Joint Sensor Calibration System for Autonomous Vehicles

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current sensor calibration methods for autonomous vehicles are inadequate for high-precision joint calibration of multiple sensors, as they focus on simple conditions and lack accuracy in complex urban environments, requiring a more systematic approach for accurate positioning and calibration of millimeter-wave radars, laser radars, and image acquisition apparatuses.

Innovation Solution

A system and method for joint calibration of sensors in vehicles, utilizing a vehicle positioning unit to guide the vehicle to a predetermined position and a target unit with multiple targets for precise calibration, including display and fixed-style targets, to determine sensor positions in a vehicle coordinate system, ensuring high accuracy and adaptability to various sensor types.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If single sensor calibration is performed under simple conditions, then calibration process is simple and fast, but calibration accuracy is insufficient for complex urban environments and high-precision requirements

Engineering Contradiction:
Improvecalibration accuracyVSAvoidcalibration system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The calibration system is segmented into multiple independent target components (first target for millimeter-wave radar, second target for laser radar, third target for image acquisition apparatus) positioned at different locations. Each target can be independently designed and positioned to optimize calibration for specific sensor types while maintaining overall system functionality.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The calibration system uses a unified coordinate system framework that can simultaneously calibrate multiple sensor types (millimeter-wave radar, laser radar, image acquisition apparatus) with different measurement principles. The system performs multi-functional calibration tasks through a single integrated setup, reducing the need for separate calibration systems for each sensor type.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Reliability

If multiple sensor types are calibrated simultaneously with multiple target types, then comprehensive calibration accuracy improves, but system complexity and setup requirements increase

Engineering Contradiction:
Improvesensor fusion accuracyVSAvoidcalibration operation difficulty
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

A unified coordinate system acts as an intermediary framework that connects multiple sensor calibration processes. The system establishes coordinate transformations between the vehicle coordinate system and global coordinate system, enabling consistent spatial reference across different sensor types and simplifying the integration of calibration results.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary positioning of multiple target types at predetermined locations before actual sensor calibration. Vehicle positioning units pre-position the vehicle relative to the targets, and target positions are pre-established in the calibration area, reducing operational complexity during the actual calibration process.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If high-precision joint calibration is implemented for autonomous driving, then perception accuracy improves, but calibration time and resource requirements increase

Engineering Contradiction:
Improvesensor positioning precisionVSAvoidcalibration time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The calibration system enables continuous simultaneous measurement by multiple sensors (millimeter-wave radar, laser radar, image acquisition apparatus) of different target types. All sensors operate concurrently rather than sequentially, maintaining continuous useful action throughout the calibration process and reducing total calibration time.

Inventive Principle:
Principle #20Continuity of useful action

Solution Approach 2:

The system utilizes spatial dimensionality by positioning targets at different locations and orientations in three-dimensional space. Targets are arranged to provide calibration data from multiple spatial perspectives, enabling comprehensive sensor calibration through spatial diversity rather than requiring extended temporal measurement periods.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP4177629A1System and method for joint calibration of sensors, vehicle and storage medium
Publication Date: 2023.05.10 NIO TECH ANHUI CO LTD
  • EP4177629A1 patent drawingFigure 1
  • EP4177629A1 patent drawingFigure 2
  • EP4177629A1 patent drawingFigure 3

AI summary

The disclosure relates to a system and method for joint calibration of sensors, a vehicle, and a storage medium. The method for joint calibration of sensors includes the following steps: guiding a vehicle to a predetermined position; positioning at least two types of targets relative to the predetermined position; and measuring the at least two types of targets by using sensors in the vehicle that correspond to the at least two types of targets to jointly calibrate the sensors. According to the method for joint calibration of sensors, efficient and accurate joint calibration of the sensors in the vehicle may be implemented.