EPS Rack Force Determination Using Multi-Model Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Electric power steering (EPS) systems struggle to provide accurate feedback to drivers about vehicle behavior and road conditions due to high inertia and disturbances from various driving situations, such as different road friction coefficients and vehicle loading, which limits the driver's perception of understeering or oversteering.

Innovation Solution

A method to determine the rack force for EPS systems using multiple models optimized for specific driving situations, combining actual and modeled rack forces through a PI controller to generate a resulting steering torque that adapts to changing conditions, including variables like steering angle, vehicle speed, and lateral acceleration, while minimizing disturbances.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a single model is used to determine rack force, then the system is simple, but it cannot accurately represent actual rack force under all driving conditions

Engineering Contradiction:
Improverack force determination accuracyVSAvoidmodeling system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the rack force determination into multiple separate models, each optimized for specific driving situations (normal driving, parking, high-lateral acceleration). This segmentation allows each model to specialize in particular conditions, improving overall accuracy without requiring a single overly complex universal model.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically selects and switches between different models based on current driving conditions detected by sensors (steering angle, vehicle speed, lateral acceleration). This dynamic adaptation ensures the most appropriate model is used for each situation, maintaining high accuracy across varying operational contexts.

Inventive Principle:
Principle #15Dynamics

2Loss of information

If actual rack force is used for steering torque generation, then feedback about vehicle behavior is accurate, but disturbances from road conditions and vehicle loading degrade the steering feel

Engineering Contradiction:
Improvefeedback quality to driverVSAvoiddisturbances from road conditions
Core Design Contradiction:
Loss of informationVSObject-affected harmful factors

Solution Approach 1:

The patent extracts and removes disturbance components (road surface irregularities, vehicle loading effects, friction variations) from the actual rack force signal. By separating these harmful factors from the useful information about vehicle behavior, the system can provide clean, accurate feedback to the driver while filtering out unwanted noise.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces modeled rack force as an intermediary between the actual rack force and the steering torque generation. This modeled force acts as a mediator that represents the essential vehicle behavior information while excluding disturbances, thereby improving feedback quality without directly using the problematic actual rack force signal.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Adaptability or versatility

If multiple models are used to determine rack force, then accuracy across different driving situations improves, but system complexity increases

Engineering Contradiction:
Improveadaptation to different driving situationsVSAvoidcontrol system complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent creates a universal control architecture that handles multiple driving situations (normal driving, parking, high-lateral acceleration) through a unified framework. The same basic structure and switching logic manage all scenarios, allowing the system to be versatile across conditions without proportionally increasing complexity.

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

Solution Approach 2:

The system uses dynamic condition-based switching to activate only the necessary model for current driving conditions. This dynamic approach ensures high adaptability to different situations while maintaining manageable complexity by avoiding the need to continuously process all possible models simultaneously.

Inventive Principle:
Principle #15Dynamics

4Ease of operation

If disturbances are filtered out completely, then steering feel is smooth, but safety-related information about road conditions is lost

Engineering Contradiction:
Improvesteering feel qualityVSAvoidsafety-related road information
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The patent selectively extracts only the harmful disturbance components (road surface irregularities, loading effects) while preserving the useful safety-related information (lateral forces, vehicle behavior). This selective extraction maintains smooth steering feel while retaining critical safety information for driver awareness.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The system provides controlled feedback to the driver through the steering torque, giving appropriate information about vehicle behavior and road conditions while filtering out unwanted disturbances. The feedback mechanism ensures safety-related information reaches the driver in a usable, non-distressing form.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP2590850B1Method to determine a rack force for the steering system of a vehicle
Publication Date: 2016.08.17 ROBERT BOSCH AUTOMOTIVE STEERING
  • EP2590850B1 patent drawingFigure 1
  • EP2590850B1 patent drawingFigure 2
  • EP2590850B1 patent drawingFigure 3

AI summary

The invention relates to a method for determining a toothed rack force on a steering device (2) in a vehicle, said toothed rack force (forZS) being ascertained dependent on a plurality of models. A toothed rack force (forZS) component (forESM) relating to a driving process is generated by means of a first model (52), and a toothed rack force (forZS) component relating to a parking process is generated by means of a second model (54).