Vehicle Sensor Extrinsic Calibration Using Relevant Point Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor calibration methods for vehicles, particularly those involving multiple sensor modalities, suffer from inaccuracies due to the need for precise extrinsic calibration without effective filtering of irrelevant data points, leading to potential errors in sensor alignment and reliability.
Innovation Solution
A method for calibrating vehicle sensors that involves obtaining data from multiple sensors, filtering out irrelevant points based on position, and determining calibration parameters using a priori knowledge and detected objects, with adjustments for steering direction and sensor orientation to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all data points from multiple sensors are used for calibration without filtering, then the calibration process can be performed with available data, but the accuracy and reliability of calibration deteriorates due to inclusion of irrelevant points
Solution Approach 1:
The patent extracts and removes irrelevant data points from sensor data before performing calibration. A filtering unit identifies and eliminates points that do not contribute to accurate calibration, keeping only the relevant subset of data points for the calibration process. This extraction of useful information from the total data set improves calibration reliability without requiring complex additional processing systems.
2Measurement precision
If filtering is applied to remove irrelevant data points, then calibration accuracy improves, but the processing time and computational load increase
Solution Approach 1:
The patent performs filtering as a preliminary action before the actual calibration process. By pre-processing the sensor data to remove irrelevant points beforehand, the system prepares clean calibration data in advance. This preliminary filtering step, while requiring computational resources, enables faster and more accurate calibration execution, as the main calibration algorithm receives already-filtered data without needing to process irrelevant points.
3Reliability
If extrinsic calibration is performed between multiple sensor modalities, then comprehensive sensor alignment is achieved, but errors propagate due to accumulation of calibration uncertainties
Solution Approach 1:
The patent extracts and eliminates sources of calibration error by filtering out data points that could introduce uncertainty. By removing irrelevant and potentially erroneous data points before calibration, the system prevents error propagation that would otherwise accumulate through multiple sensor modalities. This extraction of problematic data preserves the accuracy of calibration information across camera, radar, and lidar sensors.
Data Source
AI summary
The present invention provides a method for calibrating a first and a second sensor of a vehicle, the method comprising:obtaining first data from the first sensor and second data from the second sensor,filtering at least the second data based on positions of data points of the second data, anddetermining one or more parameters of a calibration between the first and the second sensor based on the first data and the filtered second data.