Single UAV 3D Localization Using 5G Reference Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current UAV-based localization systems face challenges in accurately locating victims in three-dimensional environments, such as those obscured by obstacles like snow, water, or debris, due to limitations in satellite visibility and communication infrastructure, and existing solutions often require multiple drones or complex coordination, which is costly and complex.
Innovation Solution
A single-UAV 3D cellular search and rescue solution leveraging 5G-NR technology estimates the location of user equipment using reference signals to compute distance and direction, employing machine learning algorithms for position correction and predicting UAV trajectories for efficient localization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple drones are used for localization, then localization accuracy is improved, but system complexity and deployment cost increase
Solution Approach 1:
The patent transitions from two-dimensional localization to three-dimensional localization by adding vertical position estimation. The system estimates 3D position using reference signals and signal processing techniques, enabling accurate localization in complex terrain and environments where 2D approaches fail, thus improving accuracy without requiring multiple drones
Solution Approach 2:
The patent replaces the mechanical approach of using multiple physical drones with a single drone equipped with advanced signal processing capabilities. By using reference signals and machine learning algorithms for position correction, the system achieves high accuracy without the mechanical complexity of coordinating multiple vehicles
2Area of stationary object
If GNSS is used for localization, then coverage area is large, but reliability decreases when obstacles block satellite visibility
Solution Approach 1:
The patent introduces an intermediary approach by using a single drone as a mobile relay between the user equipment and the localization system. The drone receives reference signals from the UE and processes them to determine position, eliminating the need for direct satellite visibility and maintaining reliability in obstructed environments
Solution Approach 2:
The patent makes the localization system universal by enabling it to function in both open skies and obstructed environments. The single-drone system with reference signals and machine learning correction can operate regardless of satellite visibility, providing consistent localization service across diverse scenarios
3Device complexity
If 2D localization approach is used, then system simplicity is maintained, but accuracy decreases in three-dimensional environments
Solution Approach 1:
The patent enhances the 2D localization approach by adding the vertical dimension to create a comprehensive 3D positioning capability. The system estimates azimuth, elevation, and distance angles to calculate three-dimensional position, enabling accurate localization of victims buried under snow, water, or debris where 2D methods fail
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 approach enables rapid and accurate 3D localization of victims within a single frame, achieving a median error of 1.1 meters, improving upon existing systems by simplifying the deployment and reducing the complexity of multi-drone coordination.
Implementation Method 1
estimating the distance and the angle between at least one user equipment (or at least one cluster) located around the specified region of interest and the UAV, by using at least one known reference signal transmitted by the UE and the at least one reference signal received by the UAV
Data Source
Figure 1
Figure 2~3B
Figure 3C~3D
AI summary
A computer implemented method, a system and computer programs for locating a user equipment using an unmanned aerial vehicle. The method comprises directing an UAV over a specified region of interest; estimating a distance and angles between at least one user equipment located around the specified region and the UAV by using a reference signal transmitted by the user equipment and the reference signal received by the UAV in a single time slot; computing an estimated position of the user equipment by defining a coordinate system using the estimated distance and angles; updating and correcting the computed estimated position of the user equipment using a machine learning algorithm, providing an updated and corrected position of the user equipment as a result; and defining a trajectory of the UAV including the vertical position around the user equipment by executing a prediction trajectory algorithm over the updated and corrected position of the user equipment.