Tire Wear Estimation Using Multi-Model Reliability Weighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing tire wear monitoring systems face challenges such as sensor damage, durability issues, cost, and inaccurate indirect estimation due to external parameter changes, lacking real-time combination of prediction techniques for reliable wear state estimation.
Innovation Solution
A tire wear state estimation system using sub-models and a supervisory model that combines tire and vehicle parameters to generate a comprehensive wear state estimate, employing Bayesian Networks for reliability scoring and inference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If wear sensors are placed in the tire tread to directly measure tire wear, then measurement precision is improved, but reliability deteriorates due to sensor damage during curing and durability issues
Solution Approach 1:
The patent introduces an intermediary estimation system that uses multiple indirect measurement techniques (rolling radius, slip, frictional energy, vibration, cornering stiffness, braking stiffness, footprint length) as mediators to estimate tire wear without directly placing sensors in the tire tread. This intermediary approach avoids the reliability issues of direct sensor placement while still providing wear estimation.
Solution Approach 2:
The patent segments the tire wear measurement problem into multiple independent sub-models, each using a different indirect measurement technique. By dividing the overall estimation task into separate sub-models that can be independently evaluated and combined, the system achieves reliable wear estimation without relying on a single fragile direct sensor.
2Ease of operation
If wear sensors are made small to avoid uniformity problems at high speeds, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent uses intermediary measurement techniques that do not require small sensors in the tire tread. By measuring parameters like rolling radius, vibration, and stiffness through vehicle-mounted sensors and indirect measurements, the system achieves accurate wear estimation without the constraints of small sensor size.
3Measurement precision
If multiple indirect estimation techniques are used to improve accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the complex estimation problem into multiple independent sub-models, each handling a specific indirect measurement technique. This segmentation allows the system to manage complexity by treating each sub-model separately while combining their results through a supervisory model that evaluates reliability based on operating conditions.
Solution Approach 2:
The patent introduces dynamic adaptability by having the supervisory model select and weight sub-models based on real-time operating conditions (weather, road surface, tire temperature, vehicle load). This dynamic approach allows the system to simplify operationally by activating only the most reliable sub-models for current conditions while maintaining high precision when needed.
4Measurement precision
If direct wear sensors are used to achieve accurate measurement, then measurement precision is improved, but cost increases
Solution Approach 1:
The patent replaces expensive direct wear sensors with intermediary indirect measurement techniques that use existing vehicle sensors and processing. By measuring parameters like rolling radius, vibration, and stiffness through standard vehicle-mounted sensors, the system achieves accurate wear estimation without the high cost of specialized tire-mounted wear sensors.
Data Source
AI summary
A tire wear state estimation system includes at least one tire that supports a vehicle. A sensor is mounted on the tire and measures tire parameters. At least one sensor is mounted on the vehicle and measures vehicle parameters. Each one of a plurality of sub-models receives selected tire parameters from the tire mounted sensor and selected vehicle parameters from the vehicle mounted sensor. Each one of the sub-models generates a sub-model wear state estimate, and a model reliability is determined for each one of the sub-models. A supervisory model receives the wear state estimate from each sub-model and the model reliability for each sub-model, and generates a combined wear state estimate for the tire.