Performance evaluation of track fusion with information matrixfilter

In a multisensor environment, each sensor detects multiple targets and creates corresponding tracks. Fusion of tracks from these, possibly dissimilar, sensors yields more accurate kinematic and attribute information regarding the target. Two methodologies have been employed for such purpose, which are: measurement fusion and state vector fusion. It is well known that the measurement fusion approach is optimal but computationally inefficient and the state vector fusion algorithms are more efficient but suboptimal, in general. This is so because the state vector estimates to be fused obtained from two sensors, are not conditionally independent in general due to the common process noise from the target being tracked. It is to be noted that there are three approaches to state vector fusion, which are: weighted covariance, information matrix, and pseudomeasurement. This research is restricted solely to performance evaluation of the information matrix form of state vector fusion. Closed-form analytical solution of steady state fused covariance has been derived as a measure of performance using this approach. Note that the results are derived under the assumptions that the two sensors are synchronized and no misassociation or merged measurement is considered in the study. Results are compared with those using Monte Carlo simulation, which was used in the past to predict fusion system performance by various authors. These results provide additional insight into the mechanism of track fusion and greatly simplify evaluation of fusion performance. In addition, availability of such a solution facilitates the trade-off studies for designing fusion systems under various operating conditions     

A multi-rate sensor fusion approach using information filters for estimating aero-engine performance degradation

Gas-path performance estimation plays an important role in aero-engine health management, and Kalman Filter(KF) is a well-known technique to estimate performance degradation. In previous studies, it is assumed that different kinds of sensors are with the same sampling rate, and they are used for state estimation by the KF simultaneously. However, it is hard to achieve state estimation using various kinds of sensor measurements at the same sampling rate due to a complex network and physical characteristic differences between sensors, especially in an advanced multisensor architecture. For this purpose, a multi-rate sensor fusion using the information filtering approach is proposed based on the square-root cubature rule, which is called Multi-rate Squareroot Cubature Information Filter(MSCIF) to track engine performance degradation. Soft measurement synchronization of the MSCIF is designed to provide a sensor fusion condition for multiple sampling rates of measurement, and a fault sensor is isolated by maximum likelihood validation before state estimation. The contribution of this paper is to supply a novel multi-rate informationfilter approach for sensor fault tolerant health estimation of an aero-engine in a multi-sensor system. Tests are conducted for aero-engine performance degradation estimation with multiple sampling rates of sensor measurement on both digital simulation and semi-physical experiment.Experimental results illustrate the superiority of the proposed algorithm in terms of degradation estimation accuracy and robustness to sensor failure in a multi-sensor system.     

Efficient multisensor fusion using multidimensional dataassociation

We present the development of a multisensor fusion algorithm using multidimensional data association for multitarget tracking. The work is motivated by a large scale surveillance problem, where observations from multiple asynchronous sensors with time-varying sampling intervals (electronically scanned array (ESA) radars) are used for centralized fusion. The combination of multisensor fusion with multidimensional assignment is done so as to maximize the “time-depth” in addition to “sensor-width” for the number S of lists handled by the assignment algorithm. The standard procedure, which associates measurements from the most recently arrived S-1 frames to established tracks, can have, in the case of S sensors, a time-depth of zero. A new technique, which guarantees maximum effectiveness for an S-dimensional data association (S⩾3), i.e., maximum time-depth (S-1) for each sensor without sacrificing the fusion across sensors, is presented. Using a sliding window technique (of length S), the estimates are updated after each frame of measurements. The algorithm provides a systematic approach to automatic track formation, maintenance, and termination for multitarget tracking using multisensor fusion with multidimensional assignment for data association. Estimation results are presented for simulated data for a large scale air-to-ground target tracking problem     

多传感器融合状态估计方法综述

随着传感器网络技术的发展，多传感器融合状态估计凭借其鲁棒性、灵活性、可扩展性以及便于故障检测等优点，长期受到国内外学者的广泛关注，并取得了大量研究成果。数据融合的方法为融合状态估计奠定了理论基础，也是早期研究的主要方向，从20世纪70年代到20世纪末，相继发展出了集中式和分散式滤波架构及相应算法。无线通信技术的成熟以及一致性算法的出现使得分布式状态估计的研究进入了快车道，自2005年以来，大量基于一致性的分布式滤波算法被提出，其中不乏实用的经典方法和优秀的开创性方法。旨在梳理多传感器融合状态估计的发展，探究从数据融合到分布式滤波的内在联系，并对一些经典方法进行了总结。     

On optimal track-to-track fusion

Track-to-track fusion is an important part in multisensor fusion. Much research has been done in this area. Chong et al. (1979, 1986, 1990) among others, presented an optimal fusion formula under an arbitrary communication pattern. This formula is optimal when the underlying systems are deterministic, i.e., the process noise is zero, or when full-rate communication (two sensors exchange information each time they receive new measurements) is employed. However, in practice, the process noise is not negligible due to target maneuvering and sensors typically communicate infrequently to save communication bandwidth. In such situations, the measurements from two sensors are not conditionally (given the previous target state) independent due to the common process noise from the underlying system, and the fusion formula becomes an approximate one. This dependence phenomena was also observed by Bar-Shalom (1981) where a formula was derived to compute the cross-covariance of two track estimates obtained by different sensors. Based on this results a fusion formula was subsequently derived (1986) to combine the local estimates which took into account the dependency between the two estimates. Unfortunately, the Bayesian derivation made an assumption that is not met. This work points out the implicit approximation made and shows that the result turns out to be optimal only in the ML (maximum likelihood) sense. A performance evaluation technique is then proposed to study the performance of various track-to-track fusion techniques. The results provide performance bounds of different techniques under various operating conditions which can be used in designing a fusion system.     

Sequence comparison techniques for multisensor data fusion andtarget recognition

A new class of techniques for multisensor fusion and target recognition is proposed using sequence comparison by dynamic programming and multiple model estimation. The objective is to fuse information on the kinematic state and “nonkinematic” signature of unclassified targets, assessing the joint likelihood of all observed events for recognition. Relationships are shown to previous efforts in pattern recognition and state estimation. This research applies “classical” speech processing-related and other sequence comparison methods to moving target recognition, extends the efforts of previous researchers through improved fusion with kinematic information, relates the proposed techniques to Bayesian theory, and applies parameter identification methods to target recognition for improved understanding of the subject in general. The proposed techniques are evaluated and compared with existing approaches using the method of generalized ambiguity functions, which lends to a form of Cramer-Rao lower bound for target recognition     

Expert Systems in the Fighter of the 1990s

In the Advanced Tactical Fighter (ATF) to be deployed in the 1990s, the role of expert systems will enhance mission success. This paper discusses the utilization of two expert systems for handling multisensor data fusion and situation assessment. In multisensor data fusion, each sensor operates over a different region of the surveillance volume asynchronously and provides different measurements. In some instances, more than one sensor may yield the same measurement but with a different measurement accuracy. In this regard, the paper describes, in layman's terms, a system block diagram for processing the autonomous sensor track files and the possible need for a ``smart' fusion processor. This expert system is shown to manage the sensor outputs in both the temporal and spatial domains to maximize target identification confidence as well as kinematic state vector accuracy. The paper delineates the features needed by the fusion expert in order to assign a quality factor to each composite track file entry. A second expert system uses the output from fusion and other mission-related data to formulate the best picture of the surveillance volume at hand. This second expert system will show how historical data and real-time sensor data are merged for purposes of display parameters to the pilot, weapon cueing, countermeasures response management, and feedback to the fusion expert processor for individual sensor communication and data collection direction. The paper concludes with a tabular summary of the subprocesses of which these two expert systems may consist.     

A Polynomial Prediction Filter Method for Estimating Multisensor Dynamically Varying Biases

The estimation of the sensor measurement biases in a multisensor system is vital for the sensor data fusion. A solution is provided for the estimation of dynamically varying multiple sensor biases without any knowledge of the dynamic bias model pa- rameters. It is shown that the sensor bias pseudomeasurement can be dynamically obtained via a parity vector. This is accom- plished by multiplying the sensor uncalibrated measurement equations by a projection matrix so that the measured variable is eliminated from the equations. Once the state equations of the dynamically varying sensor biases are modeled by a polynomial prediction filter, the dynamically varying multisensor biases can be obtained by Kalman filter. Simulation results validate that the proposed method can estimate the constant biases and dynamic biases of multisensors and outperforms the methods reported in literature.     

Comparison of two-sensor tracking methods based on state vectorfusion and measurement fusion

There are two approaches to the two-sensor track-fusion problem. Y Bar-Shalom and L. Campo (ibid., vol.AES-22, 803-5, Nov. 1986) presented the state vector fusion method, which combines state vectors from the two sensors to form a new estimate while taking into account the correlated process noise. The measurement fusion method or data compression of D. Willner et al. (1976) combines the measurements from the two sensors first and then uses this fused measurement to estimate the state vector. The two methods are compared and an example shows the amount of improvement in the uncertainty of the resulting estimate of the state vector with the measurement fusion method     

Self-Tuning Multisensor Weighted Measurement Fusion Kalman Filter

For the multisensor systems with unknown noise variances, based on the solution of the matrix equations for the correlation function, the on-line estimators of the noise variance matrices are obtained, whose consistency is proved using the ergodicity of sampled correlation function. Further, two self-tuning weighted measurement fusion Kalman filters are presented for the multisensor systems with identical and different measurement matrices, respectively. Based on the stability of the dynamic error system, a new convergence analysis tool is presented for a self-tuning fuser, which is called the dynamic error system analysis (DESA) method. A new concept of convergence in a realization is presented, which is weaker than the convergence with probability one. It is rigorously proved that the proposed self-tuning Kalman fusers converge to the steady-state optimal Kalman fusers in a realization or with probability one, so that they have asymptotic global optimality. A simulation example for a target tracking system with 3 sensors shows their effectiveness.     

Fusion of distributed extended forgetting factor RLS state estimators

For single-target multisensor systems, two fusion methods are presented for distributed recursive state estimation of dynamic systems without knowledge of noise covariances. The estimator at every local sensor embeds the dynamics and the forgetting factor into the recursive least squares (RLS) method to remedy the lack of knowledge of noise statistics, developed before as the extended forgetting factor recursive least squares (EFRLS) estimator. It is proved that the two fusion methods are equivalent to the centralized EFRLS that uses all measurements from local sensors directly and their good performance is shown by simulation examples.     

应用神经网络信息融合诊断航空发动机故障

研究了基于神经网络信息融合技术,同时结合模糊集合论对发动机气路部件进行故障诊断的方法,并以某型涡轴发动机为对象进行了仿真分析.研究结果表明该方法的故障诊断过程相对简单,对模型的精度要求不高,能够降低虚警、误报、漏报等情况的发生.      

Joint sensor registration and track-to-track fusion for distributed trackers

Sensor registration deals with the correction of registration errors and is an inherent problem in all multisensor tracking systems. Traditionally, it is viewed as a least squares or a maximum likelihood problem independent of the fusion problem. We formulate it as a Bayesian estimation problem where sensor registration and track-to-track fusion are treated as joint problems and provide solutions in cases 1) when sensor outputs (i.e., raw data) are available, and 2) when tracker outputs (i.e., tracks) are available. The solution to the latter problem is of particular significance in practical systems as band limited communication links render the transmission of raw data impractical and most of the practical fusion systems have to depend on tracker outputs rather than sensor outputs for fusion. We then show that, under linear Gaussian assumptions, the Bayesian approach leads to a registration solution based on equivalent measurements generated by geographically separated radar trackers. In addition, we show that equivalent measurements are a very effective way of handling sensor registration problem in clutter. Simulation results show that the proposed algorithm adequately estimates the biases, and the resulting central-level trucks are free of registration errors.     

The modeling and estimation of asynchronous multirate multisensor dynamic systems

An asynchronous data fusion problem based on a kind of multirate multisensor dynamic system is studied. The system is observed by multirate sensors independently, with the state model known at the finest scale. Under the assumption that the sampling rates of sensors decrease successively by any positive integers, the discrete dynamic system models are established based on each single sensor and an asynchronous multirate multisensor state fusion estimation algorithm is presented. Theoretically, the estimate is proven to be unbiased and the optimal in the sense of linear minimum covariance, the fused estimate is better than the Kalman filtering results based on each single sensor, and the accuracy of the fused estimate will decrease if any of the sensors' information is neglected. The feasibility and effectiveness of the algorithm are shown through simulations.     

Voting fusion adaptation for landmine detection

This paper presents a voting fusion application for use with a remotely controlled multisensor vehicle platform for antitank landmine detection. Data from three landmine detection sensors mounted at the front of the vehicle enhance the probability of detection and, when combined via data fusion, limit the false alarm density to practical levels. The performance of the voting fusion scheme presented in this paper is contrasted with a heuristic data fusion approach developed by General Dynamics Canada.     

Performance evaluation for MAP state estimate fusion

This paper presents a quantitative performance evaluation method for the maximum a posteriori (MAP) state estimate fusion algorithm. Under ideal conditions where data association is assumed to be perfect, it has been shown that the MAP or best linear unbiased estimate (BLUE) fusion formula provides the best linear minimum mean squared estimate (LMMSE) given local estimates under the linear Gaussian assumption for a static system. However, for a dynamic system where fusion is recursively performed by the fusion center on local estimates generated from local measurements, it is not obvious how the MAP algorithm will perform. In the past, several performance evaluation methods have been proposed for various fusion algorithms, including simple convex combination, cross-covariance combination, information matrix, and MAP fusion. However, not much has been done to quantify the steady state behavior of these fusion methods for a dynamic system. The goal of this work is to present analytical fusion performance results for MAP state estimate fusion without extensive Monte Carlo simulations, using an approach developed for steady state performance evaluation for track fusion. Two different communication strategies are considered: fusion with and without feedback to the sensors. Analytic curves for the steady state performance of the fusion algorithm for various communication patterns are presented under different operating conditions.     

Data fusion of decentralized local tracker outputs

An approach for fusing offboard track-level data at a central fusion node is presented. The case where the offboard tracker continues to update its local track estimate with measurement and system dynamics models that are not necessarily linear is considered. An algorithm is developed to perform this fusion at a central node without having access to the offboard measurements, their noise statistics, or the location of the local estimator. The algorithm is based on an extension of results that were originally established for linear offboard trackers. A second goal of this work is to develop an inequality constraint for selecting the proper sampling interval for the incoming state estimates to the fusion node. This interval is selected to allow use of conventional Kalman filter algorithms at the fusion node without suffering error performance degradation due to processing a correlated sequence of track state estimates     

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。     

基于D-S证据理论的目标属性辨识研究

D—S证据组合理论已经成为不确定性推理的一种重要方法，基于该理论的多传感器决策层信息融合已得到广泛应用。本文在简要介绍D—S证据组合理论框架基础上，阐述了基于D—S证据组合理论的信息融合一般步骤及决策层判决方法；分析了发生证据严重冲突时，Dempster证据组合规则出现的不可靠和不合理的本质原因，介绍了针对D—S证据组合规则进行修正的几种主要方法；最后，提出了需要继续深入探讨的问题。     

多传感器数据融合方法在加力燃烧室试验中的应用研究

在航空发动机加力燃烧室试验中,为实现同一截面真实气流状态参数的多传感器精确测量,提出了1种多传感器数据融合方法。该方法运用格拉布斯准则剔除多传感器测量数据中的无效数据,运用方差自适应加权融合方法对有效数据进行融合计算并获取真实参数,并分别利用算术平均值方法和自适应加权融合方法对试验数据进行处理分析。结果表明:通过多传感器融合方法获取的数据均方差明显优于算术平均值方法获取的方差,具有更高的测试精度和可靠性。