Onboard Runway Incursion Detection Using Sensor Fusion and Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current airport runway incursion detection systems rely heavily on human assessment and ADS-B data, which is unreliable and environment-dependent, leading to potential catastrophic incidents due to the dynamic nature of airport environments and the inability of all vehicles to report their positions accurately.
Innovation Solution
An onboard system that uses multiple data sources, including visual sensing, ADS-B, and sensor fusion, to automatically detect and report runway incursions with confidence, utilizing convolutional neural networks and object tracking to enhance situational awareness for pilots and autonomous systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If ADS-B system is used for runway incursion detection, then real-time position data can be obtained, but the system is unreliable and subject to erroneous reporting
Solution Approach 1:
The patent combines multiple detection systems (ADS-B, ground radar, visual sensing) into a unified detection framework. By merging these diverse data sources, the system achieves both real-time detection capability and improved reliability through cross-validation of multiple independent sources, thereby resolving the contradiction between speed and reliability.
Solution Approach 2:
The patent introduces an intermediary verification layer that processes ADS-B data through multiple validation mechanisms including ground radar cross-checking and visual sensing confirmation. This intermediary layer filters out erroneous ADS-B reports while maintaining real-time detection capability, thus resolving the reliability issue without sacrificing detection speed.
2Reliability
If ground radar-based detection system is deployed, then detection coverage can be improved, but infrastructure dependence increases
Solution Approach 1:
The patent designs a multi-functional detection system that can operate with ground radar infrastructure when available, but seamlessly transitions to using only onboard sensors (visual sensing, ADS-B) when ground infrastructure is unavailable. This universal approach provides comprehensive detection coverage while reducing infrastructure dependence through flexible operational modes.
Solution Approach 2:
The system dynamically adjusts its operational mode based on the availability of ground radar infrastructure. When ground radar is available, the system utilizes it for enhanced coverage; when unavailable, it automatically relies on onboard sensors. This dynamic adaptation resolves the contradiction by making infrastructure dependence conditional rather than fixed.
3Loss of information
If human assessment is used for runway incursion detection, then situational understanding can be achieved, but response time is insufficient for high-speed operations
Solution Approach 1:
The patent replaces human cognitive assessment with automated sensor fusion algorithms and machine learning models that process multi-source data (ADS-B, ground radar, visual sensing) to achieve comprehensive situational understanding. This substitution eliminates human response time limitations while maintaining or improving situational awareness through faster computational processing of the same information.
Solution Approach 2:
The system implements real-time feedback loops where detection results from multiple sources are continuously cross-validated and processed. This automated feedback mechanism provides comprehensive situational understanding instantaneously, replacing slow human assessment while maintaining the depth of situational awareness through continuous multi-source data integration.
4Measurement precision
If multiple detection sources are integrated, then detection accuracy can be improved, but system complexity increases
Solution Approach 1:
The patent segments the complex multi-source detection system into modular functional components: data acquisition modules (ADS-B receiver, ground radar interface, visual sensing), data processing modules (sensor fusion algorithms, machine learning models), and decision-making modules (incursion detection, alert generation). This segmentation improves detection accuracy through specialized processing while managing complexity through modular architecture that allows independent development and testing of each component.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
An on-board system and method for leveraging multi-modal means of detecting runway incursions to increase aircraft safety, and to provide a necessary enabler for both reduced crew operations and autonomous flights in the future. The system and method detect incursions using visual perception for all types of vehicles and objects that may incur, and an ADS-B for self-reporting vehicles. The system and method fuse prior known information (such as ownship pose and map data) with detections from sensors and other sources. The system actively tracks detections and computes the current and predicted future states to decide if there is an on-going incursion on the runway. The system reports the runway state to the pilots and/or autonomous receivers.