Predictive Navigation Images for GPS-Denied Inertial Aiding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems, such as GPS and FMV, are vulnerable to spoofing and jamming in contested environments, making navigation difficult or impossible in environments with changing landscapes.
Innovation Solution
A predictive navigation system that generates time-sequenced and geo-sequenced images using machine learning, incorporating library and training data to predict future landscapes, which are then presented to the pilot or platform control system to aid navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS and FMV navigation aids are used, then navigation accuracy is improved, but vulnerability to spoofing and jamming increases
Solution Approach 1:
The patent introduces an offline-generated predictive image database as an intermediary between the navigation system and the pilot. This database contains pre-rendered images of future positions along the mission route, serving as a mediator that provides navigation guidance without requiring real-time vulnerability to GPS/FMV spoofing or jamming.
Solution Approach 2:
The system performs preliminary actions by pre-generating and storing predictive images of future mission route positions before the actual mission occurs. These images are created offline using terrain data and mission parameters, allowing the navigation system to reference pre-computed visual information rather than relying on real-time vulnerable signals.
2Reliability
If real-time navigation aids are used, then navigation reliability is improved, but susceptibility to electronic countermeasures increases
Solution Approach 1:
The predictive image database acts as a trusted intermediary that decouples navigation reliability from real-time electronic signals. By referencing pre-generated images stored locally, the system maintains reliable navigation guidance without exposing itself to electronic countermeasures that target real-time signal transmission.
Solution Approach 2:
The system creates copies of future navigation scenes in the form of predictive images stored in the offline database. These copied visual representations allow the pilot to navigate reliably by comparing actual views against pre-copied expected views, eliminating dependence on real-time vulnerable electronic navigation aids.
3Adaptability or versatility
If predictive images are generated using machine learning models, then adaptability to changing landscapes is improved, but computational complexity increases
Solution Approach 1:
The system performs the computationally intensive machine learning image generation in advance, before the mission. The offline generation of predictive images transfers the computational burden to pre-mission processing, allowing the actual mission system to use simpler storage and retrieval operations while still benefiting from adaptive landscape predictions.
Solution Approach 2:
The navigation problem is segmented into offline preparation phase and online execution phase. The complex adaptive image generation is separated and performed offline, while the online phase uses simpler processes of loading and comparing pre-generated images, reducing real-time computational complexity while maintaining adaptability.
Data Source
Figure 1
Figure 2A~2B-4
Figure 3A~3D
AI summary
A method for generating a predictive navigation system includes defining, based on a mission tasking, a mission route, identifying multiple mission planning images that correspond to a predetermined mission route, identifying geospatial data corresponding to the mission planning images, providing to a model, a mission plan, the mission planning images, and the geospatial data. The model is operated to generate predicted time sequenced and geo-sequenced images that are , provided to a platform control system. The predicted time sequenced and geo-sequenced images are presented during the mission to a pilot in real mission time. The mission plan includes multiple location markers that correspond to locations along the mission route. The time sequenced and geo-sequenced images depict landscapes along the mission plan corresponding to the mission plan. The predicted time sequenced and geo-sequenced images are provided to the platform control system before a mission.