A study published in Satellite Navigation presents a novel real-time sound speed profile (SSP) correction scheme that significantly improves the accuracy of underwater navigation for autonomous and remotely operated vehicles. The research, conducted by scientists from collaborating institutions, addresses a critical challenge in deep-sea operations: the degradation of acoustic positioning due to temporal and spatial variations in seawater sound speed.
Underwater navigation systems typically rely on fusion of Strap-down Inertial Navigation Systems (SINS) and Ultra-Short Baseline (USBL) acoustic positioning, as satellite signals cannot penetrate water. However, sound speed varies with temperature, salinity, and pressure, causing refraction that introduces systematic errors in travel-time and angle measurements. Pre-measured SSPs quickly become outdated during long missions, leading to accumulating navigation drift. Traditional correction methods, such as static conductivity-temperature-depth (CTD) profiler measurements or empirical models, fail to adapt to real-time conditions.
The proposed method models temporal SSP variability using acoustic ray-tracing theory based on Snell's law, deriving partial differential relationships between sound-speed disturbance and horizontal/vertical displacements. A quasi-observation model estimates SSP perturbation by comparing SINS-derived and USBL-measured travel times. A two-order SSP disturbance representation separates the shallow-water mixed layer, thermocline, and deep isothermal layer to reflect realistic depth-dependent sound-speed distribution.
To fuse navigation data, the researchers designed an Adaptive Two-stage Information (ATI) filter that integrates SINS, Doppler Velocity Log (DVL), Pressure Gauge (PG), and USBL observations. The filter updates position, velocity, and attitude errors while simultaneously detecting USBL anomalies through a Generalized Likelihood Ratio test and refining SSP estimation via recursive least squares. Simulations using CTD datasets from the MVP system showed that without correction, USBL horizontal positioning errors reached several meters; with the algorithm, RMS error dropped markedly.
Sea trials in the South China Sea demonstrated RMS position improvement from 0.45 m to 0.08 m northward and 0.23 m to 0.07 m eastward, enhancing precision by over 80%. According to the authors, "Traditional navigation often depends on static sound speed profiles, which quickly become outdated during long missions. Our model integrates physical ray-tracing with adaptive filtering, enabling ARVs to sense and correct sound-speed changes rather than rely on fixed inputs."
This SSP correction framework provides a practical path toward self-adaptive deep-sea navigation systems, reducing dependence on external CTD surveys and improving resilience to acoustic distortion. The method is well-suited for autonomous remotely operated vehicles (ARVs) and autonomous underwater vehicles (AUVs) performing seabed mapping, ecological monitoring, mineral exploration, under-ice routing, or long-range autonomous missions. Further developments could integrate machine-learning-based SSP prediction or multi-sensor oceanographic data for proactive correction, with potential to improve efficiency and data reliability in future deep-sea exploration and marine resource assessment.
The study was published in Satellite Navigation (DOI: 10.1186/s43020-025-00181-w). Funding was provided by the National Natural Science Foundation of China, National Key Research and Development Program of China, Shandong Province Natural Science Foundation, and other sources.
