多目标数据关联的多维Lagrangian松驰法

将多目标跟踪中的数据关联问题提成为一个离散最优化问题 ,并采用 L agrangian松驰算法来克服该问题面临的计算困难 ,这是目前多祯多目标测量数据关联研究发展的重要方向。首先将多祯多目标数据关联问题建立为一个带约束的多维分配问题 ;并在维数大于 2时 ,采用 L agrangian松驰算法来松驰约束条件 ,使问题降维为计算上可接受的二维分配问题。

Multi-target Data Association with Multi-dimensional Lagrangian Relaxation

Multi-target data association problem has been widely treated as a multidimensional assignment problem recently.The major obstacle to this approach lies in its inapplicable computational cost when the assignment dimension is greater than 2,NP-hard as it turns to be.A Lagrangian relaxation mechanism to handle this problem is introduced in this paper.It iteratively updates the Lagrangian multipliers and relaxes multidimensional constraints to reduce higher dimensional assignment problems to 2D ones,which have polynomial complexity and available standard algorithm packages.