Machine Learning Code Change Detection for Virtual Control Units

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing code-generating tools like GitHub Copilot and Google Gemini are not workflow-oriented and require use-case-specific adaptations to create relevant results for virtual artifacts such as virtual control units, necessitating laborious manual processing of diverse materials for automated driving functions in motor vehicles.

Innovation Solution

A computer-implemented method using a generative deep learning model applies a machine learning algorithm to source code segments to identify required code changes, providing a data set that includes annotations for faulty code, missing dependencies, redundant information, and cycle time corrections to generate a virtual control unit efficiently.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If conventional code-generating products (GitHub Copilot, Google Gemini, Amazon Code Whisperer) are used, then code suggestions can be obtained, but the tools are not workflow-oriented and require use-case-specific adaptations, resulting in laborious manual processing

Engineering Contradiction:
Improveease of creating virtual artifactsVSAvoidtime for manual processing of diverse materials
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The system performs preliminary actions by automatically gathering, processing, and analyzing diverse project materials (requirements, specifications, existing code) before the virtual artifact creation process begins. The machine learning model is pre-trained on code repositories and documentation to understand project contexts, eliminating the need for manual material processing and preparation steps that currently consume significant time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system enables self-service by allowing the virtual artifact generation process to autonomously identify required code changes, suggest modifications, and generate test cases without continuous human intervention. The machine learning algorithm independently analyzes project materials and generates relevant code suggestions tailored to the specific use case, reducing reliance on manual processing while maintaining workflow-oriented relevance.

Inventive Principle:
Principle #25Self-service

2Reliability

If manual processing of diverse materials is performed to obtain relevant information, then relevant information for virtual artifact creation can be obtained, but the process is laborious and time-consuming

Engineering Contradiction:
Improverelevance of information for virtual artifact creationVSAvoidefficiency of virtual artifact creation
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The system replaces the mechanical manual processing system with an automated machine learning-based information processing system. The machine learning model automatically extracts, analyzes, and processes relevant information from diverse project materials (requirements, specifications, existing code), substituting human manual efforts while maintaining high relevance through context-aware analysis and generating code suggestions aligned with project-specific needs.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If existing code-generating tools are used without adaptation, then general code suggestions can be provided, but they lack workflow-oriented relevance for specific use cases like virtual control units

Engineering Contradiction:
Improveadaptability to use-case-specific requirementsVSAvoidcomplexity of extending tools for specific use cases
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system achieves universality by designing a multi-functional machine learning model that can handle diverse use cases (virtual control units, software components, test cases) through a single unified platform. The model is trained on multiple code repositories and documentation types, enabling it to adapt to different project contexts and requirements without requiring separate specialized tools for each use case, thereby reducing overall system complexity while maintaining high adaptability.

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

Data Source

PatentEP4597303A1Computer-implemented method for identifying a required code change
Publication Date: 2025.08.06 DSPACE SE & CO KG
  • EP4597303A1 patent drawingFigure 1~3
  • EP4597303A1 patent drawing

AI summary

The invention relates to a computer-implemented method and system (1) for identifying at least one code change (14) required to generate a virtual artifact (10), in particular a virtual control unit, for testing and/or validating a predetermined function, in particular an automated driving function, of a motor vehicle, comprising the steps of providing (S1) at least one source code segment (12) for testing and/or validating the predetermined function of the motor vehicle; applying (S2) a machine learning algorithm (A) to the at least one source code segment (12) to identify the at least one required code change (14); and outputting (S3) a data set (16) of information about the code change (14) required to generate the virtual artifact (10) for testing and/or validating the predetermined function of the motor vehicle.Furthermore, the invention relates to a computer-implemented method for providing a machine learning algorithm (A).