OMEGA is a mobile bathymetric survey system designed to collect underwater depth measurements along a traveled path and reconstruct a mapped representation of submerged terrain from filtered spatial samples. It combines field sensing, real-time validation, structured logging, and post-processing into a single acquisition and mapping pipeline.
Not every field measurement is equally useful for mapping. Depth values must be interpreted in context, evaluated against position and motion quality, and filtered before they are used to reconstruct a surface. OMEGA preserves both the complete raw log of the survey and a reduced mapping dataset containing only higher-quality accepted points.
The platform operates in four stages:
This architecture allows the system to remain lightweight in the field while still supporting structured bathymetric reconstruction and geographic export in post-processing.
The purpose of the system is to produce a usable spatial survey record from a moving field platform. It is intended to:
OMEGA is a full measurement-to-map workflow.
The system is organized into four main functional layers.
The sensor layer gathers both mapping variables and measurement context.
Typical inputs include:
GPS
Depth sensor / sonar
DS18B20 temperature sensor
BNO085 IMU
These measurements are combined into structured observations that represent both the environment and the state of the survey platform.
The validation layer determines whether a recorded observation is suitable for mapping. It evaluates:
Its role is to prevent unreliable or redundant measurements from dominating the mapping dataset.
The logging layer writes two parallel outputs:
This separation preserves the full field record while creating a cleaner input for later reconstruction.
The reconstruction layer uses filtered spatial samples to generate:
Each logged record is treated as a structured measurement event rather than a single sensor reading.
A typical observation may include:
This structure makes it possible to judge depth measurements in context rather than in isolation.
The mapping dataset is ultimately reduced to samples of the form:
(lat, lng, depth)
These accepted points serve as the basis for reconstruction.
Additional fields are retained because they help determine whether a point should be trusted. These include GPS quality indicators, motion state, and environmental correction inputs.
The system follows a staged processing pipeline:
Sensor input → measurement assembly → real-time validation → raw log + filtered map log → post-processing and reconstruction → map outputs
Sensor readings are combined into one observation that associates:
Each observation is checked against quality constraints. A measurement may be logged but rejected from the mapping dataset if its conditions are poor.
The system writes:
Accepted mapping points are then used to produce visual and geographic outputs.
Real-time filtering is a central design feature of the system. It improves mapping quality by screening measurements as they are collected instead of waiting until the end of the survey.
A point is only eligible for mapping when position quality meets defined requirements, such as:
This helps ensure that accepted depth samples are tied to credible locations.
The platform’s motion state affects both measurement geometry and spatial sampling quality. The system evaluates:
This reduces the likelihood that unstable measurements become control points in the reconstructed surface.
Depth values are tested for:
These checks suppress noisy readings while preserving actual terrain variation.
A minimum spacing rule is used to prevent accepted mapping points from clustering too densely. This improves point distribution and reduces oversampling of nearly identical locations.
OMEGA writes two CSV datasets with different roles.
dataXX.csvThis is the complete acquisition log. It preserves the full field record and may include:
This file supports review, debugging, and future reprocessing.
mapXX.csvThis is the filtered mapping dataset. It contains only the observations that pass the acceptance logic and are intended for surface reconstruction.
This structure allows the system to:
The raw log preserves everything the system saw. The mapping file preserves what the system judged suitable for map generation.
Depth is treated as a measured signal that may require refinement before reconstruction.
A moving average filter is used to reduce short-term noise and isolated spikes in the depth series. This improves stability and reduces the visual impact of transient sensor noise.
Because acoustic depth sensing depends on the speed of sound in water, and that speed changes with temperature, the system applies a temperature-based correction using measured water temperature. This improves depth consistency across varying conditions.
The reconstruction stage begins with filtered spatial samples:
(lat, lng, depth)
These samples are irregularly spaced and do not directly define a continuous terrain surface.
A moving survey platform only measures specific locations along its path. The bottom surface must therefore be estimated between recorded points.
OMEGA reconstructs the surface using Inverse Distance Weighting (IDW). In this method:
IDW is a practical choice because it works well with irregular point spacing and is easy to interpret.
The reconstructed surface should therefore be interpreted as a modeled estimate constrained by accepted field samples.
The system supports multiple output types.
These preserve the complete acquisition history and the filtered mapping dataset.
Scatter maps display accepted points directly. They are useful for:
These produce a continuous estimated depth surface from accepted spatial samples.
Contour maps show depth zones and transitions, making bottom structure easier to read.
These include:
The system is affected by several interacting forms of uncertainty.
Depth sensing may be influenced by:
Position may be affected by:
Motion affects both geometry and sample distribution through:
Measurements may also be affected by:
Interpolation introduces assumptions about the behavior of unsampled areas and depends heavily on the spatial distribution of accepted points.
Several design choices shape the system.
Filtering in the field improves the quality of the mapping dataset early and reduces later cleanup.
This preserves both traceability and flexibility. The raw file keeps the complete field history, while the filtered file remains focused on mapping.
Map generation is deferred until after acquisition so the logging system remains lightweight and reconstruction settings can be tuned afterward.
IDW provides a practical balance between simplicity, interpretability, and computational cost for this kind of irregular mobile survey data.
The current system has several important limits:
These limitations define the system’s intended scope.
Possible future directions include:
OMEGA is a structured bathymetric acquisition and reconstruction system built around the distinction between measurement and mapping-quality measurement. Its core strength is the integration of real-time filtering, dual-file logging, and post-processed surface generation into a lightweight field workflow.
Rather than treating every observation as equally trustworthy, the system preserves the full record while selectively constructing a cleaner mapping dataset for reconstruction. That separation between acquisition, validation, logging, and surface generation is the core architectural principle of the platform.