Steering Angle Detection Using Nonlinear Learning Maps

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional electric power steering systems face challenges in accurately estimating motor shaft and steering shaft angles due to nonlinear elements in the reduction mechanism, leading to difficulties in diagnosing and backing up angle sensors, which affects reliability and increases costs.

Innovation Solution

A vehicle steering angle detecting apparatus with dual-system angle sensors that learn and estimate motor shaft and steering shaft angles by dividing nonlinear elements into static, dynamic, and delay characteristics, allowing for accurate backup and abnormality judgment, even when one sensor fails.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If dual-system angle sensors are equipped with steering shaft and motor shaft, then reliability and redundancy are improved, but device complexity and cost increase

Engineering Contradiction:
Improvesensor reliabilityVSAvoidsensor system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system uses feedback control by comparing the steering shaft angle detected by the first angle sensor with the motor shaft angle detected by the second angle sensor, and adjusts the motor current based on the angle difference to maintain accurate steering assistance even when one sensor fails

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The control unit acts as an intermediary that processes data from both angle sensors, calculates the angle difference, and determines sensor abnormality, enabling the system to utilize information from both sensors to maintain reliability without simply duplicating the entire sensor system

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If nonlinear elements of reduction mechanism are learned and compensated, then measurement precision of angle estimation is improved, but device complexity and processing time increase

Engineering Contradiction:
Improveangle estimation precisionVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system pre-acquires steering shaft angle and motor shaft angle data during normal operation, and performs learning processing to create compensation value maps for nonlinear elements before they are needed, so that accurate angle estimation can be achieved without complex real-time calculations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The learning processing is performed dynamically based on actual operating conditions, with the system adaptively updating compensation values for static characteristics, dynamic characteristics, and delay characteristics according to the specific steering工况, rather than using fixed complex models

Inventive Principle:
Principle #15Dynamics

3Manufacturing precision

If learning processing is performed for static, dynamic, and delay characteristics, then manufacturing precision of angle compensation is improved, but loss of time for learning increases

Engineering Contradiction:
Improveangle compensation precisionVSAvoidlearning time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The learning processing is performed periodically under specific steering conditions (such as when the vehicle is stationary or during normal steering operation), dividing the learning process into manageable segments for static characteristics, dynamic characteristics, and delay characteristics rather than requiring continuous extensive learning

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The system performs learning processing selectively based on whether the current operating conditions meet the predetermined criteria, and only learns the necessary characteristics (static, dynamic, or delay) based on the specific situation, rather than always performing complete learning for all three characteristics

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3345808B1Vehicle steering angle detection device and electric power steering apparatus equipped with same
Publication Date: 2019.12.11 NSK LTD
  • EP3345808B1 patent drawingFigure 1
  • EP3345808B1 patent drawingFigure 2
  • EP3345808B1 patent drawingFigure 3

AI summary

[Problem] An object of the present invention is to provide a vehicle steering angle detecting apparatus and an electric power steering apparatus equipped therewith having reasonable price that estimate a motor shaft angle and a steering shaft angle with a high accuracy by learning nonlinear elements, and can back up both angle sensors by utilizing estimating angles of the both angle sensors. [Means for solving the problem] The present invention is the vehicle steering angle detecting apparatus that connects a motor, which assist-controls a steering system of a vehicle, to the steering shaft via a reduction mechanism, and comprises the first angle sensor which detects the steering shaft angle of the steering shaft and the second angle sensor which detects the motor shaft angle of the motor, comprising: a function that updates a static characteristic map and a dynamic characteristic map by iteratively learning a static characteristic and a dynamic characteristic of the nonlinear elements including the reduction mechanism, and mutually estimates the steering shaft angle and the motor shaft angle by using the static characteristic map and the dynamic characteristic map. The present invention is also the electric power steering apparatus equipped with the vehicle steering angle detecting apparatus.