Vehicle Sensor Self-Calibration via Static Object Cross-Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle environment detection systems require redundant sensors for calibration, leading to increased complexity and the need for additional sensors with overlapping fields of view, which is inefficient and prone to errors.
Innovation Solution
A method that uses at least two different types of environment sensors to detect and categorize objects, where static objects are identified using a stereo camera and their positions are validated by a second sensor type, such as ultrasonic sensors, without requiring overlapping fields of view, allowing for continuous position updating and error detection beyond the camera's field of view using vehicle movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If redundant sensors with overlapping fields of view are used for calibration and error detection, then measurement reliability is improved, but device complexity and sensor quantity increase
Solution Approach 1:
The patent makes existing environment sensors serve dual functions: their primary function for detecting dynamic objects and an additional function for detecting static objects and self-calibration. This eliminates the need for dedicated redundant calibration sensors, reducing device complexity while maintaining measurement reliability through cross-validation of sensor data.
Solution Approach 2:
The system performs self-calibration using its own existing sensors without requiring external calibration equipment or additional redundant sensors. The sensors calibrate themselves by comparing measurements of static objects against each other and against known reference positions, enabling the system to maintain reliability autonomously.
2Manufacturing precision
If additional sensors are installed to ensure overlapping fields of view for calibration, then calibration accuracy is improved, but the quantity of sensors and system complexity increase
Solution Approach 1:
Existing environment sensors are made multi-functional by adding calibration and static object detection capabilities to their primary dynamic object detection function. This eliminates the need for additional dedicated calibration sensors, maintaining calibration accuracy while reducing the total quantity of sensors required.
Solution Approach 2:
The patent merges the calibration function with the existing environment sensing function. Instead of separating calibration sensors from operational sensors, the system combines both functions into the same sensor suite, using the same hardware resources for multiple purposes to reduce overall sensor quantity.
3Measurement precision
If static objects are tracked only within the field of view of the first sensor type, then measurement precision is maintained, but the system cannot detect errors beyond the field of view
Solution Approach 1:
The patent uses static objects as intermediary reference points that bridge the fields of view of different sensors. By tracking static objects that appear in multiple sensor fields of view at different times and using vehicle movement to update their positions, the system creates a continuous reference framework that enables cross-validation and error detection beyond any single sensor's immediate field of view.
Solution Approach 2:
The system preliminarily identifies and tracks static objects within the first sensor's field of view before the vehicle moves. These pre-identified static objects serve as reference points that can later be used to validate measurements from other sensors, enabling error detection capability that extends beyond the initial field of view while maintaining measurement precision through continuous position updates.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables reliable detection of sensor errors without additional redundant sensors, simplifies troubleshooting, and integrates seamlessly into existing systems by comparing position data from different sensors to identify deviations beyond statistical fluctuations.
Implementation Method 1
objects in the environment of the vehicle are detected with a first environment sensor type... the relative position of the detected static objects to the vehicle is determined
Implementation Method 2
this position is compared with a position determined via a second type of environment sensor... such as ultrasonic sensors
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
The invention relates to a method for checking an environment detection system of a vehicle (10), wherein the environment detection system comprises at least two different types of environment sensors (12, 14) and wherein objects in the environment of the vehicle (10) are detected with a first type of environment sensor (12), the objects are categorized into static and dynamic objects using the data from the first type of environment sensor (12), the relative position of the detected static objects to the vehicle (10) is determined, this position is compared with a position determined by a second type of environment sensor (14) and if there is a deviation above a limit value, a fault is inferred.wherein the field of view (18) of the first type of environmental sensor (12) does not overlap with the field of view (20) of the second type of environmental sensor (14) and the relative position of the detected static objects is updated after they leave the field of view (18) of the first type of environmental sensor (12), taking into account the movement of the vehicle (10). The invention further relates to a computer program and a device for carrying out the method.