NNAKF: A NEURAL NETWORK ADAPTED KALMAN FILTER FOR TARGET TRACKING

Sami Jouaber (1, 2) , Silvere Bonnabel (1, 3) , Santiago Velasco-Forero (4) , Marion Pilte (5)

1 CAOR - Centre de Robotique
2 THALES [France]
3 ISEA - Institut de sciences exactes et appliquées
4 CMM - Centre de Morphologie Mathématique
5 Thales Air Systems

An adaptive three-dimensional Kalman filter for the tracking of maneuvering targets in three dimensions is proposed. In the radar industry, numerous trackers are based on a constant velocity model, with a process noise covariance matrix Q which is adapted in real time to enhance tracking: it is kept at moderate values during straight lines where the constant velocity assumption applies and is increased during maneuvers. In the present paper we advocate a novel method to increase Q during maneuvers (and hence the Kalman gains) based on a recurrent neural network (RNN). The difficulty and the interest of our approach lies in the fact the neural network is trained together with the filter, by backpropagation through the filter, and hence learns the covariance matrix such as to directly maximize the accuracy of the final output.

Mots clés: Recurrent Neural Network (RNN); Kalman Filter; Radar Tracking

Domaines: Sciences de l'ingénieur [physics]; Intelligence artificielle [cs.AI]; Traitement du signal et de l'image [eess.SP]; Réseau de neurones [cs.NE]; Apprentissage [cs.LG]

Sami Jouaber, Silvere Bonnabel, Santiago Velasco-Forero, Marion Pilte. NNAKF: A NEURAL NETWORK ADAPTED KALMAN FILTER FOR TARGET TRACKING. ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Jun 2021, Toronto, France. pp.4075-4079, ⟨10.1109/ICASSP39728.2021.9414681⟩. ⟨hal-03463046⟩