Automated Vehicle Sensor Calibration Using Map-Referenced Targets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for calibrating sensors in automated vehicles are inefficient and require extensive manual effort, delaying operations and needing significant engineer time, especially for long-range sensors where small angular miscalibrations can result in large errors.
Innovation Solution
The use of specially designed stationary calibration targets and preconfigured geographic information associated with each target, allowing controllers to automate the sensor calibration process by referencing expected values and adjusting observed values accordingly, without the need for complex setups like target jungles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual calibration methods are used, then calibration can be performed, but the process requires multiple data collection and processing steps that delay operations and require significant engineer effort
Solution Approach 1:
The calibration system performs self-calibration by automatically comparing sensor observations of calibration targets with pre-stored expected values from maps. The vehicle's own sensors and processors conduct the calibration without external intervention, eliminating the need for manual engineer involvement in data collection and processing steps.
Solution Approach 2:
Expected values for calibration targets are pre-configured and stored in maps before calibration is needed. This preliminary preparation of reference data allows the calibration process to proceed automatically by simply comparing current sensor readings against these pre-established benchmarks, eliminating time-consuming manual setup.
2Measurement precision
If target jungle setups are used, then sensor calibration can be performed, but complex permanent setups requiring significant effort to build and teardown are required
Solution Approach 1:
The calibration system extracts only the essential calibration targets needed for accurate sensor calibration, removing the need for complex target jungle setups. By using a minimal set of strategically placed targets with pre-stored expected values, the system achieves the same calibration precision without the overhead of building and teardown of extensive target arrays.
Solution Approach 2:
The calibration system uses pre-configured map data that can serve multiple vehicles and multiple calibration scenarios. The same map with expected target values can be universally applied across different vehicles and locations, eliminating the need for vehicle-specific or location-specific complex setups.
3Measurement precision
If target jungle setups are used, then sensor calibration can be performed, but significant labor force and time are required to establish and maintain
Solution Approach 1:
The system performs calibration autonomously using its own sensors and processors to compare observations with pre-stored expected values. This self-service capability eliminates the need for labor forces to manually set up, maintain, and teardown calibration targets, dramatically improving calibration efficiency.
Solution Approach 2:
All calibration reference data is pre-configured in maps before deployment. This preliminary preparation eliminates the need for on-site setup efforts and allows calibration to occur automatically during normal vehicle operation, maximizing productivity by eliminating manual labor requirements.
4Adaptability or versatility
If not every location has target jungles established, then deployment flexibility is improved, but calibration capability is reduced
Solution Approach 1:
Instead of requiring physical target jungles at every location, the system uses copies of target information stored in digital maps. The expected values and positions of calibration targets are replicated in map data, allowing any vehicle with access to the appropriate map to perform calibration at any location without physical targets being present at every site.
Solution Approach 2:
The system extracts only the essential calibration information needed from the target jungle concept and stores it digitally in maps. This extraction allows calibration to occur using minimal or no physical targets at the test location, providing deployment flexibility while maintaining calibration precision through the use of pre-configured expected values.
Data Source
AI summary
Disclosed herein include systems and methods for calibrating perception sensors of an automated vehicle using stationary calibration targets and preconfigured geographic information for the calibration targets. A controller of the automated vehicle references map data indicating locations of each calibration target, which is then recorded by the automated vehicle's perception sensors. The automated vehicle is configured to use a corrected geographical position system (GPS) process (e.g., RTK, PPK) to determine positions and orientations of each sensor and/or the automated vehicle. The controller uses these values to generate accurate calibrations of the automated vehicle and the sensors.