Multiple hypothesis S‐dimensional assignment algorithm for data association of angle‐only sensors with limited fields of view

This paper focuses on how to obtain globally optimal multisensor data association results under the condition that number of targets is unknown. problem likely occur in scenarios where sensors’ fields view (FoVs) are limited. The authors proved that, ideally, cost result a convex function assumed target and with minimal value only if correctly assumed. Based this, multiple hypothesis S-dimensional assignment (MHSDA) algorithm proposed, which replaces traditional irrevocable onetime assumption approach counteract effect uncertainty. Furthermore, hybrid implementation architecture based Union Find plus MHSDA proposed ensure real-time performance. requires neither prior knowledge nor FoVs sensors applicable dense scenarios. Simulation different FoV, measurement noise configurations demonstrate effectiveness algorithm.