Virtual Tire Sensor Using CAN Data for Real-Time Lateral Guidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional tire monitoring systems fail to consider real-time and real-world conditions, leading to underutilization of tire and vehicle potential, as they rely on predetermined assumptions rather than dynamic tire performance data.
Innovation Solution
A method using a neural network within a vehicle's control system to determine real-time tire performance factors based on CAN data, including vehicle dynamics and pilot inputs, allowing for dynamic adjustment of lateral guidance performance profiles to optimize cornering capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional tire monitoring systems use predetermined assumptions for tire performance, then the system complexity is reduced, but the measurement precision and reliability of tire performance data deteriorates
Solution Approach 1:
The patent creates a virtual tire sensor that copies the functionality of physical tire sensors by using a neural network to process CAN data and generate virtual sensor readings. This approach achieves measurement precision comparable to physical sensors while avoiding the complexity of installing and maintaining actual sensor hardware on the tires.
Solution Approach 2:
The patent replaces the mechanical/physical tire sensor system with a computational model (neural network) that processes existing vehicle data. This substitution eliminates the need for physical sensors while providing dynamic, real-time tire performance estimates that adapt to changing conditions.
2Measurement precision
If physical tire sensors are installed to obtain real-time tire data, then the measurement precision improves, but the device complexity and cost increase
Solution Approach 1:
The virtual tire sensor replicates the measurement capabilities of physical sensors by processing existing vehicle data through a trained neural network, providing equivalent information without the hardware installation complexity.
Solution Approach 2:
The system uses existing vehicle data infrastructure (CAN bus data already present in the vehicle) to generate tire performance information, eliminating the need for additional sensors and their associated installation and maintenance requirements.
3Ease of operation
If predetermined tire performance assumptions are used, then the ease of operation is improved, but the adaptability to real-world conditions deteriorates
Solution Approach 1:
The neural network model dynamically adapts tire performance estimates based on real-time vehicle operating conditions by processing live CAN data, allowing the system to adjust to changing conditions automatically while maintaining ease of operation through automated computation.
Solution Approach 2:
The system continuously processes new vehicle data through the neural network, creating a feedback loop that constantly updates tire performance estimates based on current conditions, enabling the system to adapt to real-world variations without manual intervention.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Devices, systems, and methods related to prediction of tire performance using existing CAN data to improve overall vehicle performance. Machine learning tools are applied to CAN data, for example pilot data and/or vehicle dynamics data, to predict tire performance factors for use in a vehicle control system to provide vehicle lateral guidance control.