Sensor Fault Detection via Virtual Signal Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern vehicle control systems face unpredictable results due to faulty sensors, and existing monitoring systems require redundant sensors, which are expensive and not always necessary.
Innovation Solution
A method and system that detect sensor faults by computing virtual values based on data from other sensors using a vehicle model, calculating residual differences, and generating control values from these virtual values when a fault is detected, allowing the system to function without the faulty sensor.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If redundant sensors are used for fault detection, then sensor fault detection capability is improved, but system cost increases
Solution Approach 1:
The patent creates virtual copies of sensor signals through mathematical modeling. Instead of using physical redundant sensors, the system generates virtual sensor signals by computing from a vehicle model and actual sensor data from other sensors. This virtual copying approach provides fault detection capability without requiring additional physical sensors, thereby resolving the contradiction between detection capability and system cost.
Solution Approach 2:
The patent replaces the mechanical approach of using redundant physical sensors with a computational/mathematical approach. The system uses vehicle dynamics models and signal processing algorithms to generate virtual sensor readings and detect faults, substituting physical hardware redundancy with intellectual/computational redundancy. This eliminates the need for additional sensor hardware while maintaining fault detection capability.
2Reliability
If a sensor fault occurs, then system reliability deteriorates, but with the invention the system can continue operating
Solution Approach 1:
The patent prepares compensatory measures in advance by continuously maintaining a vehicle model that can generate virtual sensor signals. When a sensor fault occurs, the system is already equipped with the mathematical framework and alternative signal sources needed to compensate for the faulty sensor. This beforehand preparation allows seamless transition to fault-tolerant operation, cushioning the impact of sensor failures on system reliability.
Solution Approach 2:
The patent introduces virtual sensor signals as an intermediary between the faulty physical sensor and the control system. When a sensor fails, the virtual signals act as a mediator that bridges the gap, allowing the control system to continue receiving necessary input data without directly relying on the faulty sensor. This intermediary approach maintains system operation continuity while isolating the fault's harmful effects.
Data Source
AI summary
Methods and systems are provided for detecting faults in a sensor and reconstructing an output signal without use of the faulty sensor. In one embodiment, a method includes: receiving, by a processor, sensor data indicating a measured value from a first sensor; receiving, by a processor, sensor data indicating measured values from a plurality of other sensors; computing, by a processor, virtual values based on a vehicle model and the sensor data from the plurality of other sensors; computing, by a processor, a residual difference between the measured value from the first sensor and the virtual values; detecting, by a processor, whether a fault exists in the first sensor based on the residual difference; and when a fault in the sensor is detected, generating, by a processor, a control value based on the virtual values instead of the measured value.