Vehicle Auto-Hold Control Using Learned Driver Braking Patterns

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The auto-hold function in vehicles requires manual activation by the driver, which is inconvenient and limits its usability, as it needs to be manually turned on and off, especially in situations where the vehicle needs to be kept stationary.

Innovation Solution

A method that automatically controls the auto-hold function by learning the driver's operation patterns and habits using collected data, generating a categorization model to determine when to engage or disengage the auto-hold system based on real-time vehicle conditions, without requiring manual intervention.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the auto-hold function requires manual activation by the driver, then the system can ensure driver intention to use the function, but the ease of operation deteriorates due to repeated manual switching

Engineering Contradiction:
Improvedriver intention confirmationVSAvoidmanual switching frequency
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The system performs learning and categorization automatically without requiring driver intervention. The controller collects information on vehicle traveling conditions, selects learning data, performs learning to generate a categorization model, and automatically determines when to activate auto-hold based on learned driver patterns, making the system self-configuring and self-operating

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system uses real-time vehicle condition data and learned driver behavior patterns to dynamically adjust auto-hold activation. The controller continuously monitors vehicle traveling conditions, compares them against the categorization model, and automatically activates or deactivates auto-hold based on the comparison results, creating a closed-loop feedback system

Inventive Principle:
Principle #23Feedback

2Extent of automation

If the auto-hold switch is manually turned on by the driver, then the driver maintains control over the function, but the extent of automation deteriorates

Engineering Contradiction:
Improveautomatic operation modeVSAvoidlearning and categorization system
Core Design Contradiction:
Extent of automationVSDevice complexity

Solution Approach 1:

The controller performs multiple functions: it collects vehicle traveling condition information, selects and stores learning data, performs learning to generate a categorization model, and executes automatic control decisions. This multi-functionality is achieved within a single control unit, avoiding the need for separate dedicated systems for each function

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

3Ease of operation

If the auto-hold function operates only when manually activated, then the system remains simple to understand, but the ease of operation deteriorates in frequent stop-start scenarios

Engineering Contradiction:
Improveconvenience in stop-start situationsVSAvoidtime for manual switching
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The system performs learning and generates a categorization model in advance, before actual automatic control is needed. By pre-processing and storing driver behavior patterns and vehicle condition data, the system is prepared to make immediate automatic decisions when activation conditions are met, eliminating real-time manual intervention delays

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS12077167B2Method of automatically controlling vehicle auto-hold
Publication Date: 2024.09.03 HYUNDAI MOTOR CO LTD
  • US12077167B2 patent drawing
  • US12077167B2 patent drawing
  • US12077167B2 patent drawing

AI summary

A method of automatically controlling vehicle auto-hold includes: collecting information on a vehicle traveling condition that applies when a driver operates a brake pedal while a vehicle is traveling; selecting learning data for learning on a pattern of a driver's operation from among pieces of the information collected in the collecting of the information on the vehicle traveling condition and storing the selected learning data; performing the learning on the pattern of the driver's operation based on the learning data and generating a categorization model for the pattern of the driver's operation according to a result of the learning; and determining whether or not to cause the auto-hold switch to enter an automatic operation mode while the vehicle is traveling, using the categorization model, and selectively causing the auto-hold switch to enter the automatic operation mode according to a result of the determining.