Variational track management

A system for tracking a plurality of moving objects comprising:

• a sensor system configured to provide a set of measurements representing at least respective positions of the plurality of moving objects;
• a track state updating component configured to determine an estimate of a posterior probability distribution for a plurality of track states from an estimate of the posterior probability distribution for a plurality of possible assignments of the set of measurements to a set of tracks representing trajectories of the plurality of moving objects and the set of measurements;
• a track assignment updating component configured to determine a new estimate of the posterior probability distribution for the plurality of possible assignments of the set of measurements to the set of tracks from the set of measurements and the estimate of a posterior probability distribution for a plurality of track states;
• and
• a lower bound computation component configured to compute a variational lower bound, representing a lower bound for a marginal probability of the set of measurements given a model defined by the new estimate of the posterior probability distribution for the plurality of possible assignments of the set of measurements to the set of tracks and the estimate of the posterior probability distribution for a plurality of track states, from the new estimate of the posterior probability distribution for the plurality of possible assignments of the set of measurements to the set of tracks, the estimate of a posterior probability distribution for a plurality of track states, and the set of measurements;
• wherein each of the track state updating component, the track assignment updating component, and the lower bound computation component collectively perform an iterative determination of the posterior probability distribution for the plurality of possible assignments of the set of measurements to the set of tracks and the posterior probability distribution for a plurality of track states until the variational lower bound is less than a threshold value.