Real‑Time Automated Hydrographic Seabed Survey Processing

Project overview and selected deliverables. References and demonstrations available on request.

Overview

To support a client conducting large-scale hydrographical survey operations, JAK Services designed and developed a component-based automated onboard hydrographic data-processing pipeline. Its purpose was to automatically and continuously generate a preliminary dataset closely approximating the final processed data, enabling rapid detection of areas requiring re-survey before the vessel had moved too far from the site.

Workflow integration

The pipeline was fully orchestrable and integrated seamlessly with—while complementing—existing workflows, such as EIVA’s Workflow Manager (WFM) for Multi Beam Echo Sounder (MBES) data. The orchestration application was designed to be payload-agnostic and could also be used with Side-Scan Sonar (SSS) and other sensor types.

Real-time data ingestion and preprocessing

The pipeline illustrated in the image above featured:

Automated sorting and WFM integration

Once generated—and after any corresponding ancillary data was processed—the SBD files were continuously sorted according to active site and area names, then automatically imported and organized into EIVA WorkFlow Manager (WFM) for linking with ancillary data, cleaning, and exporting as point-cloud files.

Automated QC and analysis outputs

These point-cloud files were then processed by the JAK Services–developed orchestration application, which executed command-line tools to:

GIS deliverables and visualization

The orchestration application then generated low-volume GIS deliverables—including DTM, grayscale and color shaded reliefs, track plots, slope maps, red/green density charts, coverage layers, gap polygons, and more—ready for transfer to a web portal for visualization and download.

Final deliverables

Finally, the same applications were used to export the final merged deliverables once the data had been quality-checked by an operator.

Technologies: Python, C#, C++, GIS, AI

Back to portfolio