Geophysical Field Navigation With ML De-Noising and Map Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Global Positioning System (GPS) based navigation is unreliable in GPS-denied environments, necessitating alternative methods for accurate navigation.
Innovation Solution
Implementing geophysical field sensing-based navigation using magnetic and/or gravitational data, which involves real-time data processing with machine learning models to de-noise sensor data and integrate it with pre-surveyed maps for precise position estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GPS-based navigation is used, then navigation reliability is improved in open environments, but navigation fails completely in GPS-denied environments
Solution Approach 1:
The patent introduces geophysical field sensors (magnetometers, gravimeters) as intermediary devices that detect environmental fields to provide navigation capability when GPS is unavailable. These sensors act as mediators between the navigation system and the environment, enabling operation in GPS-denied scenarios through field-based position estimation.
Solution Approach 2:
The system transitions from relying on satellite-based electromagnetic signals to utilizing local geophysical field parameters (magnetic field strength, gravity acceleration). By changing the fundamental measurement parameters from space-based signals to earth-based physical fields, the system achieves adaptability across different operational environments.
2Adaptability or versatility
If geophysical field sensors are used for navigation, then navigation capability is maintained in GPS-denied environments, but measurement precision deteriorates due to platform and environmental noise
Solution Approach 1:
The patent extracts and removes noise components from geophysical measurements through dedicated noise modeling and subtraction. Platform-induced magnetic anomalies and environmental interference are identified and extracted from the raw sensor data, leaving only the relevant position-dependent field signatures for navigation calculations.
Solution Approach 2:
The system employs feedback mechanisms where measured geophysical fields are continuously compared against pre-surveyed map data, and position estimates are refined iteratively. The navigation filter uses residual errors between measured and predicted field values to correct position estimates, improving precision over time despite noisy measurements.
3Measurement precision
If real-time de-noising processing is implemented, then position estimation precision is improved, but computational complexity increases
Solution Approach 1:
The patent performs noise characterization and de-noising model training in advance, before real-time navigation operations. Platform magnetic signatures are pre-measured and stored, and de-noising algorithms are pre-configured based on known platform properties. This preliminary preparation significantly reduces the computational burden during real-time operation, as the system only needs to apply pre-computed corrections rather than perform complex analysis on-the-fly.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables real-time, accurate navigation by removing platform and environmental noise from geophysical data, reducing errors in position estimation, and facilitating map matching for precise location determination.
Implementation Method 1
receiving magnetic field data from a magnetometer sensor
Implementation Method 2
receiving gravitational field data from a gravimeter sensor
Data Source
AI summary
Disclosed are exemplary computer-implemented methods and systems for geophysical field sensing based navigation. One example of a computer-implemented method includes: receiving geophysical field data from at least one geophysical field sensor; synchronizing timing of the geophysical field data; de-noising, using a de-noising machine learning model, the geophysical field data removing noise from local sources of noise for the at least one geophysical field sensor to produce de-noised geophysical field data, the de-noising machine learning model trained using ground truth map data and training data corresponding to the ground truth map data; receiving map data from a geophysical map engine; performing error estimation by comparing the de-noised geophysical field data with the map data; and updating a position estimation based at least in part on the error estimation.