Autonomous Vehicle Control Unit Modification via Virtual Scenario Recreation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for developing autonomous vehicle control units are inefficient in handling unexpected situations, as they rely on artificially generated test cases and require human interaction, which limits their ability to respond effectively to real-world disengagement events.

Innovation Solution

A method that utilizes disengagement events to identify and modify the control unit by recreating real-life scenarios in a virtual environment, using specially calibrated sensors and a deterministic simulator to analyze and improve the control unit's components, without manual intervention, focusing on critical parts of the system.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If artificially generated test cases are used for safety testing, then basic control unit development is enabled, but unexpected cases and real-world disengagement events cannot be effectively handled

Engineering Contradiction:
Improvehandling of unexpected situationsVSAvoidresponse to real-world disengagement events
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent creates virtual copies of real-world disengagement scenarios by recording actual sensor data from autonomous vehicle disengagement events and reconstructing these scenarios in a virtual environment. This allows the control unit to be tested and modified using authentic real-world data rather than artificially generated test cases, thereby improving both reliability for unexpected situations and adaptability to real-world events.

Inventive Principle:
Principle #26Copying

Solution Approach 2:

The patent performs preliminary recording of sensor data during actual disengagement events, storing this data for later analysis and virtual scenario reconstruction. By capturing and preserving real disengagement data in advance, the system can subsequently create accurate virtual representations of these events to improve the control unit's handling of unexpected situations without requiring the original real-world events to recur.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If extensive safety testing is performed to ensure control system readiness, then safety is improved, but development time and resources increase significantly

Engineering Contradiction:
Improvecontrol system safetyVSAvoiddevelopment time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces physical road testing with virtual simulation by transferring recorded real-world sensor data into a virtual environment where scenarios can be reproduced and tested computationally. This substitution of mechanical/physical testing with digital simulation dramatically reduces development time while maintaining or improving safety validation, as virtual scenarios can be executed rapidly and iteratively without the constraints of real-world testing.

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

Solution Approach 2:

The patent creates a universal testing framework where a single set of recorded disengagement data can be used to generate multiple virtual scenarios for comprehensive control unit testing. This multi-functional approach allows the same real-world data to serve various testing purposes simultaneously, reducing the overall time and resources needed for extensive safety testing while ensuring thorough validation.

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

3Ease of operation

If human interaction is required in simulation steps, then scenario construction flexibility is maintained, but automation and efficiency are reduced

Engineering Contradiction:
Improvescenario construction flexibilityVSAvoidmanual intervention requirement
Core Design Contradiction:
Ease of operationVSExtent of automation

Solution Approach 1:

The patent enables the system to automatically reconstruct virtual scenarios from recorded sensor data without requiring manual human intervention. The framework self-services by taking raw disengagement data, processing it through the virtual environment, and generating test scenarios automatically. This automation maintains scenario construction flexibility while eliminating the need for human interaction in the simulation steps, thereby improving efficiency and consistency.

Inventive Principle:
Principle #25Self-service

4Quantity of substance

If manually generated test scenarios are used, then coverage of basic cases is achieved, but critical disengagement scenarios cannot be captured

Engineering Contradiction:
Improvenumber of test casesVSAvoidaccuracy of disengagement scenario representation
Core Design Contradiction:
Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent creates accurate copies of real disengagement scenarios by directly transferring sensor data from actual autonomous vehicle events into the virtual environment. This copying process preserves the precise details, conditions, and characteristics of real disengagement events, thereby achieving high measurement precision in scenario representation while maintaining comprehensive coverage through the accumulation of multiple recorded events.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP3877740B1Method and system for modifying a control unit of an autonomous car
Publication Date: 2023.06.07 AIMOTIVE KFT
  • EP3877740B1 patent drawingFigure 1
  • EP3877740B1 patent drawingFigure 2
  • EP3877740B1 patent drawingFigure 3

AI summary

The invention is a method for modifying a control unit of an autonomous car equipped with sensors, in the course of which, - disengagement related data are recorded during an autonomous test action performed in autonomous driving mode of the autonomous car in an analysis time window before a disengagement event (S40), - a virtual scenario is generated in a simulator apparatus for the analysis time window based on the disengagement related data, and a first virtual run is performed, so that the first virtual run lead to the disengagement event in the virtual scenario (S50), - based on the virtual scenario an inadequately operating component of the control unit is identified (S60), and - the inadequately operating component of the control unit is modified, a second virtual autonomous run is performed (S70), and it is checked whether the disengagement event is avoided in the second virtual autonomous run. (Fig. 2)