Map-Assisted Sensor Calibration Using Static Road Objects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor calibration methods for autonomous vehicles are inefficient and require time-consuming recalibration processes, often necessitating the vehicle to return to a garage, and struggle to accurately distinguish between static and moving objects.
Innovation Solution
Utilizing high-definition LiDAR map data and a semantic mapping layer to extrinsically calibrate sensors by identifying static objects and compensating for ego-motion, allowing for quicker recalibration on the road based on static points determined through histogram analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional sensor calibration methods are used, then calibration accuracy can be maintained, but the recalibration process becomes time-consuming and requires the vehicle to return to a garage
Solution Approach 1:
The system performs preliminary identification of static objects using map data before actual calibration occurs. By pre-processing the environment understanding and identifying calibration targets in advance, the system reduces the time required for actual calibration execution while maintaining accuracy requirements.
Solution Approach 2:
Map data serves as an intermediary between the sensor calibration system and the environment. The map provides pre-existing information about static objects, which mediates the calibration process by eliminating the need for time-consuming real-time identification of calibration targets, thus reducing recalibration time while preserving accuracy.
2Reliability
If traditional sensor calibration methods are used, then calibration can be performed, but the system struggles to accurately distinguish between static and moving objects
Solution Approach 1:
Map data acts as an intermediary that provides prior knowledge about the environment's static structures. This intermediary information helps the system reliably distinguish static objects from moving ones without requiring complex real-time analysis, thereby improving object differentiation accuracy while managing system complexity.
Solution Approach 2:
The system performs preliminary analysis of the environment using map data to identify static objects before the calibration process. This advance preparation improves reliability in distinguishing static from moving objects by establishing a baseline understanding of the environment, reducing the need for complex real-time differentiation during calibration.
3Measurement precision
If frequent recalibration is performed to maintain accuracy, then calibration precision is improved, but processing load and time consumption increase
Solution Approach 1:
By performing preliminary identification of static calibration objects using map data, the system prepares the calibration environment in advance. This allows frequent recalibration to be performed efficiently without proportionally increasing processing load, as the heavy lifting of environment understanding is already done.
Solution Approach 2:
Map data serves as an intermediary that reduces the processing burden during recalibration. By providing pre-processed environmental information, the map allows the system to perform frequent calibration operations with reduced computational load, improving productivity while maintaining calibration precision.
Data Source
AI summary
System, methods, and computer-readable media for extrinsically calibrating sensors on an autonomous vehicle (AV) by leveraging map data. By leveraging offline data, such as the map data on the AV, the extrinsic calibration of the sensors may be performed on road to update or check previous calibrations. Static objects are determined in the scene by leveraging a map with a semantic map layer and sensors are extrinsically calibrating using a portion of a set of static points associated with the determined static objects by determining relative positions and angles the sensors.