Real-Time Driver Class Identification Using Speed Profile Matching

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current driver assistance systems lack the ability to identify a driver's driving class in real-time without altering their driving conditions, making it difficult to provide personalized and adaptive speed recommendations for enhanced safety.

Innovation Solution

A method that determines a reference driving class by comparing the driver's speed to predetermined speed profiles, using a hierarchical classification based on observed speed and acceleration data, and continuously updates the classification using a sliding window approach to ensure accurate and real-time identification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If real-time driver identification is implemented, then personalized speed recommendations can be provided, but computational power requirements increase

Engineering Contradiction:

Improvepersonalized speed recommendationsVSAvoidcomputational power requirements

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent segments the continuous speed profile into discrete driving classes (e.g., aggressive, moderate, conservative drivers). By categorizing drivers into predefined classes rather than creating unique profiles for each driver, the system reduces computational complexity while maintaining personalized recommendations. The segmentation allows the system to match drivers to pre-defined templates rather than performing complex real-time analysis of every driver's complete speed curve.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent uses a sliding window approach that processes only recent speed data rather than analyzing the entire driving history continuously. This disposable approach to data processing - where only the most recent window of speed measurements is considered - reduces computational load while maintaining accuracy in identifying current driving behavior patterns.

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

2Measurement precision

If comprehensive speed monitoring is performed, then accurate driving class identification is achieved, but system complexity increases

Engineering Contradiction:

Improvedriving class identification accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the essential features needed for driving class identification - specifically speed and acceleration data - from the complete set of possible vehicle parameters. By focusing on these two key measurements and ignoring other potentially useful but computationally expensive parameters, the system achieves accurate classification with reduced complexity.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent transforms the continuous speed measurements into discrete driving class categories. By changing the parameter representation from continuous speed values to discrete class labels, the system simplifies the identification process while maintaining sufficient accuracy for personalized recommendations.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If continuous real-time identification is implemented, then up-to-date driving class information is available, but processing time increases

Engineering Contradiction:

Improveup-to-date driving class informationVSAvoidprocessing time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent implements periodic re-identification of driving class using a sliding window approach, where the system periodically reassesses the driver's class based on recent speed data rather than continuously processing every speed measurement. This periodic action maintains up-to-date information while reducing processing time compared to continuous analysis.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The patent performs preliminary classification by comparing current speed data against pre-defined driving class templates. This preliminary action allows the system to quickly match drivers to classes without performing complex real-time analysis, thereby reducing processing time while maintaining reliability of the identification.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentEP3472015B1Method for determining a reference driving class

Publication Date: 2020.06.10 MICHELIN & CO (CIE GEN DES ESTAB MICHELIN)

Figure 1~2

AI summary

The invention relates to a method for determining a reference driving class of a driver from among a set of predetermined driving classes, each predetermined driving class being characterized by a speed profile, the method comprising the following steps: -the speed of a driver over a given journey is determined, -the observed speed is compared with the speed profiles of each of the predetermined driving classes so as to obtain a disparity between the driver and each of the driving classes, -a driving class is determined, detected as being that which minimizes this disparity, and -the reference driving class is determined as a function of this detected driving class.