提出了确定性核粒子群的粒子滤波跟踪算法(Deterministic Core Particle Swarm Particle filter, DCPS-PF),推导了该算法的理论误差性能下界(Cramer Rao Low Bound, CRLB),与粒子滤波算法相比,仿真结果表明了该算法的有效性和优越性。(5)三维运动声阵列对双点声源角跟踪指向性能研究。阐述了多点声源干扰的基本原理,建立运动声阵列在双点声源下的角度跟踪指向性能数学模型。从三个方面即：伪目标与真实目标的声信号在频率上一致,声压幅值保持线性关系、声压幅值一致,频率保持线性关系、声信号的声压幅值和频率均保持线性关系,分析了频率值比、声压幅值比及两声源的相位差与运动声阵列角度跟踪指向性能之间的关系。提出了包含运动声阵列的飞行速度、侧向过载,战斗部有效毁伤半径,运动声阵列的弹道倾角及两点声源对声阵列张角等参数的角度干扰指数(Bearing-only Deflection Index, BODI)作为运动声阵列角度跟踪指向性能评价指标,为进一步研究三维运动声阵列对多声源目标跟踪理论奠定了基础。更多还原

【Abstract】 Conventional detection systems, such as radar and sonar, have encountered more and more threats with the increasing complexity of circumstances in modem battle field. Passive acoustic tracking teleology has been paid more and more attention because of its significant advantage in self-hiding, the capability to avoid attack and extensive applicability. In this paper, the tracking theory of3D dynamic acoustic array for acoustic target has been researched based on the application background of brainpower antitank submunition. The main research contents contain the dynamic model of3D dynamic acoustic array tracking system, optimal arrangement of the four-sensor dynamic acoustic array, the observation signal preprocessing teleology, tracking filter algorithms and the bearing-only tracking pointing performance of3D dynamic acoustic array based on dual acoustic sources. More specifically, the main research contents are as follows(1) The3D dynamic acoustic array tracking system dynamic model. The mechanism and characteristic of acoustic signal were elaborated, after that, the signal characteristic, such as signal reflection, signal refraction, signal scattering, signal attenuation, and signal Doppler Effect, have been analyzed, then the physical phenomena of3D dynamic acoustic array tracking system also is clearly. Combine the dissertation; the basic assumptions of3D dynamic acoustic array tracking system are obtained, so the state model and observation model are established based on the Cartesian coordinate system and correction polar coordinate, respectively. Then the model parameters are analyzed, and the missile-borne height measurement and record device is designed based on digital pressure sensor.(2) Optimal arrangement of the3D dynamic acoustic array. The arrangement of tracking observing system composed of four sensors has been studied, and the indicator guideline which evaluates observation precision of dynamic acoustic array tracking measurement system has been proposed, such as the optimal position dilution of precision function (PDOPF). After that, the location coordinates of acoustic array are solved based on the optimal position dilution of precision function of2D target, so the theory optimal arrangement of four-sensors dynamic acoustic array tracking measurement system is obtained, and the static experimental study on sound-source bearing estimation is designed. The theoretical results are compared with the experimental results of the present study.(3) The signal preprocessing technology of3D dynamic acoustic array. The interference signal has been analyzed at battlefield environment, the orthogonal multi-scale wavelet signal preprocessing algorithm is proposed, and the semi-physical simulation results confirm the effectiveness and stability. Based on the EMD theory, the spectral characteristics of IMF also been analyzed, and the single channel signal preprocessing algorithm is proposed. In addition, the current average change energy (CACE) is proposed for geometric window, then an algorithm of maneuvering detection based on CACE is deduced, and the current average change energy is modulated to CACE, so the current average change energy maneuvering detection criterion is deduced. Finally, the signal preprocessing software based on Matalb is also designed.(4) The tracking filter algorithm of3D dynamic acoustic array. The tracking filter algorithm of3D dynamic acoustic array to2D acoustic target have been studied based on dynamic acoustic array tracking system dynamic model from Gaussian linear, Gaussian nonlinear, and non-Gaussian nonlinear.(A) The tracking filter algorithm based on Gaussian linear. The traditional linear filter algorithm is elaborated, such as Kalman filter. The Multi-scale Sequential Kalman Filter algorithm is obtained based kalman fitler recursive theory; the matlab simulation results show the validity and superiority of the presented algorithm. In order to solve the problem of real-time for tracking and filter, the based on CACE maneuvering detection and variable dimension adaptive Kalman filtering algorithm (CACEMD-VDAKF) is obtained, and the matlab simulation results confirm the effectiveness and stability of CACEMD-VDAKF algorithm for two-dimensional target tracking.(B) The tracking filter algorithm based on Gaussian nonlinear. The traditional nonlinear filter algorithm is elaborated, such as Expand Kalman Filter (EKF), and the filter error of EKF has been analyzed, the adaptive interacting multiple models unscented particle filter algorithm based on measured residual is proposed. The Matlab simulation results between different algorithms show the validity and superiority of the presented algorithm in tracking accuracy, stability and real-time capability.(C) The tracking filter algorithm based on non-Gaussian nonlinear. The deterministic core particle swarm particle filter algorithm is proposed based on particle filter, and the Cramer Rao Low Bound(CRLB) is also be deduced, compared with the traditional particle filter algorithm, the Matlab simulation results show the validity and superiority of the presented algorithm.(5) The bearing-only tracking pointing performance of3D dynamic acoustic array based dual acoustic sources. The basic principle of multi-source interference is analyzed, and the mathematical model of dynamic acoustic array bearing-only tracking pointing performance is established at dual acoustic sources interference. The relationship of frequency, sound pressure and phase difference between real target and pseudo-target has been analyzed. Finally, the evaluation indicator of bearing-only tracking pointing performance in point acoustic source interference also is established based on the hypothesis of equal power at dual acoustic sources interference, which is as the theoretical basis for further research the multi-source tracking.更多还原