一种融合物理约束的水陆两栖车动力学参数动态识别方法

By combining four-dimensional dynamic clustering and multilayer perceptron models with physical constraints and optimizing neural network parameters, the problem of identifying dynamic parameters of amphibious vehicles under complex working conditions was solved, achieving high-precision dynamic modeling and autonomous adaptive control.

CN121980813BActive Publication Date: 2026-07-17HEFEI UNIV OF TECH
View PDF 4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
HEFEI UNIV OF TECH
Filing Date
2026-02-10
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Traditional amphibious vehicle dynamics modeling cannot accurately describe the nonlinearity, time-varying nature, and coupling caused by drastic changes in the medium and environment. Existing methods cannot achieve continuous and unified parameter identification across all media, especially at the water-land interface and in complex working conditions, where the parameter identification results are coarse and cannot meet the requirements of amphibious missions.

Method used

By employing a method that integrates physical constraints, a multilayer perceptron model is constructed to identify working conditions through four-dimensional dynamic clustering. By combining boundary conditions, initial conditions, and data in-point loss, the neural network parameters are optimized to achieve dynamic identification of dynamic parameters.

Benefits of technology

It improves the accuracy of vehicle dynamics modeling and environmental adaptability under complex working conditions on land and water, realizes high-precision identification and autonomous adaptation of dynamic parameters, and supports high-performance motion control of vehicles in multi-task environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121980813B_ABST
    Figure CN121980813B_ABST
Patent Text Reader

Abstract

本发明适用于车辆动力学技术领域，提供了一种融合物理约束的水陆两栖车动力学参数动态识别方法，该方法包括以下步骤：根据关键状态量进行工况识别，得到工况分类结果；基于工况分类结果，采用神经网络模型从工况分类后的状态数据学习动力学参数相关特征；基于水陆两栖车辆动力学模型，构建融合物理机理的损失函数，并结合边界条件损失、初始条件损失及数据内点损失，得到融合物理约束的加权损失函数；基于加权损失函数进行反向传播，优化神经网络模型的参数及车辆动力学参数，实现动力学参数动态识别。本发明可以提升车辆在陆上和水上多种复杂运行场景下的工况识别精度，以及实现车在陆上和水上多种复杂运行场景下的动力学参数动态识别。
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Self-adaptive identification method for kinetic parameters of self-driving automobile based on SQP and GRNN

    CN119644724A

  • Vehicle kinetic parameter identification method based on constrained nonlinear optimization

    CN121167872A

  • Tire force estimation method based on physical information fusion recurrent neural network

    CN119047303A

  • Wheel hub motor vehicle control method and device based on curve working condition and storage medium

    CN120756458A