基于驾驶人使用模式的汽车底盘集成控制研究

发布时间：2018-01-22 15:39

本文关键词： 车辆工程底盘集成控制驾驶人使用模式仿生智能算法多变量频域控制　出处：《吉林大学》2016年硕士论文　论文类型：学位论文

【摘要】：汽车底盘集成控制以整车性能最优为目标,能够综合协调汽车底盘各动力学子系统,已经成为现今车辆动力学研究领域的热点问题。然而,当前底盘集成控制系统大多基于车辆动力学特性进行均一设计,忽略了驾驶人对控制系统的作用。事实上,驾驶人与主动控制系统对车辆的作用具有强烈的耦合关系,因此,汽车底盘集成控制系统设计过程中应考虑驾驶人特性的影响。本文依托国家自然科学基金(项目编号51105169、51475206)等项目,开展了基于驾驶人使用模式的汽车底盘集成控制研究,设计并搭建了驾驶人使用模式数据采集系统,进行了典型工况下的驾驶人使用模式数据采集;引入仿生智能算法对驾驶人使用模式进行了辨识;在此基础上,基于改进多变量频域控制算法建立了基于驾驶人使用模式的汽车底盘集成控制策略,并选取典型工况利用MATLAB/Simulink与Car Sim联合仿真平台进行了验证分析。本文的主要研究内容包括:(1)驾驶人使用模式数据采集为了进行驾驶人使用模式的分析、识别和量化,首先基于d SPACE DS1006实时仿真平台设计搭建了驾驶人使用模式数据采集系统。利用所搭建的系统对典型工况下的驾驶人使用模式数据进行了采集,所采集数据包括驾驶人方向盘转角、油门踏板和制动踏板开度以及车辆状态信息等。(2)基于仿生智能算法的驾驶人使用模式辨识引入驾驶人预瞄最优曲率模型来描述驾驶人使用模式,通过预瞄时间、神经延迟时间和肌肉惯性延迟时间来表征不同类型驾驶人使用模式的差异,在此基础上,分别采用改进粒子群算法和BP神经网络算法设计了驾驶人使用模式离线和快速在线辨识策略,并对辨识结果进行了分析验证。(3)基于驾驶人使用模式的汽车底盘集成控制策略设计提出了基于驾驶人使用模式的汽车底盘集成控制策略架构。通过对驾驶人使用模式的分析、识别、量化和建模,得到预测的驾驶人方向盘转角,将预测方向盘转角输入给二自由度车辆参考模型,从而建立基于驾驶人使用模式的控制系统参考模型。分析了主动转向和基于主动制动的电子稳定性控制之间的干涉和耦合,采用考虑了车速变化和轮胎侧偏刚度非线性的改进逆奈奎斯特阵列法(Inverse Nyquist Array method,INA)进行了集成控制器的设计,有效消除了两种主动控制系统之间的耦合。(4)控制策略验证搭建了MATLAB/Simulink与Car Sim联合仿真平台,选取双移线和蛇形两种工况,在不同车速、不同路面附着系数的情况下对控制策略进行了仿真测试。结果表明,所提出的控制策略能够有效提高车辆的轨迹跟随能力,有效提高车辆的操纵稳定性能,有效减轻驾驶人的操作负荷。
[Abstract]:The vehicle chassis integrated control, which aims at the optimal performance of the whole vehicle, can comprehensively coordinate the dynamic subsystems of the vehicle chassis, which has become a hot issue in the field of vehicle dynamics research. At present, the chassis integrated control system is mostly based on the vehicle dynamics characteristics for uniform design, ignoring the role of the driver on the control system. The driver and the active control system have the strong coupling relation to the vehicle function, therefore. The influence of driver's characteristics should be considered in the design of automobile chassis integrated control system. This paper relies on the National Natural Science Foundation of China (project number 51105169 / 51475206) and so on. The integrated control research of automobile chassis based on driver usage mode is carried out, and the data acquisition system of driver usage mode is designed and built, and the data collection of driver usage mode is carried out under typical operating conditions. The bionic intelligent algorithm is introduced to identify the driving mode. On this basis, a vehicle chassis integrated control strategy based on driving mode is established based on the improved multivariable frequency domain control algorithm. At the same time, the typical working conditions are selected and verified by the joint simulation platform of MATLAB/Simulink and Car Sim. The main research contents of this paper include: 1). The driver uses the pattern data collection to carry on the driver to use the pattern analysis. Identification and quantification. First based on d SPACE. DS1006 real-time simulation platform designed and built the driver use mode data acquisition system, using the built system to the typical operating conditions of the driver use mode data collection. The data collected include the steering wheel angle of the driver. Throttle pedal and brake pedal opening degree and vehicle state information etc.) based on the bionic intelligent algorithm driver use pattern identification introduces the driver preview optimal curvature model to describe the driver use mode. Preview time, nerve delay time and muscle inertia delay time were used to characterize the difference of different drivers' usage patterns. Improved particle swarm optimization algorithm and BP neural network algorithm are used to design off-line and fast on-line identification strategies for drivers. The identification results are analyzed and verified. The design of vehicle chassis integrated control strategy based on driver usage mode. The framework of vehicle chassis integrated control strategy based on driver usage mode is proposed. The predicted steering wheel angle is obtained by identifying, quantifying and modeling, and the predicted steering wheel angle is input to the two-degree-of-freedom vehicle reference model. The reference model of control system based on driving mode is established, and the interference and coupling between active steering and electronic stability control based on active braking are analyzed. An improved inverse Nyquist Array method method, which takes into account the variation of vehicle speed and the nonlinearity of tire lateral stiffness, is used to improve the inverse Nyquist array method. INA) is used to design the integrated controller. The coupling between two active control systems is eliminated effectively. The verification of control strategy builds a joint simulation platform of MATLAB/Simulink and Car Sim. The control strategy is simulated under the conditions of different speed and different road adhesion coefficient under two working conditions of double moving line and snake shape. The results show that. The proposed control strategy can effectively improve the tracking ability of the vehicle, improve the handling and stability of the vehicle, and effectively reduce the operating load of the driver.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U463.1

【相似文献】