Feedback Control Fault Detection Using Multi-Instance Output Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing feedback control systems face challenges in detecting and compensating for hardware faults, particularly random faults caused by environmental factors like neutron bombardment and electromagnetic interference, which can compromise system performance and safety, and existing methods incur significant resource and power overheads.
Innovation Solution
The system employs multiple instances of the same control program to execute simultaneously, with sensor data input to each instance and output comparisons made to detect hardware faults, using methods such as majority voting and machine learning to identify faulty outputs, while maintaining processor overhead similar to a single instance execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple instances of control program are executed simultaneously to detect hardware faults, then system reliability is improved, but processor overhead increases
Solution Approach 1:
The patent segments the control program into multiple independent instances that execute simultaneously. Each instance processes sensor data independently and generates control commands, allowing fault detection through comparison of outputs from different instances without requiring a single monolithic program to handle all processing
Solution Approach 2:
The patent implements periodic execution of multiple control program instances at different time periods. By alternating which instances execute during different time periods and comparing their outputs, the system achieves fault detection while managing processor load through periodic rather than continuous multi-instance execution
2Reliability
If redundant hardware blocks are used to detect and compensate for faults, then fault detection capability is improved, but area and resource overhead increase
Solution Approach 1:
The patent creates software copies of the control program rather than hardware copies. Multiple instances of the same control program are executed on the existing hardware processor, eliminating the need for redundant hardware blocks while achieving the same fault detection capability through software-based parallel execution
Solution Approach 2:
The patent replaces the mechanical/hardware-based fault detection system with a software-based system. Instead of using redundant hardware blocks, error correcting codes, or parity circuits, the system uses multiple software instances of the control program that can be executed on standard processors without additional hardware overhead
3Reliability
If software-based error detection techniques are implemented, then fault detection is improved, but computing overhead and implementation complexity increase
Solution Approach 1:
The patent creates a universal fault detection mechanism that works with any control program without requiring program-specific modifications. The same multi-instance execution and output comparison approach can be applied to different control programs, making the system universally applicable while reducing implementation complexity compared to specialized software-based error detection techniques
Data Source
AI summary
Systems and methods for detecting hardware faults in computer-based feedback control systems. Multiple instances of the system control program(s) are run on system processors. System sensor data are input to each instance, and the control commands output by each instance are compared. As instantiations of the same programs receive largely the same sensor data, differences between output commands may indicate the presence of one or more hardware faults.