A novel multi-sensor multiple model particle filter with correlated noises for maneuvering target tracking

Aiming at the effective realization of particle filter for maneuvering target tracking in multi-sensor measurements,a novel multi-sensor multiple model particle filtering algorithm with correlated noises is proposed.Combined with the kinetic evolution equation of target state,a multi-sensor multiple model particle filter is firstly constructed,which is also used as the basic framework of a new algorithm.In the new algorithm,in order to weaken the adverse influence from random measurement noises in the measuring process of particle weight,a weight optimization strategy is introduced to improve the reliability and stability of particle weight.In addition,considering the correlated noise existing in the practical engineering,a decoupling method of correlated noise is given by the rearrangement and transformation of the state transition equation and measurement equation.Since the weight optimization strategy and noise decoupling method adopt respectively the center fusion structure and the off-line way,it improves the adverse effect effectively on computational complexity for increasing state dimension and sensor number.Finally,the theoretical analysis and experimental results show the feasibility and efficiency of the proposed algorithm.

A Survey on Joint Target Tracking and Classification

The JTC technology deals with the problem of target tracking and target classification simultaneously within a unified framework.The fundamental idea of the JTC technology is that by taking advantage of the mutual exchange of useful information between the tracker and classifier,significant improvements in performance of both target tracking and target classification can be expected.The principle of JTC technology is introduced.The existing JTC technologies are broadly categorized into two classes,i.e.,point-target-motion-model-based JTC and rigid-target-motion-based JTC,which are then compared in detail.The advance of the JTC technology is surveyed with comments on some related literatures.Finally,some opening topics of the JTC technology are discussed.

Algorithm for Multi-laser-target Tracking Based on Clustering Fusion

Multi-laser-target tracking is an important subject in the field of signal processing of laser warners. A clustering method is applied to the measurement of laser warner, and the space-time fusion for measurements in the same cluster is accomplished. Real-time tracking of multi-laser-target and real-time picking of multi-laser-signal are introduced using data fusion of the measurements. A prototype device of the algorithm is built up. The results of experiments show that the algorithm is very effective.

Tracking method based on separation and combination of the measurements for radar and IR fusion system

A new distributed fusion method of radar/infrared （IR） tracking system based on separation and combination of the measurements is proposed by analyzing the influence of rate measurement. The rate information separated from the radar measurements together with measurements of IR form a pseudo vector of IR, and the corresponding filter is designed. The results indicate that the method not only makes a great improvement to the local tracker＇s performance, but also improves the global tracking precision efficiently.

Surrounding Objects Detection and Tracking for Autonomous Driving Using LiDAR and Radar Fusion

Radar and LiDAR are two environmental sensors commonly used in autonomous vehicles,Lidars are accurate in determining objects’positions but significantly less accurate as Radars on measuring their velocities.However,Radars relative to Lidars are more accurate on measuring objects velocities but less accurate on determining their positions as they have a lower spatial resolution.In order to compensate for the low detection accuracy,incomplete target attributes and poor environmental adaptability of single sensors such as Radar and LiDAR,in this paper,an effective method for high-precision detection and tracking of surrounding targets of autonomous vehicles.By employing the Unscented Kalman Filter,Radar and LiDAR information is effectively fused to achieve high-precision detection of the position and speed information of targets around the autonomous vehicle.Finally,the real vehicle test under various driving environment scenarios is carried out.The experimental results show that the proposed sensor fusion method can effectively detect and track the vehicle peripheral targets with high accuracy.Compared with a single sensor,it has obvious advantages and can improve the intelligence level of autonomous cars.

MULTISENSOR TRACKING SYSTEM WITH ATTITUDE MEASUREMENTS

A new 3 D fusion tracking system for an anti air missile homing system based on radar and imaging sensor is developed. The attitude measurements from the imaging sensor are used to improve the tracking performance. Computer simulation results show that the tracking system greatly reduces the tracking errors compared with trackers without attitude measurements, and achieves small miss distances even when the target has a big maneuver.

THEORETICAL ANALYSIS OF IMPROVEMENT OF TRACK LOSS IN CLUTTER WITH MULTISENSOR DATA FUSION

The paper analyses the improvement of track loss in clutter with multisensor data fusion.By a determination of the transition probability density function for the fusion prediction error, one can study the mechanism of track loss analytically. With nearest-neighbor association algorithm. The paper we studies the fused tracking performance parameters, such as mean time to lose fused track and the cumulative probability of lost fused track versus the normalized clutter density, for track continuation and track initiation, respectively. A comparison of the results obtained with the case of a single sensor is presented. These results show that the fused tracks of multisensor reduce the possibility of track loss and improve the tracking performance. The analysis is of great importance for further understanding the action of data fusion.

Maneuvering Vehicle Tracking Based on Multi-sensor Fusion

Maneuvering targets tracking is a fundamental task in intelligent vehicle research. Thispaper focuses on the problem of fusion between radar and image sensors in targets tracking. Inorder to improve positioning accuracy and narrow down the image working area, a novel methodthat integrates radar filter with image intensity is proposed to establish an adaptive vision window.A weighted Hausdor? distance is introduced to define the functional relationship between image andmodel projection, and a modified simulated annealing algorithm is used to find optimum orientationparameter. Furthermore, the global state is estimated, which refers to the distributed data fusionalgorithm. Experiment results show that our method is accurate.

天空地一体化多目标跟踪算法研究综述

为实现全时全域“泛在连接”,构建天空地一体化网络已成为国家重大需求,而基于天空地一体化网络下跨域协同系统进行多目标跟踪是其中一个重要的发展方向,其在军民用领域都极具应用价值。本文详细阐述了天空地一体化网络背景下多目标跟踪方法研究进展。首先,介绍了天空地一体化跨域协同多目标跟踪的研究背景与意义。其次,从基于视觉的多目标跟踪、基于模型的多目标跟踪和基于多模态融合的多目标跟踪三个方面概述了当前的代表性研究方法:在基于视觉的多目标跟踪算法方面,介绍了单摄像头和多摄像头融合的多目标跟踪方法;对于基于模型的多目标跟踪,先介绍了单传感器多目标跟踪方法,以及在多种复杂场景下的改进,然后介绍了多传感器融合方法;在基于多模态信息融合的目标跟踪方面,在对多传感器时空配准方法和有代表性的多模态信息融合方法介绍的基础上,概括了基于多模态融合的多目标跟踪算法。最后探讨了当前存在的问题和未来发展方向:无论基于视觉的还是基于模型的多目标跟踪方法都有不少问题有待解决,特别是两种方法的结合值得深入研究;在面临复杂干扰时,基于多传感器信息融合的多目标跟踪由于能实现信息的互补,成为未来的主流发展方向;此外,跨域协同系统,由于能利用更多的资源和信息,其多目标跟踪问题研究极具价值,不过其中通信安全问题和多目标跟踪模型轻量化问题值得探讨。本文对从事目标跟踪及空天地一体化协同控制相关理论与技术研究的科研工作者具有重要参考价值。

面向目标跟踪的多传感器数据融合研究综述

随着现代目标跟踪场景的日益复杂化,单一传感器探测系统已无法满足目标跟踪的实际需求,多传感器数据融合技术成为复杂场景下目标跟踪问题的重要解决方案。本文综述了多传感器数据融合中的数据关联和估计融合这两项关键过程的基本原理与研究现状,整理了神经网络、强化学习等人工智能技术在多传感器数据融合领域应用的相关研究,并对多传感器数据融合方法在目标跟踪领域的发展进行了展望。

ADAS系统视觉与毫米波雷达分布式抗差卡尔曼滤波融合算法

自动驾驶车辆常常利用多传感器对周围目标进行检测和跟踪,但受限于传感器多源异构特性和复杂多变的驾驶环境,准确的多目标检测和跟踪仍是实现自动驾驶的一大困难和挑战。本文面向高级驾驶辅助系统(ADAS)的多目标检测与跟踪任务,采用了基于1个视觉传感器和5个毫米波雷达(1V5R)的传感器配置方案,且设计了基于分布式抗差卡尔曼滤波器的多传感器信息融合算法以实现对周围目标的准确感知。首先,针对不同传感器数据特征,采用不同的线性卡尔曼滤波器和扩展卡尔曼滤波器进行数据融合,并基于分布式卡尔曼滤波建立了1V5R多传感器信息融合框架。其次,为降低传感器动态误差对于融合精度的影响,在卡尔曼加权观测融合的基础上,引入抗差估计方法,实现了对传感器动态误差的实时估计和修正。最后,通过离线仿真和实车道路试验对所提出的基于分布式抗差卡尔曼滤波的多传感器融合算法进行了验证。试验结果表明,与单一传感器的测量值相比,所提出的算法能有效融合多个传感器的信息以提升目标的检测与跟踪精度,且鲁棒性较好。

单站水下方位频率机动目标运动分析方法

水下方位频率机动目标运动分析常需已知中心频率先验信息且存在机动检测错误等,因此提出了一种基于方位与多线谱信息融合的自适应无迹卡尔曼滤波算法。该算法将方位与多线谱频率信息融合处理,重新构建了方位−多线谱目标运动分析模型。将中心频率信息引入状态向量中构建实时迭代估计,同时引入了自适应时变渐消因子,实时调整过程噪声协方差矩阵使其适应机动目标。仿真与海试数据处理结果表明,所提方法的机动目标运动分析性能最优,其目标运动分析结果能以最快的速度逼近克拉美罗下界。

多源传感器箱粒子LMB滤波算法

随着复杂跟踪场景的大量涌现,常规多源传感器多目标跟踪算法存在计算量大、跟踪精度低、无法估计目标航迹等不足,已无法满足现代战争的需求。笔者以主动传感器和被动传感器组成的多源传感器系统为背景,重点研究多源传感器多目标跟踪问题的实现算法。针对“多主动+多被动”多源传感器系统量测无法充分融合且整体算法计算复杂度较高的问题,提出一种多源传感器箱粒子标签多伯努利(MS-BPF-LMB)滤波算法。首先,对传感器依据不同主动传感器进行分组,即将所有传感器划分为若干“单主动+多被动”传感器组;然后,通过并行运算,对各传感器组运用基于角度关联的多传感器信息融合算法,得到跟踪所需的有效量测;最后,在跟踪滤波阶段,通过引入箱粒子滤波数值计算方法,将获取的量测点划分为若干箱粒子,并对箱粒子滤波下的多传感器量测更新系数进行重新定义,以较低的计算复杂度实现LMB(Labeled multi-Bernoulli,LMB)滤波。仿真结果表明,所提算法在保证跟踪精度的前提下,误差明显降低且算法复杂度下降约40%,能够有效处理异构数据多源信息融合问题。

融合时空信息的Transformer单目标跟踪算法

目前,主流的基于孪生网络的单目标跟踪方法,通过计算模板与搜索区域之间的相似度来匹配目标,缺乏对目标时空状态信息的利用。特别是当场景中存在多个相似目标时,孪生网络跟踪器往往无法精确区分目标,从而导致跟踪错误。针对上述问题,提出一种融合时空信息的Transformer单目标跟踪算法(SIFTransT)。该算法通过MixFormer(end-to-end tracking with iterative mixed attention)跟踪器获取初步的跟踪结果,设计了一个目标状态计算模块,用于计算并存储目标的状态信息,包括目标位置、边界框、速度、加速度、运动方向等,以此深入挖掘目标状态信息。构建了一个基于Transformer的时空信息融合模块,利用编码器的自注意力和解码器的交叉注意力,深入融合目标最近一段时间的状态信息,从而更加准确地对目标状态进行建模,提高目标跟踪的准确性。在LaSOT数据集上的实验结果表明,相比基准算法MixFormer,SIFTransT算法在AUC指标提高了2.8个百分点,PNorm指标提升了2.6个百分点,P指标提升了2.1个百分点,在搭载RTX8000显卡的服务器上平均每秒处理帧数达28帧。

基于特征融合及混合注意力的小目标船舶识别

[目的]旨在解决由于小目标船舶特征不显著导致网络模型识别率低的问题,提出基于图像与运动特征融合方法。[方法]在小目标船舶图像特征不明显的情况下,融合运动特征信息,丰富小型船舶的特征表达,同时提出混合注意力模型。在数据驱动条件下加入船舶目标先验信息,提高模型对关键特征的感知和利用能力。[结果]在720P分辨率图像中能实现10×10像素的小目标船舶识别,识别距离达到4 km范围,实现了广域船舶识别与定位功能。[结论]改进后的目标识别网络既具备像素级别的小目标船舶检测能力,同时又具备对环境噪声的抗干扰能力,突破了网络模型对小目标船舶识别率低的瓶颈。

基于IMM-UIF的多无人机纯角度机动目标跟踪

针对单无人机不能及时捕捉到目标的运动状态信息,很容易跟丢目标的问题,结合无迹信息滤波(UIF)算法和交互多模型(IMM)算法,提出了基于IMM-UIF的多无人机分布式融合估计算法。将各个无人机上的观测信息传输至中心节点,并统一优化各无人机的控制输入。仿真结果表明,基于IMM-UIF的多无人机分布式融合估计算法比基于IMM-UIF的单无人机跟踪精度提高了约30%,有效融合多无人机平台的量测信息,实现对目标稳定的高精度跟踪。

有限视距多传感器集群目标可扩展跟踪方法

在实际应用中,单传感器的视距、计算资源通常是有限的,多传感器网络的发展和应用为解决具有挑战性的目标跟踪问题提供了更多的可能性。相比于多目标跟踪,集群目标跟踪由于群内目标距离近、协同运动、数目多以及集群分裂合并等因素,会面临更具挑战性的数据关联和计算上的问题,而这些问题在多传感器融合系统中会进一步复杂化。针对有限视距情形下的多传感器集群目标跟踪问题,该文提出了一种可扩展的多传感器集群目标信念传播跟踪方法。该方法在贝叶斯框架下考虑集群结构的不确定性,构建多传感器集群目标联合后验概率密度分解和相应的因子图,以及通过在设计的因子图上运行信念传播算法高效求解数据关联问题。此外,该方法具有计算处理可扩展性,其计算复杂度与传感器数目、集群划分数目和观测数目呈线性关系,与目标数目呈二次关系。最后,仿真实验对比了不同方法关于GOSPA和OSPA(2)的性能,结果表明所提方法能够无缝跟踪集群目标和非群目标、充分利用多传感器信息互补优势、提升跟踪精度。

多平台异构信息融合的航空目标跟踪算法

该文以高空无人机(UAV)飞艇载双光电传感器,无人机载两坐标雷达对航空目标的精确定位跟踪为研究背景,针对参与融合的传感器均无法独立获得目标位置信息导致传统点迹关联、定位方法失效等问题,提出一种基于多平台异构信息融合的航空目标跟踪算法。首先,在坐标系转换的基础上提出基于角度-距离两级点迹关联算法,从而实现多传感器量测关联。其次,提出基于线面交汇融合定位算法,通过最小二乘法、交汇点投影、距离最近点解算及同源数据压缩确定目标的航迹起始位置。在此基础上,利用空基多平台侦察的异构信息,结合传统无迹卡尔曼滤波器(UKF)设计扩维UKF对航空目标进行跟踪。仿真结果表明,该算法实现了对航空高速目标的高精度跟踪。

机载雷达、红外、电子支援措施协同跟踪与管理

针对辐射限制下的目标跟踪问题,提出了一种机载雷达、红外传感器(infrared search and track,IRST)、电子支援措施(electronic support measure,ESM)协同跟踪与管理的方法。针对雷达、红外、ESM量测时间不一致的特点,采用顺序处理结构的多传感器集中式融合方式对目标进行跟踪,利用跟踪过程中的预测协方差与预定门限进行比较控制雷达辐射,并分析了红外、ESM不同间歇时间、不同控制门限与雷达辐射时间的相对关系。研究结论有助于提高作战飞机的抗侦察和抗干扰能力,从而提升整体的生存能力。

主被动多传感器多目标状态信息融合

研究了主被动多传感器多目标状态信息融合问题。针对被动式跟踪的特点，借助主动跟踪的距离通道值，提出类主动的被动式跟踪。在此基础上提出主被动串联状态信息融合和并联状态信息融合算法。仿真结果表明两种状态信息融合方法都可以大大提高跟踪精度，同时还可以提高系统的可靠性。