Aircraft Actuator Failure Control With Sensor Reliability Mediation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle systems lack comprehensive situational awareness and robust control mechanisms, particularly in autonomous aircraft operations, leading to potential safety risks due to unreliable sensor inputs and pilot errors, as well as inadequate contingency planning for unexpected flight conditions.
Innovation Solution
A system and method that integrate advanced sensor redundancy, machine learning for input correlation and guidance determination, and contingency planning algorithms to enhance situational awareness, improve sensor reliability, and enable autonomous decision-making, including the ability to override pilot inputs and adapt to undesired flight conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor redundancy is increased to improve reliability, then system complexity increases
Solution Approach 1:
The system segments sensor inputs into multiple independent channels with different reliability levels, processing each separately before integration. This allows the system to handle redundant sensors without proportionally increasing overall complexity, as each segment can be managed independently through modular processing blocks.
Solution Approach 2:
An intermediary processing layer is introduced between raw sensor inputs and control decisions. This intermediary layer correlates multiple sensor inputs, detects inconsistencies, and resolves conflicts before passing processed information to control systems, thereby managing complexity while maintaining reliability through systematic input validation.
2Loss of information
If machine learning algorithms are used for input correlation and guidance determination, then situational awareness improves, but computational requirements and processing time increase
Solution Approach 1:
Machine learning models are pre-trained offline on extensive flight data to learn normal sensor correlations and anomaly patterns. During actual flight operations, the pre-trained models perform rapid inference rather than learning, significantly reducing processing time while maintaining high situational awareness through pattern recognition.
Solution Approach 2:
The system implements continuous feedback loops where machine learning algorithms process sensor correlations in real-time, compare predictions with actual measurements, and adjust guidance determinations dynamically. This feedback mechanism enables adaptive situational awareness that improves over time while maintaining real-time performance through iterative refinement rather than exhaustive computation.
3Reliability
If the system is designed to override pilot inputs for safety, then safety improves, but pilot control and ease of operation deteriorate
Solution Approach 1:
The system monitors pilot inputs and aircraft state continuously, using machine learning to detect when pilot actions would lead to unsafe conditions. When anomalies are detected, the system automatically adjusts control surfaces or provides corrective guidance without requiring pilot awareness or consent, effectively making the safety system self-regulating while maintaining nominal pilot authority during normal operations.
Solution Approach 2:
The system performs preliminary analysis of pilot inputs against learned safe operating patterns before allowing execution. When potential unsafe actions are predicted, the system pre-emptsively counteracts them through automated control adjustments or warnings, preventing harmful actions before they occur while maintaining seamless operation during normal flight conditions.
Data Source
AI summary
A method, preferably including: sampling inputs, determining aircraft conditions, and/or acting based on the aircraft conditions. A method, preferably including: sampling inputs, determining input reliability, determining guidance, and/or controlling aircraft operation. A method, preferably including: operating the vehicle, planning for contingencies, detecting undesired flight conditions, and/or reacting to undesired flight conditions. A system, preferably an aircraft such as a rotorcraft, configured to implement the method.