Network-Centric Multiple Frame Association for Distributed Multiple Target Tracking

The coordinated use of these multiple distributed sensors by network communication has the potential to substantially improve track state estimates, associated feature and attribute estimates, and identification of targets (ID and CID), primarily due to geometric diversity, complementary sensor information, and different coverage areas. To achieve this substantial improvement in the fusion capability, algorithms and data processing capabilities that address a number of technical issues and are robust in challenging and anomalous conditions must be developed. For single sensors or multiple sensors in a centralized architecture, multiple frame data association methods such as MHT and MFA are superior to the single frame methods due the ability to hold difficult decisions in abeyance until more information is available or, equivalently, the opportunity to change past decisions to improve current decisions. Thus, the goal of this program is to develop a decentralized network-centric class of multiple frame data association methods (MFA/MHT) that achieves the quality of a MFA/MHT centralized tracker and SIAP across the network of platforms in a communication constrained environment.