LiDAR Composite Data Evaluation for Vehicle Algorithms in Weather

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image processing algorithms for vehicles, such as those used in autonomous driving, are not adequately evaluated due to the lack of reflection of actual surroundings in weather disturbance simulations, leading to inappropriate performance assessments.

Innovation Solution

A data processing algorithm evaluation device that includes a data storage unit for optical ranging data, a data generating unit to create composite data by reflecting disturbance information based on actual conditions, and a data processing unit to evaluate the algorithm's performance using these composite data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Extent of automation

If weather disturbance images are created using computer graphics and superimposed over actual images, then the evaluation process can be automated, but the composite image does not reflect actual surroundings and leads to inappropriate performance evaluations

Engineering Contradiction:

Improveevaluation process automationVSAvoidperformance evaluation accuracy

Core Design Contradiction:

Extent of automationVSMeasurement precision

Solution Approach 1:

The patent uses optical ranging data (LiDAR) to create a three-dimensional model of the actual surroundings, which serves as an accurate copy of the real environment. This model is then processed to generate composite data that reflects actual weather disturbances, replacing the inaccurate computer graphics-based copying method.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the traditional image processing approach with optical ranging data processing. By using LiDAR data to reconstruct the three-dimensional environment and apply disturbance models, the system achieves more accurate performance evaluations without relying on inaccurate image superposition techniques.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Ease of manufacture

If computer graphics are used to create weather disturbance images, then the generation process is simple, but the surrounding objects and environment are not accurately reflected

Engineering Contradiction:

Improvedisturbance image generationVSAvoidsurroundings reflection accuracy

Core Design Contradiction:

Ease of manufactureVSManufacturing precision

Solution Approach 1:

The patent performs preliminary processing of optical ranging data to create a three-dimensional model of the environment before generating the composite data. This preliminary reconstruction step ensures that the underlying model accurately represents actual surroundings, allowing subsequent disturbance application to be precise and accurate.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent changes the fundamental parameters of the data representation from two-dimensional images to three-dimensional optical ranging data. This parameter change enables the system to accurately model the spatial arrangement of surrounding objects and apply weather disturbances with precision, overcoming the limitations of traditional image-based approaches.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If traditional image processing algorithms are evaluated using composite images, then the evaluation can be conducted efficiently, but the results do not accurately assess the algorithm's performance in real conditions

Engineering Contradiction:

Improveevaluation efficiencyVSAvoidperformance assessment validity

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent creates a universal evaluation system that can assess image processing algorithms under various weather disturbance conditions using the same optical ranging data framework. The three-dimensional model and disturbance application mechanism serve multiple evaluation scenarios, maintaining both efficiency and reliability across different testing conditions.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent implements a feedback mechanism where the composite data generated from optical ranging information is used to evaluate algorithm performance, and the evaluation results feed back to improve the understanding of algorithm behavior in real conditions. This feedback loop ensures that the evaluation process accurately reflects real-world performance while maintaining efficiency.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20250362412A1Data processing algorithm evaluation device

Publication Date: 2025.11.27 MITSUBISHI HEAVY IND MACHINERY SYST LTD

AI summary

A data processing algorithm evaluation device includes: a data storage unit configured to store therein a plurality of pieces of optical ranging data detected from a vehicle; a data generating unit configured to acquire and interpret target data, upon receiving an input of disturbance information representing disturbance for the target data, from the plurality of pieces of optical ranging data stored in the data storage unit, and to generate composite data in which the target data is reflected with the disturbance, by editing the target data on the basis of the interpretation; and a data processing unit configured to evaluate performance of a data processing algorithm for determining situation of the vehicle, on the basis of the generated composite data.