Machine Learning Model for Satellite Positioning Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing satellite-based positioning systems face challenges in accurately estimating device location in environments with challenging signal conditions, such as urban canyons, areas with dense foliage, or near structures, due to interference and signal blockage.
Innovation Solution
A machine learning model is generated and used to assist GNSS positioning by comparing GNSS position estimates with reference positioning system estimates, incorporating parameters like pseudorange and range rate errors, to improve accuracy in challenging signal environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional GNSS positioning is used in challenging signal environments, then the system complexity remains low, but the positioning accuracy deteriorates due to signal interference and blockage
Solution Approach 1:
The patent introduces a machine learning model as an intermediary between the GNSS receiver and the positioning output. This model processes raw GNSS measurements and environmental parameters to compensate for signal degradation, acting as a mediator that translates imperfect measurements into accurate position estimates without requiring changes to the fundamental GNSS infrastructure
Solution Approach 2:
The patent replaces traditional signal processing and filtering methods with a machine learning-based approach. Instead of relying on conventional algorithms to process GNSS signals, the system uses trained neural networks or other ML models to directly estimate position from raw measurements, substituting mechanical/mathematical signal processing with intelligent data-driven processing
2Measurement precision
If machine learning model is introduced to improve positioning accuracy, then the positioning precision improves in challenging environments, but the device complexity increases
Solution Approach 1:
The patent applies preliminary action by training the machine learning model offline using extensive datasets collected from various environments before deployment. The model is pre-trained to recognize patterns in signal degradation and environmental characteristics, so that during actual positioning operations, the pre-trained model can quickly process measurements without requiring real-time training or complex computations
Solution Approach 2:
The patent changes the parameters processed by the system by incorporating not only traditional GNSS measurements but also environmental parameters such as signal strength, multipath indicators, and contextual data from other sensors. This multi-parameter approach allows the machine learning model to better characterize the signal environment and compensate for degradation effects
3Reliability
If GNSS signals are used in urban canyons and dense foliage areas, then the system simplicity is maintained, but the reliability of positioning deteriorates due to signal interference
Solution Approach 1:
The patent implements feedback mechanisms where the machine learning model continuously monitors the quality of GNSS signals and environmental conditions, adjusting its processing approach based on detected signal degradation patterns. The system uses feedback from measurement quality indicators to dynamically adapt the positioning solution, improving reliability in challenging environments where traditional fixed algorithms fail
Data Source
AI summary
A device implementing a system for estimating device location includes at least one processor configured to receive an estimated position based on a positioning system comprising a Global Navigation Satellite System (GNSS) satellite, and receive a set of parameters associated with the estimated position. The processor is further configured to apply the set of parameters and the estimated position to a machine learning model, the machine learning model having been trained based at least on a position of a receiving device relative to the GNSS satellite. The processor is further configured to provide the estimated position and an output of the machine learning model to a Kalman filter, and provide an estimated device location based on an output of the Kalman filter.