当代汽车工业进入智能时代，汽车技术迎来颠覆性变革。智能汽车是当今汽车科技创新的前沿，是我国建设汽车强国和交通强国的重要抓手，更是未来大国竞争的关键领域。底盘动力学智能控制是智能驾驶的基础。车辆行驶过程中，高复杂且不确定的交通状况、动力学非线性给底盘安全协调作动带来很大挑战：非线性特性导致动力学状态难以精确获取；复杂、不确定交通环境极易诱发底盘失稳；随着底盘电子化高度发展，如何协调各个子系统以保证动力学稳定性这一难题也亟需解决。本课题即针对智能汽车底盘动力学智能控制面临的挑战开展工作。 首先剖析底盘各部件工作特性并分析智能驾驶系统对底盘控制执行的需求，设计了底盘动力学域控制总体架构。其主要包括基于全局状态获取的模型解算和多目标多系统协调机制两大部分。需研究该技术以保证智能汽车稳定安全运行。 在全局动力学参数获取研究方面，针对轮胎动力学参数难以直接测量难题，提出了一种统一超螺旋滑模观测体系，实现了四个车轮的垂向、纵向、侧向力以及轮胎侧偏角的精确单独观测。针对稳定性关键参数观测难题，设计了零点重置积分法和自适应容积卡尔曼滤波器，进而通过自适应积分校正融合实现车身侧偏角的精确估计。通过递推最小二乘辨识和补偿算法综合估计出路面附着系数。基于轮胎六分力测试设备搭建全工况汽车动力学状态观测实验平台，进行了实验验证。 在底盘动力学稳定性协调域控研究方面，针对保证纵向智能驾驶辅助功能执行中动力学稳定性这一问题，设计了基于模型预测控制的纵向避障与动力学稳定性自适应协调控制算法，保证不确定工况下，车辆自动紧急刹车时的底盘动力学稳定性；提出了自适应巡航与直接横摆力偶矩集成控制的多目标自适应巡航控制方法，保证跟车过程中的底盘动力学稳定性，并搭建硬件在环台架进行实验验证。 针对智能驾驶辅助横向控制与动力学稳定性协调控制难题，首先设计考虑整车参数不确定性的基于鲁棒模型预测控制的路径跟踪控制器。进一步地，提出了基于前轮主动转向和直接横摆力偶矩协调控制的H∞控制器，考虑参数不确定性下通过转向和制动多系统协调保证动力学稳定性，并基于台架进行实验分析。 课题研究实现了底盘动力学关键状态的精确观测和稳定性多目标多系统协调控制，有望攻克智能汽车底盘动力学的模型失准、控制失配问题。

Nowadays, the automobile industry enters the intelligent age, and the automotive technology enters the stage of subversive change and development. The automated vehicle becomes the frontier of modern automobile science and technology innovation, is the important grasp of building automobile and transportation power in our country, and is the key field of confrontation between big powers in the future. Autonomous driving is on the basis of chassis dynamics intelligent control. The high complexity and uncertain traffic conditions and vehicle dynamics nonlinearity in the process of vehicle driving bring great challenge to the safe and coordinated execution of chassis subsystems. Vehicle dynamics nonlinearity makes it difficult to obtain vehicle dynamics states accurately. Complex unknown road environment and uncertain traffic condition can easily induce chassis dynamics instability. Additionally, with the rapid development of chassis electronics, the increasing chassis subsystems bring difficult problem of coordinating chassis subsystems to ensure dynamic stability and it also needs to be tackled urgently. This Ph.D. dissertation focuses on crucial issues of chassis dynamics intelligent control, and has conducted massive work on key technologies of chassis dynamics domain control for automated vehicles. Firstly, the research proposes the overall architecture of chassis dynamics integration and domain control based on the analysis of work characteristic of each chassis component and the dynamics requirements of intelligent functions' control and implementation. It mainly contains two parts: resolving of vehicle dynamics model based on the acquisition of global dynamics states and coordinated control mechanism of multi-objective and multi-subsystem. Both need to be studied to ensure vehicle dynamics stability and safe operation. On the acquisition of dynamics sates, this research addresses the estimation of tire dynamics parameters. A novel unified super twisting sliding mode observation system is proposed to observe accurately the vertical, longitudinal, lateral tire forces and tire side slip angle of each tire respectively. To tackle the observation of crucial parameters of dynamics stability, this research designs the zero-point-reset method and adaptive cubature Kalman filter, based on which, proposes an adaptive cubature Kalman filter based estimator with the integral correction fusion to estimate vehicle side slip angle accurately. In addition, this research synthesizes the recursive least-squares identification method and compensation algorithm to obtain the tire friction coefficient. An experimental platform for the vehicle dynamics observation under full working conditions is established based on the wheel force transducer, and experimental tests are carried out. On the chassis coordination and domain control, this research focuses on chassis dynamics stability of the execution of longitudinal intelligent assist functions. To ensure chassis dynamics stability of emergency braking under uncertain conditions, a model predictive control based lateral stability coordinated collision avoidance control algorithm is proposed. A multi-objective adaptive cruise control integrated with direct yaw moment control is designed to ensure car following performance on the premise of chassis dynamics stability. A hardware in the loop test bench is established, and experimental tests are carried out to validate the effectiveness of the proposed controller. Furthermore, this research solves the coordinated control of intelligent assist lateral control and dynamics stability. Firstly, this research presents a robust model predictive control based path tracking controller, which takes vehicle parameter uncertainties into consideration. Then, considering parameter uncertainties, a H∞ controller integrating active front steering control and direct yaw moment control is proposed to guarantee chassis dynamics stability via the coordinated control of steering and braking subsystem. Experimental results of hardware in the loop are analyzed. This research realizes accurate estimation of chassis dynamics key states and coordinated stability control of multi-objective and multi-subsystem. It promises to overcome model inaccuracy and control mismatch of automated vehicle chassis dynamics.