UAV Navigation Control Against GPS Spoofing and Hijacking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned aerial vehicles (UAVs) are vulnerable to electronic attacks such as GPS spoofing, RF blocking, and laser attacks, which can disrupt their navigation and control, leading to potential damage or collateral damage, and existing systems lack effective mechanisms to detect and counter these threats in real-time.
Innovation Solution
Equipping UAVs with a neural network-based system that analyzes real-time image data from onboard cameras to detect and respond to attacks, allowing the drone to adapt its flight path and maintain safe operation, and collaborating with high-altitude pseudosatellite platforms to enhance communication and navigation resilience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS navigation is used for UAV flight path control, then navigation accuracy is improved, but the system becomes vulnerable to GPS spoofing attacks and hijacking
Solution Approach 1:
The patent introduces an intermediary neural network system that acts as a mediator between GPS signals and flight control. The neural network analyzes image data from onboard cameras to detect spoofing attacks and determines whether to trust GPS navigation data, thereby protecting the flight control system from malicious GPS signals while maintaining navigation accuracy when GPS is legitimate.
Solution Approach 2:
The system implements feedback by continuously monitoring image data from onboard cameras and comparing it with expected visual features along the flight path. When discrepancies are detected that indicate GPS spoofing, the feedback loop triggers alternative navigation methods using visual odometry and pre-mapped environmental features, ensuring reliable navigation even when GPS is compromised.
2Reliability
If real-time image analysis using neural networks is implemented, then attack detection capability is improved, but computational load and energy consumption increase
Solution Approach 1:
The neural network does not continuously analyze all image data at full computational intensity. Instead, it performs partial analysis by focusing on key visual features and only triggering full attack detection protocols when preliminary indicators suggest potential spoofing. This reduces energy consumption while maintaining effective attack detection capability.
Solution Approach 2:
The system performs preliminary actions by pre-processing and pre-analyzing image data during normal flight conditions before attacks occur. Visual features and environmental landmarks are pre-mapped and stored, allowing the neural network to quickly compare real-time images against expected patterns without requiring intensive real-time computation during critical attack detection moments.
3Productivity
If preprogrammed flight paths are used, then mission execution efficiency is improved, but adaptability to unplanned changes and GPS drift is reduced
Solution Approach 1:
The flight path system transitions from static preprogrammed paths to dynamic adaptive paths. The neural network continuously monitors visual feedback from onboard cameras and dynamically adjusts the flight path in real-time to account for GPS drift, unexpected obstacles, and environmental changes, while still maintaining efficient mission execution by building upon the original preprogrammed trajectory where applicable.
Solution Approach 2:
The UAV performs self-service by autonomously detecting and correcting its own navigation errors using visual odometry and environmental feature recognition. When GPS drift occurs or obstacles are detected, the system independently calculates corrective flight path adjustments without requiring external intervention, maintaining both efficiency and adaptability.
Data Source
AI summary
An unmanned aerial vehicle (UAV) or “drone” executes a neural network to assist with detecting and responding to attacks. The neural network may monitor, in real time, the data stream from a plurality of onboard sensors during navigation and may communicate with a high-altitude pseudosatellite (“HAPS”) platform. For example, if the neural network detects a cyber-attack but determines that it does not interfere with external communications, it may shift navigation control of the drone to the HAPS.