Distributed Avionics Synchronization for Deterministic Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Avionics systems face challenges in achieving increased dependability and determinism in data traffic and processing, particularly in real-time systems where failure to meet bandwidth, data delivery time, and redundancy demands can be dangerous, especially in applications like 'brake-by-wire' or 'steer-by-wire', and certification costs for software development are substantial.
Innovation Solution
A distributed avionics system with interconnected processing subsystems that generate periodic processing and communication time frames, prioritize tasks based on predetermined schedules and processing rates, and use network interfaces for scheduled data packet transmission, ensuring deterministic and robust real-time processing with low jitter and high system safety.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple applications share a single processor using APEX partitioning, then space utilization and modularity are improved, but certification complexity and cost increase
Solution Approach 1:
The system divides the processor into multiple virtual partitions using APEX (Application Partitioning Executive), where each partition operates as an isolated execution environment. This segmentation allows multiple applications to share physical hardware while maintaining independent certification boundaries, thus improving space utilization without proportionally increasing certification complexity.
Solution Approach 2:
The APEX partitioning mechanism acts as an intermediary layer between applications and the physical processor. It provides abstraction and isolation, allowing certified applications to run in partitioned environments without requiring re-certification when sharing hardware resources, thereby reducing overall certification burden.
2Ease of manufacture
If COTS technology is used instead of specialized avionic technology, then cost efficiency is improved, but system dependability and determinism may deteriorate
Solution Approach 1:
The system changes the operational parameters of COTS processors by implementing strict time-triggered scheduling and deterministic communication protocols. This transforms generic commercial processors into reliable avionic-grade computing platforms, achieving cost efficiency without sacrificing dependability through parameter-controlled operation.
Solution Approach 2:
The system employs periodic time-triggered task execution and communication cycles, where all processing and data transmission occur at predetermined intervals. This periodic action ensures deterministic behavior from COTS technology, meeting real-time requirements while maintaining cost efficiency through standard commercial components.
3Manufacturing precision
If time-triggered scheduling with synchronized processing frames is implemented, then real-time determinism is improved, but system complexity increases
Solution Approach 1:
The system merges communication scheduling and processing scheduling into a unified time-triggered framework. By synchronizing data transmission and task execution to the same global time base, the system reduces the number of separate synchronization mechanisms needed, improving real-time determinism while managing complexity through integration.
4Reliability
If bandwidth and data delivery time demands are increased for safety critical control, then control reliability is improved, but system resource requirements and complexity increase
Solution Approach 1:
The system implements continuous periodic transmission of critical data within synchronized time frames, ensuring uninterrupted data flow for safety-critical control. This continuous action guarantees reliable data delivery without requiring excessive peak bandwidth, as the same data rate is maintained consistently over time.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A distributed avionics system arranged in an aerial vehicle (1) for controlling at least one avionics function, the system comprising: a plurality of avionics processing subsystems (S 1-S4) interconnected in a network (2), each avionics processing subsystems (S1-S4) comprising: processing means (4), arranged to process at least one task (53, 55-56, 62-63) so as to provide a set of data messages comprising one or more units of data related to controlling the at least one avionics function on basis of provided input data, related to conditions of the at least one avionics function, memory means (3), in operative connection with the processing means (4) arranged to store instructions for the processing means (4) to process said at least one task, wherein at least one of the avionics processing subsystems (S 1-S4) is arranged to transmit at least one synchronization cycle start signal onto the network (2).