Vehicle Sensor Extrinsic Calibration Using Combined Target
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately calibrating sensors due to manufacturing discrepancies and environmental factors, leading to inconsistent data interpretation and potential safety risks, as existing calibration methods do not adequately account for intrinsic sensor properties and extrinsic relationships between sensors.
Innovation Solution
A dynamic scene calibration system using a motorized turntable and sensor targets allows for comprehensive and efficient intrinsic and extrinsic calibration of vehicle sensors, enabling precise transformation mapping and correction of distortions, thereby enhancing sensor accuracy and consistency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional calibration methods are used, then manufacturing simplicity is maintained, but measurement precision and reliability of sensor data are insufficient
Solution Approach 1:
The patent introduces a calibration target as an intermediary object with known geometric properties. This target serves as a mediator between the sensors and the calibration process, providing reference points that enable precise measurement of sensor characteristics without requiring complex calibration equipment or procedures.
Solution Approach 2:
The patent creates a simplified model of the calibration process by using a virtual or physical representation of the calibration target. This copying approach allows the system to work with idealized geometric shapes that are easier to process and interpret, thereby improving measurement precision while maintaining practical simplicity.
2Reliability
If comprehensive intrinsic and extrinsic calibration is performed, then sensor reliability and data consistency are improved, but calibration time and computational resources increase
Solution Approach 1:
The patent performs preliminary calibration actions by pre-computing and storing calibration parameters or creating lookup tables during system setup. This allows the actual sensor calibration to be performed more quickly during operation, as the heavy computational tasks have already been completed in advance, thereby reducing calibration time while maintaining comprehensive calibration coverage.
Solution Approach 2:
The patent implements a dynamic calibration approach where the calibration process can adapt its complexity based on operational needs. The system can perform full intrinsic and extrinsic calibration when high reliability is required, or use simplified calibration methods during normal operation, thereby balancing calibration comprehensiveness with time efficiency.
3Manufacturing precision
If manufacturing discrepancies and environmental factors are accounted for, then sensor accuracy is improved, but calibration complexity and difficulty increase
Solution Approach 1:
The patent adjusts calibration parameters to account for manufacturing discrepancies and environmental factors. By modifying the calibration model to include terms for positioning errors, lens distortion, and environmental variations, the system can maintain high sensor accuracy without requiring complex calibration procedures, as the parameters are adjusted within the existing calibration framework.
Data Source
AI summary
Sensors coupled to a vehicle are calibrated, optionally using a dynamic scene with sensor targets around a motorized turntable that rotates the vehicle to different orientations. One vehicle sensor captures a representation of one feature of a sensor target, while another vehicle sensor captures a representation of a different feature of the sensor target, the two features of the sensor target having known relative positioning on the target. The vehicle generates a transformation that maps the captured representations of the two features to positions around the vehicle based on the known relative positioning of the two features on the target.