Vehicle Sensor Calibration Using Camera-Map Location Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle sensor systems face inaccuracies in location estimation due to errors in wheel speed sensors and yaw rate sensors, which are affected by tire pressure and internal disturbances like operating time and temperature, impacting the reliability of Advanced Driver Assistance Systems (ADAS) and autonomous vehicle control.
Innovation Solution
A method and apparatus that determine a scale factor for wheel speed sensors and bias for yaw rate sensors based on estimated location information acquired by matching camera images with precise maps, using a scale factor determining unit and bias determining unit to correct sensor outputs, thereby reducing errors and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If wheel speed sensor and yaw rate sensor are used for location estimation, then the system can operate without additional sensing means, but the measurement precision deteriorates due to sensor errors affected by tire pressure and internal disturbances
Solution Approach 1:
The patent introduces camera images and precise maps as intermediary elements to bridge the gap between the existing sensors and accurate location estimation. The camera captures visual information that is matched against pre-stored precise map data, creating an intermediary reference system that corrects the inaccuracies of the wheel speed and yaw rate sensors without requiring complete replacement of the existing sensor configuration.
Solution Approach 2:
The patent replaces the purely mechanical sensor-based location estimation system with a hybrid system that incorporates optical (camera) and information processing (image matching) components. This substitution transforms the mechanical measurement approach into an information-based approach, where location is determined by matching visual features against map data rather than relying solely on mechanical sensor integration.
2Measurement precision
If camera image matching with precise map is used to correct sensor errors, then the measurement precision improves, but the device complexity increases due to additional processing requirements
Solution Approach 1:
The patent makes the camera serve multiple functions: it acts as both a regular vehicle camera for general purposes and as a location correction sensor by matching its images with precise maps. This multi-functionality eliminates the need for dedicated additional sensing means, as the existing camera infrastructure is utilized for dual purposes, thereby reducing overall system complexity while improving measurement precision.
Solution Approach 2:
The system uses the vehicle's own camera and existing map data to self-correct its location estimation errors. The camera images are processed against the precise map that the vehicle already possesses, creating a self-contained correction mechanism that does not require external correction devices or additional sensing infrastructure.
3Measurement precision
If scale factor and bias correction are applied to sensor outputs, then the measurement precision improves, but the ease of operation deteriorates due to complex correction processes
Solution Approach 1:
The patent performs preliminary correction by calculating scale factors and bias values in advance based on camera-map matching, then applies these corrections to the sensor data. The correction parameters are determined beforehand through image matching processes, and the actual sensor outputs are corrected using these pre-computed values, which simplifies the real-time operation compared to performing complex corrections on-the-fly.
Data Source
AI summary
In accordance with an aspect of the present disclosure, there is provided a vehicle sensor correction information determining method comprising: acquiring estimated location information of a vehicle based on matching information between a camera image photographed by a camera of the vehicle and a landmark on a precise map; and determining a scale factor for a wheel speed sensor based on an estimated travel distance calculated based on the acquired estimated location information of the vehicle and a travel distance of the vehicle detected by the wheel speed sensor of the vehicle.