Developing a Real-Time Track Display That Operators Do Not Hate

Dept. of Electr. & Comput. Eng., Univ. of Connecticut, Storrs, CT, USA
IEEE Transactions on Signal Processing (Impact Factor: 2.79). 08/2011; 59(7):3441 - 3447. DOI: 10.1109/TSP.2011.2135346
Source: IEEE Xplore


We formulate a method of estimating target states that minimizes the mean optimal subpattern assignment (MOSPA) metric, applied suboptimally to a multi-hypothesis tracker (MHT) and optimally to a particle filter. This gives the operator a display of the targets with reduced jitter and track switching.

1 Follower
5 Reads
  • Source
    • "We have derived both Wasserstein barycenters and MMOSPA estimators from fundamental mathematical concepts, i.e., the Fréchet mean and the Wasserstein distance, and pointed out recent algorithms, variations, applications, and further developments. In particular, we revealed that the empirical MMOSPA estimate is equivalent to the Wasserstein barycenter for point clouds, which both have been recently independently discussed in [15], [17], [18], [21], [25] and [33]. The two research areas might benefit from the discovered equivalence, e.g., algorithms and insights for solving the underlying optimization problem can be shared. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The two title concepts have been evolving rather rapidly, but independent of each other. The Wasserstein barycenter, on one hand, has mostly made its appearance in image processing as it can describe a measure of similarity between images. Its minimization might, for example, suggest the best match in image alignment. On the other hand, MMOSPA estimation has been applied largely to multi-target tracking. The Optimal Sub-Pattern Assignment (OSPA) measures the distance between two sets and the Mean OSPA (MOSPA) can be minimized to give the Minimum MOPSA (MMOSPA), which improves MMSE estimation of the target locations when the labeling of the targets in the set is not important. Approximate and exact algorithms have evolved for both Wasserstein barycenters and MMOSPA estimation. Here, we draw connections between the two perspectives and elaborate how they can benefit from each other.
    Full-text · Article · Sep 2015 · IEEE Signal Processing Letters
  • Source
    • "for Gaussian mixtures. In [10] "
    [Show abstract] [Hide abstract]
    ABSTRACT: In multi-object estimation, the traditional minimum mean squared error (MMSE) objective is unsuitable: a simple permutation of object identities can turn a very good estimate into what is apparently a very bad one. Fortunately, a criterion tailored to sets—minimization of the mean optimal sub-pattern assignment (MMOSPA)—has recently evolved. Aside from special cases, exact MMOSPA estimates have seemed difficult to compute. But in this work we present the first exact polynomial-time algorithms for calculating the MMOSPA estimate for probability densities that are represented by particles. The key insight is that the MMOSPA estimate can be found by means of enumerating the cells of a hyperplane arrangement, which is a traditional problem from computational geometry. Although the runtime complexity is still high for the general case, efficient algorithms are obtained for two special cases, i.e., (i) two targets with arbitrary state dimensions and (ii) an arbitrary number of one-dimensional targets.
    Full-text · Article · May 2015 · IEEE Transactions on Signal Processing
  • Source
    • "For example, the modes of a Gaussian mixture consisting of two one-dimensional Gaussians collapse when the absolute distance between their means is less than 2σ, where σ is the standard deviation of both Gaussians [10]. A further general disadvantage of the MAP estimate is that it tends to jitter because the MAP estimate ignores everything around the maximum of the posterior, which may change frequently (see [3] for a more detailed discussion). "
    [Show abstract] [Hide abstract]
    ABSTRACT: This work is concerned with direction-of-arrival (DOA) estimation of narrowband signals from multiple targets using a planar antenna array. We illustrate the shortcomings of Maximum Likelihood (ML), Maximum a Posteriori (MAP), and Minimum Mean Squared Error (MMSE) estimation, issues that can be attributed to the symmetry in the likelihood function that must exist when there is no information about labeling of targets. We proffer the recently introduced concept of Minimum Mean OSPA (MMOSPA) estimation that is based on the optimal sub-pattern assignment (OSPA) metric for sets and hence inherently incorporates symmetric likelihood functions.
    Full-text · Conference Paper · Jun 2014
Show more