Deep ocean long range underwater navigation with ocean circulation model corrections.

This article focuses on an underwater navigation algorithm that provides a "cold start" (CSA) geo-position for submerged vehicles using travel times from multiple acoustic sources, and an enhanced method called cold start with model (CSAM) that incorporates ocean general circulation model (GCM) corrections to improve accuracy. The CSA algorithm estimates position without prior information or surfacing, while CSAM refines this position by measuring travel time offsets between modeled and received acoustic arrivals using a 4D GCM, significantly reducing mean geo-position errors from 58 m to 25 m in the Philippine Sea experiment (PhilSea10). The study demonstrates that CSAM accuracy is primarily limited by the GCM’s travel time error and is largely insensitive to initial CSA errors, with simulations confirming that CSAM maintains consistent accuracy despite increasing ocean sound speed variability. These results suggest that integrating GCM-based corrections with acoustic navigation can provide reliable, low size-weight-power (SWaP) underwater positioning suitable for autonomous underwater vehicles and oceanographic sampling.