A Distributed ADMM Approach to Informative Trajectory Planning for Multi-Target Tracking
share
This paper presents a distributed optimization method for informative trajectory planning in a multi-target tracking problem. The original multi-target tracking problem is formulated as a distributed optimization problem that can be expressed in the form of a subproblem for each target, which is formally described as a partially observable Markov decision process (POMDP). A local trajectory optimization method is utilized to solve the subproblems and is integrated with distributed Alternating Direction Method of Multipliers (ADMM). In order to reduce the computation time of the algorithm, a heuristic rule suitable for multiple-target tracking problems is proposed, as is a replanning scheme for real-time implementation, which considers the computation time and communication time. The proposed algorithm can handle both trajectory optimization and task assignment in multi-target tracking problems simultaneously. Numerical examples are presented to demonstrate the applicability of the proposed algorithm.
READ FULL TEXT
