Air Corridor Models for Safe Separation of Mixed Aerial Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The concurrent use of airspace by Advanced Aerial Mobility (AAM) vehicles, conventional aircraft, and small Unmanned Aircraft Systems (sUAS) poses challenges in maintaining safe separation and efficient operation, as existing systems lack scalable and flexible solutions for managing diverse aerial vehicles in shared airspaces.
Innovation Solution
The implementation of a Multi-Agent Reinforcement Learning (MARL)-based system that trains an air corridor model using simulated vehicles to determine safe and efficient routes, allowing aerial vehicles to maintain safe separation distances by dividing airspace into spatio-temporal data cubes and using an Octree data structure for efficient processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional air traffic management systems are used to manage diverse aerial vehicles, then existing infrastructure can be maintained, but safe separation and efficient operation cannot be ensured in shared airspaces
Solution Approach 1:
The airspace is segmented into discrete three-dimensional grid cells that are organized into corridors. Each cell can be independently managed and assigned to specific vehicle types or operations, allowing diverse aerial vehicles to operate safely in shared airspaces while maintaining proper separation through the structured grid framework
Solution Approach 2:
The air corridor system provides a universal framework that can accommodate multiple types of aerial vehicles including AAM vehicles, conventional aircraft, and sUAS. The same corridor infrastructure and management system serves all vehicle types, enabling efficient shared airspace utilization while maintaining safety through type-specific corridor assignments
2Productivity
If air corridors are established for AAM vehicles, then efficient travel routes are provided, but complex coordination is needed to maintain safe separation distances
Solution Approach 1:
By segmenting airspace into discrete grid cells within corridors, the system simplifies coordination complexity. Each cell serves as an independent management unit with clear entry and exit rules, allowing vehicles to efficiently navigate through predefined corridors while the segmented structure automatically enforces separation without requiring complex real-time coordination between all vehicles
Solution Approach 2:
The air corridor model acts as an intermediary between air traffic management and individual vehicles. It provides predefined efficient travel routes through the corridors while automatically handling separation enforcement through the grid cell structure, reducing the coordination burden on individual vehicles and simplifying the overall system
3Adaptability or versatility
If traditional air traffic management is used, then current operational procedures are maintained, but scalability to accommodate vast volume of aerial vehicles is limited
Solution Approach 1:
The segmented grid cell structure provides inherent scalability. As the number of aerial vehicles increases, new grid cells can be activated and new corridors can be defined without fundamentally changing the operational framework. This modular approach allows the system to scale from current low-volume operations to future high-volume AAM scenarios while maintaining operational simplicity through consistent cell-based procedures
Solution Approach 2:
The air corridor system is dynamic and adaptable to changing traffic volumes and vehicle types. Corridors can be dynamically created, modified, or deactivated based on operational needs, and the grid cell assignments can be adjusted to accommodate varying traffic patterns, enabling the system to scale efficiently while maintaining ease of operation through consistent underlying principles
Data Source
AI summary
The disclosure herein describes use of an air corridor model to enable aerial vehicles to travel an air corridor while maintaining safe separation distance from other vehicles. A trained air corridor model associated with an air corridor is received by an aerial vehicle, wherein the trained air corridor model is configured to maintain safe separation distance between vehicles in the air corridor. The current position of the aerial vehicle in the air corridor is identified and positions of other agents in the air corridor are received. A next position of the aerial vehicle is determined based on the trained air corridor model, the current position, and the agent positions. The aerial vehicle is controlled to travel from the current position to the determined next position, whereby the aerial vehicle and other agents are enabled to efficiently use the airspace of the air corridor while maintaining safe distances between each other.