Airport Surface Data Fusion for Runway Incursion Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for detecting runway incursions face challenges in obtaining sufficient and accurate training data for machine learning models, which affects their ability to detect incursions with high accuracy.
Innovation Solution
The proposed solution involves an airport object location system that includes vehicle location units connected to vehicles, an electro-optical sensor system connected to aircraft, and a model generator. This system generates vehicle location information and timestamps, which are correlated with video frames from the aircraft using reference timestamps and location information to form a dataset for training machine learning models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sensors and data correlation methods are used to improve detection accuracy, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent combines multiple sensor types (electro-optical sensors, induction coils, microwave position sensors, radar systems, ADS-B) into a single integrated airport surface detection system. These diverse sensors are merged to provide comprehensive coverage of the airport surface, combining their respective strengths to achieve high detection accuracy while managing system complexity through unified processing.
Solution Approach 2:
The system employs multi-functional sensors that can perform multiple detection tasks. For example, radar systems can detect both aircraft and vehicles, while electro-optical sensors can provide both visual surveillance and data for machine learning models. This multi-functionality reduces the need for separate specialized sensors, thereby managing complexity while maintaining high detection precision.
2Reliability
If comprehensive sensor data collection is implemented to improve detection capability, then reliability improves, but loss of energy increases
Solution Approach 1:
The system employs periodic scanning and sampling rather than continuous monitoring of all sensors at full capacity. Sensors are activated and deactivated in periodic cycles based on operational needs, allowing the system to maintain reliable detection capability while significantly reducing energy consumption during low-activity periods.
Solution Approach 2:
The system activates only the necessary subset of sensors based on current operational conditions and detected threats. Rather than running all sensors continuously, the system applies partial action by selecting and activating only those sensors needed for current detection tasks, thereby maintaining reliability while minimizing energy loss.
Data Source
AI summary
An airport object location system comprising a number of vehicle location units, a sensor system, and a model generator. The number of vehicle location units is connected to a number of vehicles. The number of vehicle location units generate vehicle location information for the number of vehicles in an area including an operations surface at an airport and vehicle timestamps for the vehicle location information. The sensor system is connected to a reference vehicle. The sensor system is configured to generate sensor data for the area, wherein reference timestamps and reference location information are associated with the sensor data. The model generator is configured to correlate the vehicle location information for the vehicles with the sensor data using the vehicle timestamps, the reference location information, and the reference timestamps to form a dataset.