面向绿色交通的智能车辆变工况行驶能耗反馈与耗散控制方法

Aiming at the concept of Green Transportation, a variable conditions based engergy feedback and dissipative control for Intelligent vehicle are carried out considering the integration of information sharing between vehicle and road. This study focuses on intelligent vehicle equiped with longitudinal driver assistance system, analyzes the perenergy dissipative characteristics under variable driving conditions, and developes the energy storage function integrated by safety, comfort, and fuel economy. After extracting key eigenvectors which represent the connected vehicle-road conditions, the mapping relationship between these eigenvectors and the energy storage function is solved. An inverse dynamics based energy feedback method is designed, and the Quantitative identification for the energy consumption under variable driving conditions is made. In terms of the the performance critema of Gamma dissipation, a energy storage function matching method is proposed by Backstepping based Lyapunov method and algebraic method, and the multi-objective coordinated longitudinal dissipative control algorithm is designed, which is verified by the hardware-in-the-loop based intelligent vehicle platform designed. The above works are useful for providing new ideas and methodology for both advanced driver assistant systems and autonomous driving systems.

围绕绿色交通发展理念，着眼于车路一体化信息共享，研究智能车辆变工况行驶的能耗反馈与耗散控制方法。项目以具备纵向辅助驾驶功能的智能车辆为研究对象，剖析智能车变工况行驶过程中的能量耗散特性，构建综合安全跟踪、乘坐舒适、燃油经济等性能指标的能量存储函数。提取表征车路互联工况的关键特征向量，解析其与智能车能量存储函数的映射关系。设计基于逆动力学分析的能耗反馈方法，构建变工况能耗的定量化辨识方法。针对γ耗散性能准则，提出基于Backstepping的Lyapunov直接法与“配方法”的能量存储函数匹配方法，设计基于多目标协调的智能车纵向耗散控制算法。构建智能车辆耗散动力学系统的硬件在环实验平台，并对控制算法进行验证。研究成果可以为汽车安全辅助驾驶与自动驾驶系统提供新思路与新方法。

围绕绿色交通发展理念，着眼于车路一体化信息共享，研究智能车辆变工况行驶的能耗反馈与耗散控制方法。采用本项目提出的智能车辆变工况行驶能耗反馈与耗散控制方法可以将智能车辆纵横向耦合运动的最优控制转化成以能量存储函数为优化目标的耗散控制问题。利用基于Backstepping的Lyapunov直接法与“配方法”设计的智能车纵向耗散控制律可以避免求解Riccati方程或代数迭代，这样也方便借助逐步逼近γ耗散不等式的形式，得到耗散控制的最优解。利用逐步逼近耗散控制的最优解，以达到智能车控制系统对不同工况的安全跟踪、乘坐舒适、燃油经济等多目标的最优决策，并且在仿真环节中未对刹车/油门等执行机构进行建模，而是直接将控制输入施加于车辆系统的情况下，得到了γ供给率（逼近值）可以节省5%的能量消耗率的结论，这无异于可以在保证安全行驶的基础上大大提高智能车的燃油经济性。最后利用构建的车路协同的缩微交通与缩微智能车仿真平台，对项目提出的控制方法进行了实验验证。研究成果可以为汽车安全辅助驾驶与自动驾驶系统提供新思路与新方法。