Conference Paper

Efficient Nonlinear Measurement Updating based on Gaussian Mixture Approximation of Conditional Densities

Univ. Karlsruhe, Karlsruhe
DOI: 10.1109/ACC.2007.4282269 Conference: American Control Conference, 2007. ACC '07
Source: IEEE Xplore


Filtering or measurement updating for nonlinear stochastic dynamic systems requires approximate calculations, since an exact solution is impossible to obtain in general. We propose a Gaussian mixture approximation of the conditional density, which allows performing measurement updating in closed form. The conditional density is a probabilistic representation of the nonlinear system and depends on the random variable of the measurement given the system state. Unlike the likelihood, the conditional density is independent of actual measurements, which permits determining its approximation off-line. By treating the approximation task as an optimization problem, we use progressive processing to achieve high quality results. Once having calculated the conditional density, the likelihood can be determined on-line, which, in turn, offers an efficient approximate filter step. As result, a Gaussian mixture representation of the posterior density is obtained. The exponential growth of Gaussian mixture components resulting from repeated filtering is avoided implicitly by the prediction step using the proposed techniques.

Download full-text


Available from: Marco F. Huber
  • Source
    • "Alternatively, Expectation Maximization (EM) algorithm can be used to simultaneously predict and reduce the Gaussian Mixture [26]–[29], e.g. by running the EM algorithm on synthetically generated data [27], [29]. Among the reduction schemes mentioned above, the methods that rely on removing clusters are the least computationally complex. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In many signal processing applications it is required to estimate the unobservable state of a dynamic system from its noisy measurements. For linear dynamic systems with Gaussian Mixture (GM) noise distributions, Gaussian Sum Filters (GSF) provide the MMSE state estimate by tracking the GM posterior. However, since the number of the clusters of the GM posterior grows exponentially over time, suitable reduction schemes need to be used to maintain the size of the bank in GSF. In this work we propose a low computational complexity reduction scheme which uses an initial state estimation to find the active noise clusters and removes all the others. Since the performance of our proposed method relies on the accuracy of the initial state estimation, we also propose five methods for finding this estimation. We provide simulation results showing that with suitable choice of the initial state estimation (based on the shape of the noise models), our proposed reduction scheme provides better state estimations both in terms of accuracy and precision when compared with other reduction methods.
    Preview · Article · May 2014
  • Source
    • "The key idea of this paper is to enable a closed-form calculation of the prediction and measurement step by means of wavelet expansions of the conditional densities. Compared to similar approaches using axis-aligned Gaussian mixtures as in [6] or [11], wavelets have some significant advantages . Besides their good approximation quality, no complex optimization problem has to be solved in order to obtain the decomposition of a given density. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Bayesian estimation for nonlinear systems is still a challenging problem, as in general the type of the true probability density changes and the complexity increases over time. Hence, approximations of the occurring equations and/or of the underlying probability density functions are inevitable. In this paper, we propose an approximation of the conditional densities by wavelet expansions. This kind of representation allows a sparse set of characterizing coefficients, especially for smooth or piecewise smooth density functions. Besides its good approximation properties, fast algorithms operating on sparse vectors are applicable and thus, a good trade-off between approximation quality and run-time can be achieved. Moreover, due to its highly generic nature, it can be applied to a large class of nonlinear systems with a high modeling accuracy. In particular, the noise acting upon the system can be modeled by an arbitrary probability distribution and can influence the system in any way.
    Full-text · Conference Paper · Jan 2011
  • Source
    • "This special type of a squared integral measure is defined over the corresponding cumulative distribution functions of both densities and allows quantifying the distance especially in case of the used Dirac delta functions. In contrast to our purely Gaussian mixture conditional density approximation approach in [12], this optimization problem can be solved analytically and is not restricted to time-invariant systems. The hybrid structure of the approximate conditional density allows an analytical and efficient evaluation of the prediction step as well as the measurement update step. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In nonlinear Bayesian estimation it is generally inevitable to incorporate approximate descriptions of the exact estimation algorithm. There are two possible ways to involve approximations: Approximating the nonlinear stochastic system model or approximating the prior probability density function. The key idea of the introduced novel estimator called Hybrid Density Filter relies on approximating the nonlinear system, thus approximating conditional densities. These densities nonlinearly relate the current system state to the future system state at predictions or to potential measurements at measurement updates. A hybrid density consisting of both Dirac delta functions and Gaussian densities is used for an optimal approximation. This paper addresses the optimization problem for treating the conditional density approximation. Furthermore, efficient estimation algorithms are derived based upon the special structure of the hybrid density, which yield a Gaussian mixture representation of the system state's density.
    Full-text · Conference Paper · Aug 2007
Show more