Aircraft Identification With Simulated Lidar Point Clouds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing aircraft identification systems, such as video detection and guidance systems, struggle to accurately distinguish between different aircraft types and sub-types, leading to inefficiencies and safety risks in airfield operations, while 3D Lidar sensors are expensive and data-intensive.
Innovation Solution
A computing device simulates virtual Lidar sensor data for 3D aircraft models to generate point clouds, using a classification model to identify aircraft types and sub-types, and determines their pose and location without relying on 3D Lidar sensors, enabling efficient and safe airfield navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video detection and guidance systems are used for aircraft identification, then the system is simple and low-cost, but the accuracy in distinguishing between different aircraft types and sub-types is insufficient
Solution Approach 1:
The patent uses 2D images from existing video cameras as copies of the 3D aircraft structure, processing these 2D copies to extract aircraft type and sub-type information without requiring actual 3D Lidar sensors. This allows the system to achieve high measurement precision while avoiding the complexity and cost of 3D sensing hardware.
Solution Approach 2:
The patent replaces the mechanical/optical 3D Lidar sensing system with a computational approach using 2D image processing and machine learning algorithms. This substitution maintains or improves identification accuracy while significantly reducing device complexity and cost.
2Measurement precision
If 3D Lidar sensors are used for aircraft identification, then the identification accuracy is high, but the cost and data processing requirements are expensive and intensive
Solution Approach 1:
The system creates computational 3D representations (point clouds) from 2D image copies rather than using actual 3D Lidar data. This copying approach maintains the ability to perform accurate 3D analysis while working with smaller, more manageable data sets from standard video cameras.
Solution Approach 2:
The patent extracts only the essential features needed for aircraft identification from the 2D images, rather than capturing and processing complete 3D spatial data. This extraction approach reduces data volume while maintaining identification accuracy by focusing on discriminative features.
3Measurement precision
If 3D Lidar sensors are used for aircraft identification, then the identification accuracy is high, but the system cost is expensive
Solution Approach 1:
The system uses copies of aircraft views from inexpensive 2D video cameras instead of requiring expensive 3D Lidar hardware. This copying strategy enables high-precision identification using low-cost, widely available imaging devices.
Solution Approach 2:
The patent replaces expensive, complex 3D Lidar systems with cheap, readily available 2D video camera systems. The computational processing compensates for the lower hardware quality, achieving high accuracy with low-cost components.
Data Source
AI summary
Methods, devices, and systems for aircraft identification are described herein. In some examples, one or more embodiments include a computing device comprising a memory and a processor to execute instructions stored in the memory to simulate virtual light detection and ranging (Lidar) sensor data for a three-dimensional (3D) model of an aircraft type to generate a first point cloud corresponding to the 3D model of the aircraft type, generate a classification model utilizing the simulated virtual Lidar sensor data of the 3D model of the aircraft type, and identify a type and/or sub-type of an incoming aircraft at an airport by receiving, from a Lidar sensor at the airport, Lidar sensor data for the incoming aircraft, generating a second point cloud corresponding to the incoming aircraft utilizing the Lidar sensor data for the incoming aircraft, and classifying the second point cloud corresponding to the incoming aircraft using the classification model.