Vehicle Environment Sensor Fault Detection Using Gate Region References
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting malfunctions in environment sensors of motor vehicles are inadequate, particularly in the absence of a reference truth, leading to potential false alarms and safety risks during automated driving functions, and require extensive effort for comprehensive checks and recalibration.
Innovation Solution
A method involving predefined 'gate regions' along road networks where sensor data is compared against reference data to detect deviations, allowing for automatic malfunction detection and calibration of environment sensors like cameras, radars, and lidars, using artificial neural networks and image processing algorithms, with a threshold-based approach to minimize false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If self-diagnosis algorithms are used to detect sensor malfunctions, then detection capability is improved, but false alarm rate increases due to lack of reference truth
Solution Approach 1:
The patent introduces gate region reference data as an intermediary standard for comparison. Instead of relying solely on algorithms that guess whether a sensor is malfunctioning, the system compares sensor measurements against pre-established reference data from known gate regions (specific geographic locations with known characteristics). This intermediary reference truth enables precise verification of sensor functionality while minimizing false alarms.
Solution Approach 2:
The patent performs preliminary actions by pre-collecting and storing reference data during manufacturing or initial calibration phases. Gate regions are pre-defined with known object positions and characteristics, creating a baseline database before actual operation. This preliminary preparation enables accurate malfunction detection during normal operation without requiring complex real-time analysis.
2Measurement precision
If comprehensive sensor checks are performed in service workshops, then measurement precision is improved, but loss of time and operational disruption increase
Solution Approach 1:
The patent enables the sensor system to perform self-diagnosis and self-verification by automatically comparing measurements against reference data. The control device autonomously determines whether malfunctions occur without requiring service workshop intervention. This self-service capability allows continuous operation while maintaining sensor accuracy, eliminating the need for frequent service visits.
Solution Approach 2:
By pre-establishing gate region reference data and implementing automated comparison algorithms, the system prepares everything needed for accurate sensor verification in advance. This preliminary setup enables comprehensive checks to be performed instantly during normal operation rather than requiring time-consuming service workshop appointments.
3Reliability
If detection threshold for malfunctions is lowered, then reliability of detection is improved, but false alarms increase leading to unnecessary service interruptions
Solution Approach 1:
The reference data from gate regions serves as an objective intermediary standard that enables highly sensitive detection without increasing false alarms. By comparing against known reference values, the system can confidently identify even subtle malfunctions while maintaining high productivity, as false positives are eliminated through objective comparison rather than subjective algorithmic guessing.
Data Source
AI summary
The present disclosure relates to a method for detecting a malfunction of at least one environment sensor of a motor vehicle operating while the motor vehicle passes a predefined gate region of a road network. Detection data is determined based at least in part on sensor data from the at least one environment sensor. A deviation of the detection data from reference data is determined. The reference data describes at least one object actually present in the gate region. An entry regarding a malfunction of the at least one environment sensor is stored when the deviation fulfills a predefined indicator criterion.