The current U.S. Navy heavy-weight torpedo (HWT) is a complex autonomous system that carries out several autonomous phases after being launched from a U.S. Navy asset platform. These phases, as mentioned in the solicitation, are generally search, detect, track, classify, localize, home, and prosecute. Owing largely to the complexities of the underwater acoustic environment, the HWT must operate with an intrinsic information deficit. Such challenges make detection, localization, and classification of slow-moving (low-Doppler) contacts difficult for all unmanned systems. MIKEL has successfully employed advanced beamformers such as the Basis Pursuit De-Noising (BPDN) beamformer to provide highly-resolved localization estimates of sonar contacts. The BPDN is applied in this work to separate closely-spaced, slow-moving, contacts in the context of that found with the HWT. For the HWT and other unmanned systems, the BPDN can provide superior localization performance to improve the classification of low-Doppler contacts.
Benefit: The proposed effort herein anticipates significant benefits to the U.S. Navy by providing improved localization, and subsequently classification, of slowly-moving, closely-spaced contacts on unmanned systems such as the Heavy Weight Torpedo (HWT). MIKEL proposes a relevant super-resolution Basis Pursuit De-Noising (BPDN) beamformer which has successfully employed on prior sonar data.
Keywords: Basis Pursuit, Basis Pursuit, low-Doppler