Joint Sensor Calibration System for Autonomous Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current sensor calibration methods for autonomous vehicles are inadequate for high-precision joint calibration of multiple sensors, as they focus on simple conditions and lack accuracy in complex urban environments, requiring a more systematic approach for accurate positioning and calibration of millimeter-wave radars, laser radars, and image acquisition apparatuses.
Innovation Solution
A system and method for joint calibration of sensors in vehicles, utilizing a vehicle positioning unit to guide the vehicle to a predetermined position and a target unit with multiple targets for precise calibration, including display and fixed-style targets, to determine sensor positions in a vehicle coordinate system, ensuring high accuracy and adaptability to various sensor types.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If single sensor calibration is performed under simple conditions, then calibration process is simple and fast, but calibration accuracy is insufficient for complex urban environments and high-precision requirements
Solution Approach 1:
The calibration system is segmented into multiple independent target components (first target for millimeter-wave radar, second target for laser radar, third target for image acquisition apparatus) positioned at different locations. Each target can be independently designed and positioned to optimize calibration for specific sensor types while maintaining overall system functionality.
Solution Approach 2:
The calibration system uses a unified coordinate system framework that can simultaneously calibrate multiple sensor types (millimeter-wave radar, laser radar, image acquisition apparatus) with different measurement principles. The system performs multi-functional calibration tasks through a single integrated setup, reducing the need for separate calibration systems for each sensor type.
2Reliability
If multiple sensor types are calibrated simultaneously with multiple target types, then comprehensive calibration accuracy improves, but system complexity and setup requirements increase
Solution Approach 1:
A unified coordinate system acts as an intermediary framework that connects multiple sensor calibration processes. The system establishes coordinate transformations between the vehicle coordinate system and global coordinate system, enabling consistent spatial reference across different sensor types and simplifying the integration of calibration results.
Solution Approach 2:
The system performs preliminary positioning of multiple target types at predetermined locations before actual sensor calibration. Vehicle positioning units pre-position the vehicle relative to the targets, and target positions are pre-established in the calibration area, reducing operational complexity during the actual calibration process.
3Measurement precision
If high-precision joint calibration is implemented for autonomous driving, then perception accuracy improves, but calibration time and resource requirements increase
Solution Approach 1:
The calibration system enables continuous simultaneous measurement by multiple sensors (millimeter-wave radar, laser radar, image acquisition apparatus) of different target types. All sensors operate concurrently rather than sequentially, maintaining continuous useful action throughout the calibration process and reducing total calibration time.
Solution Approach 2:
The system utilizes spatial dimensionality by positioning targets at different locations and orientations in three-dimensional space. Targets are arranged to provide calibration data from multiple spatial perspectives, enabling comprehensive sensor calibration through spatial diversity rather than requiring extended temporal measurement periods.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The disclosure relates to a system and method for joint calibration of sensors, a vehicle, and a storage medium. The method for joint calibration of sensors includes the following steps: guiding a vehicle to a predetermined position; positioning at least two types of targets relative to the predetermined position; and measuring the at least two types of targets by using sensors in the vehicle that correspond to the at least two types of targets to jointly calibrate the sensors. According to the method for joint calibration of sensors, efficient and accurate joint calibration of the sensors in the vehicle may be implemented.