[Une nouvelle méthode de moindres carrés non linéaires pour le pistage de cibles hyper-manœuvrantes]
Les trajectoires de véhicules aériens ou marins sont en général formées d'une succession de trajectoires lisses, séparées par des manœuvres brusques. En particulier, les cibles hyper-manœuvrantes modernes peuvent changer de cap de façon très abrupte, avec des accélérations pouvant aller jusqu'à 15g. Le comportement de la cible est donc modélisé par des modèles de Markov déterministes par morceaux, grâce à des paramètres à sauts. Depuis quelques années, les méthodes de lissage en temps réel sont devenues une alternative aux habituels algorithmes de filtrage, tels que le filtre de Kalman étendu (EKF). Dans cet article, de telles méthodes de lissage sont utilisées pour les phases sans sauts et sont combinées à une approche probabiliste pour déterminer les instants de sauts. L'algorithme ainsi proposé est comparé à un algorithme multi-modèles, l'IMM (Interactirng Multiple Model), en particulier pour l'estimation de la vitesse de la cible, sur un ensemble de trajectoires représentatif et difficile.
Trajectories of aerial and marine vehicles are typically made of a succession of smooth trajectories, linked by abrupt changes, i.e. maneuvers. Notably, modern highly maneuvering targets are capable of very brutal changes in the heading, with accelerations of up to . As a result, we model the target behavior using piecewise deterministic Markov models, driven by parameters that jump at unknown times. Over the past years, real-time (or incremental) optimization-based smoothing methods have become a popular alternative to nonlinear filters, such as the Extended Kálmán Filter (EKF), owing to the successive relinearizations that mitigate the linearization errors that inherently affect the EKF estimates. In the present paper, we propose to combine such methods for tracking the target during non-jumping phases with a probabilistic approach to detect jumps. Our algorithm is shown to compare favorably to the state-of-the-art Interacting Multiple Model (IMM) algorithm, especially in terms of target's velocity estimation, on a set of meaningful and challenging trajectories.
Mot clés : Pistage mono-cible, Radar, Lissage non linéaire, Estimation d'état non linéaire, Approche probabiliste
@article{CRPHYS_2019__20_3_228_0,
author = {Marion Pilt\'e and Silv\`ere Bonnabel and Fr\'ed\'eric Livernet},
title = {A novel nonlinear least-squares approach to highly maneuvering target tracking},
journal = {Comptes Rendus. Physique},
pages = {228--239},
publisher = {Elsevier},
volume = {20},
number = {3},
year = {2019},
doi = {10.1016/j.crhy.2019.05.019},
language = {en},
}
TY - JOUR AU - Marion Pilté AU - Silvère Bonnabel AU - Frédéric Livernet TI - A novel nonlinear least-squares approach to highly maneuvering target tracking JO - Comptes Rendus. Physique PY - 2019 SP - 228 EP - 239 VL - 20 IS - 3 PB - Elsevier DO - 10.1016/j.crhy.2019.05.019 LA - en ID - CRPHYS_2019__20_3_228_0 ER -
Marion Pilté; Silvère Bonnabel; Frédéric Livernet. A novel nonlinear least-squares approach to highly maneuvering target tracking. Comptes Rendus. Physique, Volume 20 (2019) no. 3, pp. 228-239. doi : 10.1016/j.crhy.2019.05.019. https://comptes-rendus.academie-sciences.fr/physique/articles/10.1016/j.crhy.2019.05.019/
[1] Rao-blackwellised particle filtering for dynamic Bayesian networks, Proceedings of the Sixteenth Conference on Uncertainty in Artificial Intelligence, Morgan Kaufmann Publishers Inc., 2000, pp. 176-183
[2] Variable rate particle filters for tracking applications, Proc. IEEE/SP 13th Workshop on Statistical Signal Processing, 2005, IEEE, 2005, pp. 1280-1285
[3] Dynamical models for tracking with the variable rate particle filter, Proc. 15th International Conference on Information Fusion (FUSION), IEEE, 2012, pp. 1769-1775
[4] Particle smoothing algorithms for variable rate models, IEEE Trans. Signal Process., Volume 61 (2013) no. 7, pp. 1663-1675
[5] Estimation with Applications to Tracking and Navigation: Theory Algorithms and Software, Wiley, 2004 https://books.google.fr/books?id=xz9nQ4wdXG4C
[6] Target tracking through a coordinated turn, Proc. ICASSP '05, IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005, vol. 4, 2005, pp. 21-24 | DOI
[7] An innovative nonlinear filter for radar kinematic estimation of maneuvering targets in 2d, Proc. 18th International Radar Symposium (IRS), IEEE, 2017, pp. 1-10
[8] The graph slam algorithm with applications to large-scale mapping of urban structures, Int. J. Robot. Res., Volume 25 (2006) no. 5–6, pp. 403-429
[9] Square root sam: simultaneous localization and mapping via square root information smoothing, Int. J. Robot. Res., Volume 25 (2006) no. 12, pp. 1181-1203
[10] iSAM: incremental smoothing and mapping, IEEE Trans. Robot., Volume 24 (2008) no. 6, pp. 1365-1378 | DOI
[11] iSAM2: incremental smoothing and mapping using the Bayes tree, Int. J. Robot. Res., Volume 31 (2012) no. 2, pp. 216-235 | DOI
[12] et al. FastSLAM: a factored solution to the simultaneous localization and mapping problem, AAAI/IAAI, 2002, pp. 593-598
[13] Motion tracking with fixed-lag smoothing: algorithm and consistency analysis, Proc. IEEE International Conference on Robotics and Automation (ICRA), IEEE, 2011, pp. 5655-5662
[14] Sliding window filter with application to planetary landing, J. Field Robot., Volume 27 (2010) no. 5, pp. 587-608
[15] System Identification: Theory for the User, Prentice-Hall, 1987
[16] The interacting multiple model algorithm for systems with Markovian switching coefficients, IEEE Trans. Autom. Control, Volume 33 (1988) no. 8, pp. 780-783
Cité par Sources :
Commentaires - Politique