Fly-By-Wire Redundancy Management via Segmented Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fly-by-wire (FBW) systems face challenges in mitigating generic fault effects due to the complexity of components and communications paths, making it difficult to implement dissimilar redundancy effectively, which is critical for ensuring system integrity and availability, especially in highly critical functionalities like aircraft flight control systems.
Innovation Solution
The implementation of a control system with a common processing partition and two distinct processing partitions (normal mode and direct mode) that operate in conjunction with a simple common partition to detect failures and switch to a backup mode when necessary, using encoded data packets for error detection and validation, thereby architecturally mitigating fault effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If dissimilar redundancy is implemented to mitigate generic fault effects, then system reliability is improved, but device complexity increases due to the need for multiple computing lanes with different designs
Solution Approach 1:
The system is divided into multiple computing lanes (first computing lane, second computing lane) with distinct architectural designs. Each lane processes flight control commands independently through segmented functional blocks (e.g., different filter implementations, distinct gain calculation paths), enabling fault isolation while maintaining redundancy. This segmentation allows the system to detect and bypass failed lanes without compromising overall reliability.
Solution Approach 2:
Different computing lanes implement locally optimized algorithms tailored to specific flight control functions. For example, one lane may use a first-order filter while another uses a second-order filter for the same control surface. This local quality differentiation enhances fault detection capability while managing complexity through function-specific optimization rather than universal redundancy.
2Reliability
If multiple computing lanes are used to detect electronic failures, then system availability is improved, but the difficulty of detecting and measuring faults increases due to complex communications paths
Solution Approach 1:
The system incorporates feedback mechanisms where each computing lane's output is monitored and compared against expected parameters. Validation logic in each lane provides feedback on its own operational status, and cross-validation between lanes detects discrepancies. This feedback structure simplifies fault detection by providing continuous status information without requiring complex external monitoring of communication paths.
Solution Approach 2:
A common processing partition serves as an intermediary that receives inputs from multiple computing lanes and coordinates their outputs. This intermediary structure simplifies fault detection by centralizing the comparison and validation logic, allowing the system to identify failed lanes through the intermediary's coordination function rather than through complex direct comparisons of all communication paths.
3Productivity
If complex hardware devices are used to enhance processing capability, then productivity is improved, but implementing dissimilar redundancy becomes more difficult
Solution Approach 1:
Complex processing functions are segmented into distinct modular blocks within each computing lane (e.g., separate filter modules, gain calculation modules, saturation modules). This segmentation allows dissimilar redundancy to be implemented at the module level rather than requiring complete architectural differences, simplifying the implementation of redundancy while maintaining high processing capability through modular complexity.
Solution Approach 2:
The computing lanes are designed with universal interfaces and common data structures that allow different architectural implementations to work together. Despite internal differences in processing algorithms and hardware configurations, the lanes share common input/output protocols and validation mechanisms, enabling dissimilar redundancy without requiring completely separate system architectures.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods and systems are provided for redundancy management of a fly-by-wire avionics system. A control module for producing a control signal is provided comprising a common processing partition (24) for receiving a flight input signal and at least one first mode input signal, a first processing partition (22) coupled to the common processing partition and configured to receive the first mode input signals and flight input signal from the common processing partition, and a second processing partition (26) coupled to the common processing partition. The first processing partition (22) produces a first mode output signal in response to one of the first mode input signals and flight input signal. The second processing partition (26) generates a second mode signal in response to the flight input signal when the first processing partition (22) fails. The common processing partition (24) produces the control signal in response to one of the first mode output signal and second mode signal.