Friction Value Estimation Using Geostatistical Kriging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining the coefficient of friction between a vehicle and the roadway are often inaccurate and require dedicated sensors, which are costly and inefficient, especially for predicting friction values on unmeasured road sections.
Innovation Solution
A method using geostatistical processes, such as kriging, to estimate the coefficient of friction by modeling spatial relationships between measured sections, incorporating data from multiple sources and sensors, and creating a friction value map that can be used to control vehicle functions, particularly in highly automated driving systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If dedicated friction measurement sensors are used to directly measure the coefficient of friction, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent uses existing vehicle sensors (acceleration, steering angle, brake pressure sensors) as intermediaries to indirectly determine the coefficient of friction. Instead of directly measuring friction with dedicated sensors, the system processes data from these existing sensors through a friction model to obtain friction values, thereby avoiding the need for complex dedicated friction measurement equipment
Solution Approach 2:
The patent replaces mechanical/direct measurement systems with a computational/model-based approach. By substituting physical friction sensors with a software-based friction determination system that processes data from existing vehicle sensors, the solution reduces hardware complexity while maintaining measurement capability
2Measurement precision
If friction values are determined for measured road sections only, then measurement precision is improved, but adaptability to unmeasured sections deteriorates
Solution Approach 1:
The system continuously receives friction values from multiple vehicles and uses this aggregated data to update and refine friction determinations for road sections. By incorporating feedback from multiple sources and using it to improve future predictions, the system enhances both accuracy and adaptability to unmeasured or newly measured sections
Solution Approach 2:
The friction determination system is designed to serve multiple road sections and vehicles simultaneously. By creating a universal friction model that can predict friction values for any road section based on aggregated data from multiple vehicles, the system achieves both precision for measured sections and adaptability for unmeasured sections
3Reliability
If data from multiple vehicles and sensors is aggregated, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent combines data from multiple vehicles and existing sensors into a unified friction determination system. By merging these diverse data sources and processing them through a common friction model, the system improves reliability through data aggregation while avoiding the need for each vehicle to independently process complex multi-source data
Data Source
Figure 1~2
Figure 3~4
AI summary
The invention relates to a method for determining a friction value for a contact between a tyre of a vehicle (102) and a road. The method involves a step of processing sensor signals (140) using a processing rule in order to generate processed sensor signals. The sensor signals (140) represent status data input by at least one detection device (104, 106, 108, 109) and which can be correlated with the friction value. The processed sensor signals represent at least one provisional friction value relating to at least one measured sub-section of the road. The method also involves a step of determining the friction value for a desired sub-section of the road using the processed sensor signals and a geostatistical process.