AV Sensor Calibration Check in a Drive-Through Target Lane
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for an efficient calibration check process for sensors on autonomous vehicles (AVs) that can be performed in a space-constrained environment, such as a parking garage, and allows for remote supervision.
Innovation Solution
A drive-through calibration process that utilizes a mapped environment with a series of targets, including camera, LiDAR, and RADAR targets, to check the calibration of AV sensors. The process involves the AV driving along a path that includes turns and straight segments, allowing for the collection of data from various sensors, which is then processed to determine the status of calibration checks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a traditional calibration process is used for AV sensors, then calibration accuracy can be maintained, but the process requires large space and cannot be performed in space-constrained environments like parking garages
Solution Approach 1:
The calibration environment is segmented into virtual targets superimposed on the real-world parking garage space. Instead of requiring a large physical calibration course, the system divides the limited physical space into multiple virtual calibration points that can be sequentially visited by the AV, enabling comprehensive calibration in a constrained area.
Solution Approach 2:
A computer-generated virtual target acts as an intermediary between the physical parking garage environment and the sensor calibration process. The virtual target, superimposed on real-world features, mediates the calibration by providing measurable reference points that bridge the gap between limited physical space and comprehensive calibration requirements.
2Extent of automation
If manual calibration supervision is implemented, then calibration accuracy can be verified, but the process requires significant human involvement and time
Solution Approach 1:
The AV performs self-calibration by autonomously navigating to virtual targets and collecting sensor data. The system automatically compares collected data against expected calibration parameters, eliminating the need for manual supervision while maintaining calibration accuracy through automated verification protocols.
Solution Approach 2:
The system implements automated feedback loops where sensor data collected during drive-through calibration is immediately processed and compared against reference values. The automated system provides real-time feedback on calibration status, verifying accuracy without human intervention and enabling iterative adjustments if needed.
3Productivity
If comprehensive sensor calibration is performed, then all sensor alignment issues can be detected, but the calibration process becomes time-consuming and complex
Solution Approach 1:
The virtual target system serves multiple calibration functions simultaneously - it can calibrate cameras, LIDAR, and other sensors using the same infrastructure. The drive-through process universally tests multiple sensor types and calibration parameters in a single integrated workflow, reducing overall process complexity while maintaining comprehensiveness.
Solution Approach 2:
The calibration process is made dynamic through the drive-through approach, where the AV moves through different positions and angles relative to virtual targets. This dynamic data collection from multiple perspectives enables comprehensive calibration verification without requiring static, complex multi-position setups, improving efficiency while maintaining thoroughness.
Data Source
AI summary
System, methods, and computer-readable media for a calibration check process of sensors on an autonomous vehicle (AV) via a drive-through environment mapped with a series of targets. The targets are used to check the calibration of the sensors on the AV as a quality check before the AV is determined fit to drive autonomously. Supervising a calibration process that collects data from the sensors on the AV based on exposure to a series of targets or augmented camera targets. The collected data may be used for extrinsic calibration aimed to determine that extrinsic parameters that define the rigid relationship between sensors are in set checker bounds.