Abstract
The aim of syntactic tracking is to classify spatio-temporal patterns of atarget's motion using natural language processing models. In this paper, wegeneralize earlier work by considering a constrained stochastic context freegrammar (CSCFG) for modeling patterns confined to a roadmap. The constrainedgrammar facilitates modeling specific directions and road names in a roadmap.We present a novel particle filtering algorithm that exploits the CSCFG modelfor estimating the target's patterns. This meta-level algorithm operates inconjunction with a base-level tracking algorithm. Extensive numerical resultsusing simulated ground moving target indicator (GMTI) radar measurements showsubstantial improvement in target tracking accuracy.
Quick Read (beta)
loading the full paper ...