Probabilistic data association in information space for generic sensor data fusion

In this contribution, the probabilistic data association for single and multi-target scenarios is adapted for the use with information filters to realize a more generic sensor interface. Therefore the calculation of the weights and the estimation equations for the use with information filters are derived, an approximation for the gating volume using the new approach is introduced and the approximation for the Mahalanobis distance in information space is reviewed. The resulting probabilistic multi-target filter using information measurements instead of state space measurements is evaluated in two simulations and one real data scenario via the optimal subpattern assignment metric.