摄动影响下汽车主动避撞系统安全距离不确定性及μ协调控制研究

Active collision avoidance is a new technique in the field of vehicle active safety. There are multidimensional uncertain perturbation and external interference that exist in the person-car-road-environment, which make the safety distance appear uncertain characteristics and collision avoidance system performance deterioration, even endanger traffic safety. The μ control theory is an effective means to design the control system which contains perturbation factors and improve performance. This project intends to build a safe distance model and collision avoidance dynamics model including perturbation factors. Analysis of parameter perturbations and external disturbances, amplitude frequency characteristics of unmodeled errors. Application of linear fractional transformation, perturbation isolation and associated reconstruction method in μ theory, research the control system robust stability and performance robustness boundary. Analysis of the safety distance's uncertain characteristics in the perturbation effect, accurate identification method of safety distance, based on the idea of hierarchical control, application of μ theory in control system analysis and synthesis, so that the system has good dynamic performance in the effect of parameter perturbations and external disturbances, but also has both robust stability and performance robustness, better solve the contradiction between the system stability and control performance. At last improve the accuracy of calculation of safety distance and collision avoidance success rate. Providing the technical foundation and new theory for research and application of active collision avoidance.

主动避撞是汽车主动安全领域的新技术，人－车－路－环境系统中的摄动及外界干扰导致安全距离呈现不确定特性，汽车主动避撞系统性能恶化，危及交通安全。μ理论是研究含有摄动因素的控制系统设计和性能改善问题的有效手段。本项目拟构建包含摄动因素的安全距离模型及主动避撞动力学模型，分析模型中参数摄动、未建模误差及外界干扰的幅频特性，应用μ理论线性分式变换、摄动隔离及关联重构方法，研究系统鲁棒稳定性及性能鲁棒性边界。分析摄动影响下安全距离的不确定特性，提出安全距离精确辨识方法，基于分层控制思想，应用μ理论对控制系统分析与综合，使系统在参数摄动、外界干扰下既具有良好的动态性能，又同时兼顾鲁棒稳定性和性能鲁棒性，较好解决系统稳定性和控制性能之间的矛盾，以提高安全距离计算的准确性及避撞成功率。本项目研究成果将为主动避撞研究与应用提供新的理论和技术。

主动避撞系统是智能汽车的关键部件之一，其性能好坏决定了智能汽车纵向安全性能优劣。项目围绕提升安全距离模型精确性和主动避撞控制系统性能鲁棒性、稳定鲁棒性两个方面开展研究。. 在Burckhardt轮胎模型基础上，基于类比思想，创新提出一种适用于各种路面的峰值附着系数辨识算法，其辨识误差在5%-10%之内，该研究成果可以显著提高安全距离模型的准确性，为提高主动避撞系统的安全性能奠定了基础； 运用综合加权的方法分析了驾驶员状态、车辆特性及道路工况对安全车距模型计算结果的影响，提出了反应驾驶员、车辆及道路因素的DRV安全距离模型； 建立了包含主动避撞控制系统的整车动力学模型，运用μ理论分析了系统鲁棒稳定性及性能鲁棒性边界；基于传统液压制动系统，设计了并联式电控液压主动防碰撞自动制动系统，提高了智能汽车自动制动系统的性能及可靠性；针对主动避撞控制系统中存在的参数摄动、外界干扰较强的非线性时变特征，基于分层控制思想，应用 µ理论分别对节气门及制动系统设计了下位控制器，并进行参数摄动及外界干扰影响下的控制器性能仿真及整车道路试验，仿真和试验结果表明在参数摄动及外界干扰的影响下，基于µ理论设计的下位控制器能保持良好的性能鲁棒性和稳定鲁棒性。