Aircraft Control Authority Blending for Safe AI Transition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Developing aircraft control systems with artificial intelligence poses challenges in ensuring safety and reliability, as existing techniques require significant trust before they are considered safe, and pilots and regulatory bodies may be hesitant to transition from proven control systems to new AI-based systems.
Innovation Solution
An aircraft control system that integrates a trained classifier module with a traditional control module, using an authority parameter to gradually transfer control based on performance evaluation, allowing incremental adjustment of the authority between the two modules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a trained classifier module is introduced to control the aircraft, then the productivity of control system development is improved, but the reliability is worsened due to lack of trust in AI techniques
Solution Approach 1:
The patent introduces an intermediary authority parameter that mediates between the traditional control module and the trained classifier module. This parameter acts as a trust intermediary, allowing the system to gradually transfer control authority from the proven traditional module to the AI-based classifier based on evaluated performance, thereby resolving the trust deficit while enabling productivity improvements
2Adaptability or versatility
If control is fully transferred to the trained classifier, then the adaptability of the control system is improved, but the stability is worsened due to pilot and regulatory hesitation
Solution Approach 1:
The authority parameter is designed as a dynamic variable that adjusts the control distribution between traditional and AI modules in real-time based on performance evaluation. This dynamic adjustment mechanism allows the system to adapt to different operational contexts while maintaining stability through continuous monitoring and incremental changes, preventing abrupt transitions that would cause instability
3Reliability
If the authority parameter is used to gradually transfer control, then the reliability is improved through performance evaluation, but the device complexity is worsened
Solution Approach 1:
The patent employs parameter changes by modifying the authority parameter based on performance metrics. Instead of adding complex structural components, the system achieves reliable controlled transfer by dynamically adjusting this single key parameter, which determines the weight or influence of each control module's outputs, thereby improving reliability without proportionally increasing device complexity
Data Source
AI summary
An aircraft control system (100) including an aircraft control module (110), a trained classifier module (120), and an aircraft control processing engine (13). The aircraft control module (110) generates first control outputs (104a to 104c) based on received aircraft operating inputs (102a to 102d). The trained classifier module receives the aircraft operating inputs (102a to 102d) and generates second control outputs (104d to 104f). The aircraft control processing engine (130) receives the first control outputs (104a to 104c) and the second control outputs (104d to 104f) and generates operating control outputs (106a to 106c), based on the received first control outputs and second control outputs. The aircraft control processing engine (130) then controls the aircraft using the operating control outputs (106a to 106c).