智能轮胎可以利用在其内部布置的传感器提供轮胎和路面的状态信息，使得底盘控制系统能够直接感知路面。然而，轮胎的高度非线性以及与路面的复杂滚动接触导致目前缺乏完善的物理模型对智能轮胎的工作机理进行准确描述，进而导致基于智能轮胎的状态估计算法开发对大量试验的依赖。此外，智能轮胎缺乏与实际底盘控制系统，尤其是悬架系统的结合。为此，本文从理论建模、感知和控制三个层面开展了一系列创新工作：在理论建模层面，提出了一种新的智能轮胎三维物理模型。该模型不仅可以从宏观角度准确地反映轮胎纵向力、侧向力随滑移参数的非线性变化以及附着饱和现象，还可以从微观角度准确地计算出车辆在加速、减速、匀速、转向以及更加复杂的行驶工况下的轮胎力、轮胎变形、接触应力以及智能轮胎信号等，从而为基于智能轮胎的状态感知算法开发奠定了理论基础。在感知层面，针对轮胎状态感知，基于智能轮胎三维物理模型，为纵向力、侧向力、垂向力以及侧偏角提出了一系列新的估计方法。为纵向力提出了基于仿真训练的估计方法，减少了算法开发对试验的依赖；为侧向力提出了基于模型的估计方法，实现了智能轮胎信号与侧向力之间数学关系的显式表征；基于双传感器智能轮胎，利用传感器冗余和多源信息融合实现了垂向力的稳健估计；提出了基于模型的带束侧向变形重建方法来进行侧偏角的准确估计。针对车辆状态感知，提出了基于路径搜索的车辆状态观测器构建方法，开发了计算软件，并建立了基于智能轮胎的车辆状态观测器。针对路面状况感知，利用智能轮胎来测量带束振动，通过试验发现了带束振动信号对悬架控制电流变化的不敏感性，提出了基于带束振动的路面不平度分类方法，保证了悬架控制过程中对路面激励的分类精度。在控制层面，在利用智能轮胎提高车辆状态感知精度的同时，提出了一种考虑轮胎非线性的稳态目标参量计算方法以进一步提升车辆稳定性控制系统的性能。在悬架控制方面，提出了智能轮胎与悬架控制器的实时通信硬件架构，使控制参数能够根据路面激励自动调节，实现了半主动悬架的自适应控制。本文阐明了智能轮胎的复杂工作机理，对轮-地关系进行了准确的建模和表征，利用智能轮胎提高了轮胎状态、车辆状态以及路面状况的感知精度，从而促进了车辆稳定性与悬架系统控制性能的有效提升。

Intelligent tires can utilize sensors arranged inside to provide the state information on tires and road surface, enabling the chassis control systems to perceive the road directly. However, the highly nonlinearity of tires and the complex rolling contact with the road result in a lack of a comprehensive physical model to accurately describe the working mechanism of intelligent tires, which further leads to a reliance on a large number of experiments for the development of state estimation algorithms based on intelligent tires. In addition, intelligent tires lack integration with actual chassis control systems, especially with suspension systems. Therefore, this thesis carries out a series of innovative work from three levels: theoretical modeling, perception, and control:At the theoretical modeling level, a new three-dimensional physical model for intelligent tires is proposed. This model can not only accurately reflect the nonlinear changes and adhesion saturation phenomenon of tire longitudinal and lateral forces with slip parameters from a macro perspective but also accurately calculate tire forces, tire deformation, contact stresses, and intelligent tire signals when the vehicle accelerates, decelerates, drives at a constant speed, steers and under more complex driving conditions from a micro perspective, laying a theoretical foundation for the development of state perception algorithms based on intelligent tires.At the perception level, for tire state perception, a series of new estimation methods for longitudinal force, lateral force, vertical force, and slip angle are proposed based on the three-dimensional physical model for intelligent tires. A training-by-simulation-based estimation method is proposed for longitudinal forces, reducing the dependence of algorithm development on experiments. A model-based estimation method is proposed for lateral forces, achieving an explicit representation of the mathematical relationship between intelligent tire signals and lateral force. Based on a dual-sensor intelligent tire, robust estimation of vertical forces is achieved through sensor redundancy and multi-source information fusion. A model-based method for reconstructing the lateral deformation of the tread band is proposed to estimate the tire slip angle accurately. For vehicle state perception, a path-searching-based method for constructing the vehicle state observer is proposed, a calculation software is developed, and an intelligent tire-based vehicle state observer is established. For road condition perception, an intelligent tire is used to measure the tread band vibration. Through experiments, it is found that the tread band vibration signals are insensitive to the changes in suspension control current. A road roughness classification method based on tread band vibration is proposed to ensure the classification accuracy of road excitation during suspension control.At the control level, while utilizing intelligent tires to improve the accuracy of vehicle state perception, a calculation method for the steady-state target parameter considering tire nonlinearity is proposed to further improve the performance of vehicle stability control systems. In terms of suspension control, a real-time communication hardware architecture between intelligent tires and the suspension controller is proposed, which enables automatic adjustment of control parameters based on road excitation, achieving adaptive control of the semi-active suspension.This thesis elucidates the complex working mechanism of intelligent tires, models and characterizes the tire-road relationship accurately, improves the perception accuracy of tire state, vehicle state and road condition, thereby promoting effective improvement of control performance of the vehicle stability and suspension systems.