摘要
质心侧偏角是车辆动力学中的关键变量。针对现有基于模型方法严重依赖动力学模型精度和数据驱动方法在面临陌生工况场景时鲁棒性差等问题,本文提出了一种物理-数据混合驱动(DeepPhy)的质心侧偏角估计方法,旨在结合物理模型与数据驱动模型的优势,实现对质心侧偏角的可靠与准确估计。DeepPhy通过将后轴轮胎侧向力模型得到的质心侧偏角先验值与深度网络进行集成,从而能够学习物理模型未能表达的非线性映射关系,提升模型面对陌生工况的可靠性。仿真结果表明,在连续DLC工况下,DeepPhy估计结果的RMSE相较于物理模型方法和纯数据驱动方法分别降低了93%和63%,并对数据稀缺工况具有鲁棒性。实车验证进一步表明,DeepPhy具有优异的泛化能力,经过仿真训练的模型可迁移至实车环境中,并保持高精度的估计结果。
In the realm of vehicle dynamics,the sideslip angle is a critical parameter.For the challenges posed by the current model-based methods,which heavily rely on the accuracy of dynamic models,and the poor robustness of data-driven methods in unfamiliar operating conditions,in this paper a sideslip angle estimation method based on a hybrid of physics and data-driven approaches(DeepPhy) is proposed.The aim is to combine the strength of physical modeling and data-driven techniques to achieve reliable and accurate estimation of the sideslip angle.DeepPhy integrates prior values of the sideslip angle obtained from the lateral force model of the rear axle tires with a deep neural network,enabling the learning of nonlinear mapping relationship not captured by the physical model,thereby enhancing the model′s reliability in unfamiliar conditions.The simulation results indicate that under continuous DLC conditions,the RMSE of the estimation results from DeepPhy is reduced by 93% compared to the physical model method and by 63% compared to the data-driven method,exhibiting robustness in scenarios with limited data.Real-world validation further confirms DeepPhy′s exceptional generalization capabilities,as the models trained through simulation can be transferred to real-world conditions while maintaining high-precision estimation results.
作者
李琴
张博远
谢智航
王勇
汤建明
陈勇
Li Qin;Zhang Boyuan;Xie Zhihang;Wang Yong;Tang Jianming;Chen Yong(School of Mechanical Engineering,Guangxi University,Nanning 530000;School of Mechanical Engineering,Beijing Institute of Technology,Beijing 100080)
出处
《汽车工程》
北大核心
2025年第4期714-723,共10页
Automotive Engineering
基金
广西自然科学基金青年基金(2025GXNSFBA069567)
广西科技计划桂科AD基金(23026205)资助。
