Toyon has provided extensive research in the domain of feature-aided tracking algorithms that exploit the dependence between the kinematic and signature information in sensor data. Toyon’s feature-aided trackers employ theoretically rigorous approaches to quantify the probability of each possible association hypothesis using both the kinematic and signature information.
Toyon has demonstrated the benefits of such algorithms to both improve the quality of track identity estimates and extend the lifetime of tracks in challenging tracking environments. Our approach has been shown to perform well when limited a priori sensor signature data is available, which is often the case in the realistic situation where the signature database is populated “on-the-fly”. Moreover, Toyon’s feature-aided trackers have been designed to work with a wide variety of data classifiers and Automatic Target Recognition (ATR) algorithms.
Toyon’s Tracked Object Manager (TOM) algorithm is an example of our unique approach to feature-aided tracking.
For more information regarding Toyon’s Feature-Aided Tracking capabilities, please contact us.