Onboard Analytics Middleware for Aircraft Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Aircraft systems face challenges in handling the increased data transmission and processing demands of Aviation Internet of Things (AIoT), as they are not designed to manage the elevated data processing requirements, with limited onboard computer server capacities and bandwidth constraints.
Innovation Solution
The implementation of an onboard analytics system utilizing a middleware layer with an orchestrator that sends application containers to cabin and crew network smart members, each equipped with an edge member daemon, which manages resource allocation to perform analytics without interfering with primary functions, thereby leveraging underutilized resources and prioritizing primary functionality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data processing capacity is increased to handle AIoT demands, then productivity is improved, but device complexity increases
Solution Approach 1:
The system segments data processing across multiple distributed smart members (e.g., sensors, actuators, controllers) throughout the aircraft. Each smart member runs containerized analytics applications independently, dividing the overall processing load into manageable units that can operate autonomously while contributing to fleet-level analytics.
Solution Approach 2:
The patent implements a universal container runtime platform that can execute multiple different analytics applications across diverse smart members. The same infrastructure (orchestrator, container runtime, middleware) serves multiple functions: deploying applications, managing resources, handling data transmission, and coordinating analytics across different device types and locations.
2Productivity
If onboard computer server capacities are increased to handle AIoT data, then productivity is improved, but weight increases
Solution Approach 1:
The system enables smart members to perform analytics processing locally using their own embedded computing resources rather than relying on centralized onboard servers. Each smart member autonomously executes containerized applications and processes data locally, then transmits only results or aggregated data to the ground, eliminating the need for additional heavy onboard server infrastructure.
Solution Approach 2:
The patent extracts data processing functionality from traditional centralized server architectures and distributes it to edge devices (smart members) throughout the aircraft. By moving analytics capabilities to the edge where data is generated, the system eliminates the need for additional onboard server hardware while maintaining processing capacity.
3Productivity
If more smart members are joined to the middleware layer for analytics, then productivity is improved, but resource availability for primary functions decreases
Solution Approach 1:
The system dynamically manages smart member states, allowing members to transition between ready, joined, and busy states based on real-time resource availability and analytics demands. The orchestrator monitors resource usage and dynamically assigns or removes containerized applications from smart members, optimizing the balance between analytics processing capacity and primary function resource requirements.
Solution Approach 2:
When smart members complete analytics tasks or transition to busy states, the system discards active container instances and recovers computing resources for primary aircraft functions. The orchestrator manages the lifecycle of containers, terminating them when no longer needed and reallocating resources to critical flight operations.
Data Source
AI summary
An aircraft is configured to perform onboard analytics using available onboard systems. The aircraft comprises a cabin and crew network core with a middleware layer comprising an orchestrator configured to send application containers to a number of cabin and crew network smart members to perform the onboard analytics; and the number of cabin and crew network smart members each having a respective edge member daemon, each edge member daemon configured to place a respective cabin and crew network smart member in a joined state with the middleware layer and release resources of the respective cabin and crew network smart member when it leaves a joined state and enters a busy state so that the edge member daemon does not interfere with a primary function of the respective cabin and crew network smart member.