The estimation of vehicle dynamics states is crucial for vehicle stability control and autonomous driving, especially tire-force which governs vehicular motion. However, measuring tire-force directly needs expensive measurement instruments, moreover, vehicle non-linear characteristics, vehicle parameter uncertainties, unknown key variables, and sensor noise could cause great challenges in its observation. Therefore, this paper aims to develop an accurate, affordable tire-force estimator to tackle the above-mentioned issues. A novel hierarchical estimation scheme based on random-walk square-root cubature Kalman filter (SCKF) is proposed and it works without a complex tire model and considers vehicle parameter uncertainties. The estimator scheme contains three blocks. The first block estimates related key variables of tire force observation. The second block estimates both vertical tire-force and longitudinal tire-force. The longitudinal tire-force is observed based on effective tire radius identification, and a proportional integral differential (PID) method is derived based on the Lyapunov principle. Then, a random-walk SCKF algorithm is presented to estimates lateral tire-force. To validate the effectiveness of the proposed estimation scheme, CarSim & Matlab/Simulink joint simulation and real car tests are carried out. Results show that the proposed estimation scheme's accuracy performance and its potential as an affordable solution for the tire-force observation.

车辆动力学状态的估计对于车辆稳定性控制和自动驾驶至关重要，尤其是控制车辆运动的轮胎力。然而，直接测量轮胎力需要昂贵的测量仪器，此外，车辆的非线性特性，车辆参数的不确定性，未知的关键变量和传感器噪声可能会给观测带来很大的挑战。因此，本文旨在开发一种准确，负担得起的轮胎力估算器来解决上述问题。提出了一种基于随机游走的平方根库尔曼滤波（SCKF）的分层估计新方案，该方案无需复杂的轮胎模型即可工作，并且考虑了车辆参数的不确定性。估计器方案包含三个块。第一块估计轮胎力观察的相关关键变量。第二个块估计垂直轮胎力和纵向轮胎力。基于有效的轮胎半径识别，可以观察到纵向轮胎力，并根据Lyapunov原理推导比例积分微分（PID）方法。然后，提出了一种随机行走的SCKF算法来估计侧向轮胎力。为了验证所提出的估计方案的有效性，进行了CarSim＆Matlab / Simulink联合仿真和实际汽车测试。结果表明，所提出的估计方案的准确性性能及其作为轮胎力观测的负担得起的解决方案的潜力。并基于李雅普诺夫原理推导了比例积分微分（PID）方法。然后，提出了一种随机行走的SCKF算法来估计侧向轮胎力。为了验证所提出的估计方案的有效性，进行了CarSim＆Matlab / Simulink联合仿真和实际汽车测试。结果表明，所提出的估计方案的准确性性能及其作为轮胎力观测的负担得起的解决方案的潜力。并基于李雅普诺夫原理推导了比例积分微分（PID）方法。然后，提出了一种随机行走的SCKF算法来估计侧向轮胎力。为了验证所提出的估计方案的有效性，进行了CarSim＆Matlab / Simulink联合仿真和实际汽车测试。结果表明，所提出的估计方案的准确性性能及其作为轮胎力观测的负担得起的解决方案的潜力。