18182494. SYSTEMS AND METHODS FOR NEURAL ORDINARY DIFFERENTIAL EQUATION LEARNED TIRE MODELS simplified abstract (TOYOTA JIDOSHA KABUSHIKI KAISHA)
Contents
SYSTEMS AND METHODS FOR NEURAL ORDINARY DIFFERENTIAL EQUATION LEARNED TIRE MODELS
Organization Name
TOYOTA JIDOSHA KABUSHIKI KAISHA
Inventor(s)
Yan Ming Jonathan Goh of Palo Alto CA (US)
Franck Djeumou of Palo Alto CA (US)
SYSTEMS AND METHODS FOR NEURAL ORDINARY DIFFERENTIAL EQUATION LEARNED TIRE MODELS - A simplified explanation of the abstract
This abstract first appeared for US patent application 18182494 titled 'SYSTEMS AND METHODS FOR NEURAL ORDINARY DIFFERENTIAL EQUATION LEARNED TIRE MODELS
Simplified Explanation:
This patent application describes a method for creating NODE learned tire models by calculating estimated tire forces based on vehicle measurements and solving a second order differential equation to minimize error in the tire force function.
- Using predictive models to provide inflection points and initial conditions
- Integrating exponential equations to obtain the tire force function
- Applying the tire force function to estimate current tire forces based on new vehicle measurements
Key Features and Innovation:
- Utilizing predictive models to improve accuracy in estimating tire forces
- Solving differential equations to minimize error in the tire force function
- Integrating vehicle measurements to create a more precise tire model
Potential Applications:
- Vehicle dynamics simulation
- Autonomous driving systems
- Tire performance optimization
Problems Solved:
- Inaccuracies in estimating tire forces
- Lack of precision in tire models
- Difficulty in integrating vehicle measurements into tire modeling
Benefits:
- Improved accuracy in predicting tire forces
- Enhanced vehicle performance
- Better understanding of tire behavior
Commercial Applications:
- Automotive industry for vehicle design and testing
- Tire manufacturers for product development
- Research institutions for advanced vehicle dynamics studies
Questions about NODE learned tire models: 1. How do NODE learned tire models improve upon traditional tire modeling techniques? 2. What are the potential limitations of using NODE learned tire models in real-world applications?
Original Abstract Submitted
System, methods, and other embodiments described herein relate to NODE learned tire models. In one embodiment, a method includes calculating estimated tire forces based on vehicle measurements; solving a second order differential equation in a repetitive manner until an error calculation based on a tire force function and the estimated tire forces reaches a minimum value, by: using a first predictive model to provide one or more inflection points and initial conditions based on the vehicle measurements, using a second and third predictive model to act as, respectively, exponents to a positive and a negative exponential equation based on the one or more inflection points, the initial conditions, and the vehicle measurements, and integrating the exponential equations to obtain the tire force function; and applying the tire force function to new vehicle measurements to estimate current tire forces.