Hands-Off Detection Testing Using Orthogonal Arrays

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing hands-off detection algorithms in vehicles suffer from incomplete testing, leading to false negative or false positive results, which decrease the quality and satisfaction of autonomous driving systems.

Innovation Solution

A method and system using system behavior testing with an orthogonal array to determine a plurality of test cases, generating expected and actual outcomes, and identifying high failure conditions to improve the hands-off detection algorithm.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If comprehensive testing of hands-off detection algorithms is performed to improve reliability, then the number of test cases required increases significantly, but this increases testing time and resource consumption

Engineering Contradiction:
Improvehands-off detection accuracyVSAvoidtesting time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the testing process by identifying and focusing on critical failure conditions rather than exhaustively testing all possible scenarios. The orthogonal array methodology divides the test space into structured subsets that target specific interaction effects between system parameters, allowing comprehensive coverage of critical cases without requiring all possible test combinations.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies partial action by selecting a representative subset of test cases that are most likely to reveal algorithm failures. Rather than performing excessive comprehensive testing, the orthogonal array methodology identifies the minimum necessary test cases that provide maximum coverage of critical interaction effects, achieving sufficient reliability with reduced testing effort.

Inventive Principle:
Principle #16Partial or excessive action

2Productivity

If the number of test cases is reduced to decrease testing time, then testing efficiency improves, but the coverage of vehicle usage conditions decreases

Engineering Contradiction:
Improvetesting efficiencyVSAvoidcoverage of vehicle usage conditions
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent transitions from one-dimensional testing (individual parameters) to multi-dimensional testing by examining interaction effects between multiple system parameters simultaneously. The orthogonal array methodology structures tests to cover combinations of parameters across different dimensions, ensuring comprehensive coverage of vehicle usage conditions even with a reduced number of test cases.

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

Solution Approach 2:

The orthogonal array methodology creates a universal testing framework that can evaluate multiple system parameters and their interactions within a single structured test suite. This multi-functional approach allows the reduced set of test cases to cover diverse vehicle usage conditions across different steering systems, vehicle types, and operating scenarios simultaneously.

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

Data Source

PatentEP4427011B1Hands-off detection for autonomous and partially autonomous vehicles
Publication Date: 2025.08.20 ROBERT BOSCH GMBH
  • EP4427011B1 patent drawingFigure 1
  • EP4427011B1 patent drawingFigure 2
  • EP4427011B1 patent drawingFigure 3

AI summary

Systems and methods for testing a hands-off detection algorithm. The method includes determining a plurality of system behavior test conditions for the algorithm and selecting an orthogonal array defining a plurality of test cases based on the plurality of system behavior test conditions. The method includes generating, for each of the test cases, an expected test outcome. The method includes for each of the test cases, conducting a test of with the vehicle based on the orthogonal array to generate a plurality of actual test outcomes and generating a response table based on the test outcomes, including a plurality of system behavior test condition interactions. The method includes determining, for each of the interactions, a result rating based on the expected test outcomes and the actual test outcomes and identifying, within the response table, which one or more of the test conditions exhibits a high failure condition.