GPS Location Correction Using Machine Learning Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In areas with high-density tall structures, GPS signals are often obstructed, leading to inaccurate location determination due to unaccounted changes in signal travel time, causing GPS receivers to provide locations that can be off by hundreds of yards from the actual position.
Innovation Solution
A supervised machine-learning approach is used to correct GPS location data by training a model with raw location data and actual location data, accounting for factors like structure geometry, material properties, and weather, using a network architecture that includes a client device, server, and machine-learning model to process GPS data and environmental information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS signals are received directly from satellites, then location determination is simple and fast, but location accuracy deteriorates in areas with tall structures due to signal reflections and obstructions
Solution Approach 1:
The patent introduces an intermediary machine learning model that acts as a mediator between the GPS receiver and the final location determination. This model processes raw GPS signals and environmental data to correct location accuracy without requiring changes to the GPS receiver hardware itself, thus improving precision while maintaining relatively simple device architecture
Solution Approach 2:
The system performs preliminary actions by collecting environmental data about tall structures and training machine learning models in advance. This pre-processing of environmental information allows the system to compensate for signal reflections and obstructions before final location calculation, improving accuracy without adding complexity to the real-time GPS reception process
2Measurement precision
If machine learning models are used to correct GPS data, then location accuracy improves in obstructed environments, but computational complexity and data processing requirements increase
Solution Approach 1:
The patent segments the location determination process into multiple independent components: raw GPS signal reception, environmental data collection, machine learning model training, and correction application. This segmentation allows each component to be optimized independently and enables the use of pre-trained models that reduce real-time computational requirements
Solution Approach 2:
The system applies partial correction by using machine learning models to adjust only the portions of GPS data affected by environmental factors like tall structures. Rather than completely reprocessing all GPS data, the model applies targeted corrections to specific signal components, reducing overall computational complexity while maintaining improved accuracy
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
This method provides more accurate GPS location determination by accounting for complex signal transformations, improving navigation and location-dependent applications in urban and obstructed environments.
Implementation Method 1
GPS satellites broadcast radio signals providing their location, status, and time from on-board atomic clocks. A GPS receiver device receives the radio signals, noting the time of the GPS signals are received, and uses these times to calculate the distance from the GPS satellites.
Implementation Method 2
reflections of GPS signals off of tall structures may delay reception of the GPS signals by the GPS receiver device, as compared to the GPS signals being received directly by the GPS satellites
Data Source
AI summary
In one embodiment, a method includes accessing global positioning system (GPS) data indicating a raw location associated with the computing system; accessing, based on the raw location, environmental model data including one or more structures; generating GPS correction data by processing the GPS data and the environmental model data using a machine-learning (ML) model that has been trained to compensate inaccurate GPS readings due to environmental interference; and determining an accurate location by correcting the raw location using the GPS correction data.