A framework for positioning, navigation and tracking problems using particle filters (sequential Monte Carlo methods) is developed. It consists of a class of motion models and a general non-linear measurement equation in position. A general algorithm is presented, which is parsimonious with the particle dimension. It is based on marginalization, enabling a Kalman filter to estimate all position derivatives, and the particle filter becomes low-dimensional. This is of utmost importance for highperformance real-time applications. Automotive and airborne applications illustrate numerically the advantage over classical Kalman filter based algorithms. Here the use of non-linear models and non-Gaussian noise is the main explanation for the improvement in accuracy. More specifically, we describe how the technique of map matching is used to match an aircraft's elevation profile to a digital elevation map, and a car's horizontal driven path to a street map. In both cases, real-time implementations are available, and tests have shown that the accuracy in both cases is comparable to satellite navigation (as GPS), but with higher integrity. Based on simulations, we also argue how the particle filter can be used for positioning based on cellular phone measurements, for integrated navigation in aircraft, and for target tracking in aircraft and cars. Finally, the particle filter enables a promising solution to the combined task of navigation and tracking, with possible application to airborne hunting and collision avoidance systems in cars.

Detecting regions of change in multiple images of the same scene taken at different times is of widespread interest due to a large number of applications in diverse disciplines, including remote sensing, surveillance, medical diagnosis and treatment, civil infrastructure, and underwater sensing. This paper presents a systematic survey of the common processing steps and core decision rules in modern change detection algorithms, including significance and hypothesis testing, predictive models, the shading model, and background modeling. We also discuss important preprocessing methods, approaches to enforcing the consistency of the change mask, and principles for evaluating and comparing the performance of change detection algorithms. It is hoped that our classification of algorithms into a relatively small number of categories will provide useful guidance to the algorithm designer.

The particle filter offers a general numerical tool to approximate the posterior density function for the state in nonlinear and non-Gaussian filtering problems. While the particle filter is fairly easy to implement and tune, its main drawback is that it is quite computer intensive. However, due to faster computers this drawback can be overcome and as a result the particle filter has quickly become a popular tool in signal processing applications. The computational complexity increases quickly with the state dimension for the problem at hand. One remedy to this problem is a technique known as RaoBlackwellization, where states appearing linearly in the dynamics are marginalized out. The result of this is that one Kalman filter is associated with each particle. Our main contribution in this article is to derive the details for the marginalized particle filter for a general nonlinear state-space model. We will also discuss some important special cases occurring in typical signal processing applications. The marginalized particle filter is applied to an integrated navigation system for aircraft. It is demonstrated that the complete high-dimensional system can be based on the particle filter using marginalization for all but three states. Excellent performance on real flight data is reported.

Abstract not found

A number of abrupt change detection methods have been proposed in the past, among which are efficient modelbased techniques such as the Generalized Likelihood Ratio (GLR) test. We consider the case where no accurate nor tractable model can be found, using a model-free approach, called Kernel change detection (KCD). KCD compares two sets of descriptors extracted online from the signal at each time instant: the immediate past set and the immediate future set. Based on the soft margin single-class Support Vector Machine (SVM), we build a dissimilarity measure in feature space between those sets, without estimating densities as an intermediary step. This dissimilarity measure is shown to be asymptotically equivalent to the Fisher ratio in the Gaussian case. Implementation issues are addressed, in particular, the dissimilarity measure can be computed online in input space. Simulation results on both synthetic signals and real music signals show the efficiency of KCD.

We propose some Bayesian methods to address the problem of fitting a signal modeled by a sequence of piecewise constant linear (in the parameters) regression models, for example, autoregressive or Volterra models. A joint prior distribution is set up over the number of the changepoints/knots, their positions, and over the orders of the linear regression models within each segment if these are unknown. Hierarchical priors are developed and, as the resulting posterior probability distributions and Bayesian estimators do not admit closed-form analytical expressions, reversible jump Markov chain Monte Carlo (MCMC) methods are derived to estimate these quantities. Results are obtained for standard denoising and segmentation of speech data problems that have already been examined in the literature. These results demonstrate the performance of our methods.

this paper, we first consider a static nonlinear mixing model, with a successful application to real-world speech data compression. Then we discuss extraction of sources from nonlinear dynamic processes, and detection of abrupt changes in the process dynamics. In a difficult test problem with chaotic data, our approach clearly outperforms currently available nonlinear prediction and change detection techniques. The proposed methods are computationally demanding, but they can be applied to blind nonlinear problems of higher dimensions than other existing approaches

... without addressing the so-called measurement-origin uncertainty. Part I and Part II deal with target motion models. Part III covers measurement models and associated techniques. Part IV is concerned with tracking techniques that are based on decisions regarding target maneuvers. This part surveys the multiple-model methods---the use of multiple models (and filters) simultaneously---which is the prevailing approach to maneuvering target tracking in the recent years. The survey is presented in a structured way, centered around three generations of algorithms: autonomous, cooperating, and variable structure. It emphasizes on the underpinning of each algorithm and covers various issues in algorithm design, application, and performance.

Abstract: Emerging driver assistance systems, such as look-ahead cruise controllers for heavy duty vehicles, require high precision digital maps. This contribution presents a road grade estimation algorithm for fusion of GPS and vehicle real-time sensor data, with measurements from previous runs over the same road segment. The resulting road grade estimate is thus enhanced using measurements from additional traversals of known roads. Distributed data fusion is utilized to ensure that the storage requirement of known roads does not increase when additional measurements are processed. The implemented algorithm, which is based on extended Kalman filtering and smoothing, is described in detail. Experiments on a Scania test vehicle show the advantages and some of the challenges with the proposed approach.

n parameters." For ex. ample, diagonal elements of the transition probability matrix (TPM) are computed assuming prior knowledge of the mean sojourn time [5]. Clearly in situations where this prior knowledge is poor or even lacking, it would be desirable to estimate transition probabilities recursively, from the measurement data. The problem has been studied by several authors [6, 7]. Tugnait [6] developed a truncate maximum likelihood technique whereas Manuscript received January 4, 2002; released for publication May 23, 2002. IEEE Log No. T-AES/$8/3/06454. Refereeing of this contribution was handled by P. K. Willett. Australian Crown copyright. 0018-9251/02/$ 17.00 ) 2002 IEEE Jilkov and Li [7] recently derived Bayesian estimates of transition probabilities assuming prior knowledge of all moments of the TPM prior probability density function (pdf). These algorithms rely on the knowledge of the "current" time-step model likelihoods which are unknown and need to. be approximated.