LIDAR Cross-Validation Calibration for Consistent Obstacle Mapping

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

There is a lack of efficient methods for calibrating LIDAR devices in autonomous vehicles, which is crucial for accurate navigation and obstacle detection, as existing methods are not scalable for mass production.

Innovation Solution

A LIDAR calibration system that uses a coordinate converter, represented by a quaternion function, to translate LIDAR images from a local coordinate system to a global coordinate system, optimizing parameters through iterative processes to maximize consistency of obstacle detection across multiple images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional LIDAR calibration methods are used, then calibration can be performed, but the process is not efficient and not scalable for mass production

Engineering Contradiction:

Improvecalibration efficiencyVSAvoidscalability for mass production

Core Design Contradiction:

ProductivityVSEase of manufacture

Solution Approach 1:

The LIDAR calibration system performs self-calibration by automatically processing LIDAR images to determine coordinate converter parameters without requiring manual intervention or specialized calibration equipment. The system uses its own captured images and internal algorithms to calibrate itself, making the process efficient and scalable for mass production

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces traditional mechanical calibration methods with image processing and computational algorithms. By using LIDAR images and automated coordinate conversion calculations, the system eliminates the need for physical calibration artifacts and manual adjustment procedures, thereby improving both efficiency and scalability

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If coordinate converter parameters are not optimized, then processing is faster, but obstacle detection consistency across multiple images deteriorates

Engineering Contradiction:

Improveobstacle detection consistencyVSAvoiditerative optimization process

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system implements feedback by iteratively adjusting coordinate converter parameters and evaluating obstacle detection consistency across multiple LIDAR images. The optimization process uses detection results as feedback to refine parameters, ensuring high reliability while automating the complexity through algorithmic processing

Inventive Principle:

Principle #23Feedback

3Loss of time

If manual LIDAR calibration methods are used, then calibration can be performed, but the process is time-consuming and not suitable for periodic calibration requirements

Engineering Contradiction:

Improvecalibration timeVSAvoidLIDAR calibration accuracy

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The system performs preliminary action by pre-processing LIDAR images and pre-calculating coordinate converter parameters during periodic calibration intervals. This allows the system to maintain accurate calibration without requiring time-consuming manual procedures during actual autonomous operation, thus reducing time loss while preserving measurement precision

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11428817B2Automatic LIDAR calibration based on cross validation for autonomous driving

Publication Date: 2022.08.30 BAIDU USA LLC

AI summary

In one embodiment, a set of LIDAR images representing LIDAR point cloud data captured by a LIDAR device of an ADV at different points in time is received. For each of the LIDAR images, a perception method is utilized to determine a location of an obstacle captured in the LIDAR image in a local coordinate system. The LIDAR image is transformed using a coordinate converter (e.g., a LIDAR to GPS coordinate conversion logic or function) from the local coordinate system to a global coordinate system. The coordinate converter is optimized based on the transformed LIDAR images by adjusting one or more parameters of the coordinate converter and the above operations are iteratively performed to obtain a set of optimal parameters. The optimized coordinate converter can then be utilized to process subsequent LIDAR images during autonomous driving at real-time.