Abstract:In this paper, we follow the physics guided modeling approach and integrate a neural differential equation network into the physical structure of a vehicle single track model. By relying on the kinematic relations of the single track ordinary differential equations (ODE), a small neural network and few training samples are sufficient to substantially improve the model accuracy compared with a pure physics based vehicle single track model. To be more precise, the sum of squared error is reduced by 68% in the considered scenario. In addition, it is demonstrated that the prediction capabilities of the physics guided neural ODE model are superior compared with a pure black box neural differential equation approach.

What problem does this paper attempt to address?

### What problems does this paper attempt to solve? This paper aims to solve the problems of accuracy and real - time performance in vehicle dynamics modeling. Specifically, by combining neural ordinary differential equations (NODEs) with physical models, the author proposes a hybrid model to improve the prediction accuracy and computational efficiency of the vehicle single - track model. The following are the specific objectives of this study: 1. **Improve model accuracy**: Although traditional physics - based vehicle single - track models can provide a certain level of accuracy, they may not be able to fully capture the real behavior of the vehicle in some dynamic situations. In this paper, neural networks, especially neural differential equations, are introduced to enhance the expressive power of the model, thereby significantly improving the prediction accuracy of the model. 2. **Reduce the need for training data**: Pure black - box models (such as pure neural network models) usually require a large amount of training data to achieve high accuracy. The hybrid model proposed in this paper combines physical prior knowledge and a small amount of training data, and can maintain high prediction performance while reducing the data requirements. 3. **Meet the requirements of real - time computing**: In the application of autonomous vehicles, real - time computing is very important. The hybrid model proposed in this paper not only improves the prediction accuracy, but also can meet the requirements of real - time computing, thus providing better support for practical applications. 4. **Verify the advantages of the hybrid model**: By conducting comparative experiments on different models (simplified white - box models, pure black - box neural ODE models, and hybrid UDE models), the advantages of the hybrid model in terms of prediction ability and training data requirements are verified. ### Main contributions - **Propose a new hybrid model**: Combine the physical model and the neural differential equation to form a new hybrid model (UDE) for simultaneously simulating the longitudinal and lateral dynamics of the vehicle. - **Significantly improve prediction accuracy**: Compared with the traditional physical model, the sum of squared errors of the hybrid model in the considered scenario is reduced by 68%. - **Reduce the need for training data**: The hybrid model can achieve high - precision prediction with only a small amount of training data, which is of great significance in practical applications. - **First application in combined longitudinal and lateral vehicle dynamics**: As far as the author knows, this is the first time that this hybrid modeling method has been applied to the study of combined longitudinal and lateral vehicle dynamics. Through these improvements, this research provides a more efficient and accurate method for vehicle dynamics modeling, especially suitable for application scenarios with high requirements for real - time performance and accuracy, such as autonomous vehicles.

Vehicle single track modeling using physics guided neural differential equations

Nonlinear Model Predictive Path Tracking Control for Autonomous Vehicles Based on Orthogonal Collocation Method

A high-performance neural network vehicle dynamics model for trajectory tracking control

Deep Dynamics: Vehicle Dynamics Modeling with a Physics-Constrained Neural Network for Autonomous Racing

Intelligent vehicle trajectory tracking control based on physics-informed neural network dynamics model

Vehicle Lateral Dynamics-Inspired Hybrid Model Using Neural Network for Parameter Identification and Error Characterization

A Hybrid Lateral Dynamics Model Combining Data-driven and Physical Models for Vehicle Control Applications

Fine-Tuning Hybrid Physics-Informed Neural Networks for Vehicle Dynamics Model Estimation

Neural modal ordinary differential equations: Integrating physics-based modeling with neural ordinary differential equations for modeling high-dimensional monitored structures

Physics Embedded Neural Network Vehicle Model and Applications in Risk-Aware Autonomous Driving Using Latent Features

Neural Network Tire Force Modeling for Automated Drifting

Estimating model inadequacy in ordinary differential equations with physics-informed neural networks

Modeling Trajectories with Neural Ordinary Differential Equations

Bidirectional Dynamic Neural Networks with Physical Analyzability

Deep-Neural-Network-Based Modelling of Longitudinal-Lateral Dynamics to Predict the Vehicle States for Autonomous Driving

Trajectory tracking control of autonomous vehicles based on Lagrangian neural network dynamics model

Hybrid physics-data-driven online modelling: Framework, methodology and application to electric vehicles

Kinematics-aware Trajectory Generation and Prediction with Latent Stochastic Differential Modeling

Longitudinal Vehicle Dynamics: A Comparison of Physical and Data-Driven Models Under Large-Scale Real-World Driving Conditions

Vehicle State Estimation Combining Physics-Informed Neural Network and Unscented Kalman Filtering on Manifolds

Bi-level Trajectory Optimization on Uneven Terrains with Differentiable Wheel-Terrain Interaction Model