车辆模型预测巡航控制系统的分析与设计

发布时间：2018-01-08 17:03

本文关键词：车辆模型预测巡航控制系统的分析与设计　出处：《浙江工业大学》2016年硕士论文　论文类型：学位论文

【摘要】：车辆自适应巡航控制(ACC)系统是一种先进的车辆辅助驾驶系统,它能减轻驾驶人员的工作负担、减少错误驾驶和疲劳驾驶的状况,提高驾驶的舒适性和安全性以及道路通行能力,减少燃料消耗。传统ACC指在跟踪前面车辆的行驶中保持期望的车辆间距或保持期望的行驶车速。近年来,国内外相关学者对提高车辆燃油经济性和减少车辆尾气排放做了深入的研究,提出了车辆经济驾驶概念。因此,如何实现车辆经济性巡航控制是车辆ACC技术亟待解决的难点和热点问题,具有重要的理论意义和应用价值。本文在总结现有车辆巡航控制技术的基础上,运用理想点优化法和模型预测控制技术,提出了一种新的车辆多目标预测巡航控制策略,使车辆巡航系统的各个目标性能逼近理想最优点。本文主要研究工作与创新点如下:(1)定性分析了车辆巡航系统的多个目标性能之间的冲突性与矛盾性。通过数学建模将各目标性能用数学公式表示出来。(2)现有车辆巡航控制系统大多针对单个目标进行控制,无法满足车辆多个目标性能之间的同步优化控制。论文将多目标优化和模型预测控制引入到ACC系统的设计与分析中,提出了一种新的多目标模型预测巡航控制策略来解决车辆巡航控制系统中多目标性能的优化问题。这种新的预测巡航控制器使得本车可以跟踪前方车辆的行驶并且使得本车具有最优的燃油经济性。在车辆巡航控制系统中引入了理想点概念的方法。它主要是使总的目标优化函数到各自独立目标函数最小值矢量点的间距离最小化来达到优化目的。该方法的主要优点是不需要对各目标函数加上相应的惩罚项,即不需要对各目标函数的重要性进行人工的加权干预。使得车辆ACC系统具有更好的渐进稳定性。这种新算法主要是把理想点法加入到模型预测控制(MPC)方法中得到一个理想点跟踪的预测巡航控制器。进一步,结合车辆巡航控制过程中的内在外在的约束条件(如车辆速度大小、油门踏板和刹车踏板的开度等)建立车辆巡航控制系统的状态空间模型。并在不同的车辆运行工况下,将该新算法的车辆运行仿真结果与采用常规的加权模型预测控制(MPC)的算法和采用智能驱动模型(IDM)算法的车辆控制进行实验对比。显示出本文所提方法的实用性和优越性。(3)设计并开发了车辆自适应巡航控制系统演示软件,比较分析了多种自适应巡航控制算法控制效果。选择JAVA语言作为软件开发的编程语言。因为其面向对象编程的特性,能够更好的展现出我们所需要的图形化效果,使得算法的内容可视化。而且该语言具有跨平台的特性,使的软件可以不受计算机硬件和操作系统的约束而在任意计算机环境下正常运行。
[Abstract]:The vehicle adaptive cruise control system (ACCS) is an advanced vehicle-assisted driving system, which can reduce the workload of drivers and reduce the condition of wrong driving and fatigue driving. Improve driving comfort and safety and road capacity, reduce fuel consumption. Traditional ACC refers to keeping the desired vehicle spacing or speed in tracking the vehicles in front of the vehicle. In recent years. Related scholars at home and abroad have done in-depth research on improving vehicle fuel economy and reducing vehicle exhaust emissions, and put forward the concept of vehicle economic driving. How to realize the vehicle economic cruise control is a difficult and hot problem to be solved urgently in vehicle ACC technology, which has important theoretical significance and application value. This paper summarizes the existing vehicle cruise control technology. Using ideal point optimization method and model predictive control technology, a new vehicle multi-objective predictive cruise control strategy is proposed. In this paper, the main research work and innovation are as follows: 1). This paper qualitatively analyzes the conflict and contradiction between the performance of multiple targets of vehicle cruise system. The performance of each target is represented by mathematical formula by mathematical modeling. Most of the existing vehicle cruise control systems are aimed at a single target. This paper introduces multi-objective optimization and model predictive control into the design and analysis of ACC system. A new multi-objective model predictive cruise control strategy is proposed to solve the multi-objective performance optimization problem in the vehicle cruise control system. This new predictive cruise controller enables the vehicle to track the vehicle in front of the vehicle. The concept of ideal point is introduced into the vehicle cruise control system. The method is mainly to make the distance between the total objective optimization function and the minimum vector point of the respective independent objective function the distance between the total objective optimization function and the minimum vector point of the respective independent objective function. The main advantage of this method is that it does not need to add corresponding penalty terms to each objective function. That is, the importance of each objective function does not need to be artificially weighted, which makes the vehicle ACC system more asymptotically stable. This new algorithm mainly adds the ideal point method to the predictive control of the model. An ideal point tracking predictive cruise controller is obtained in the MPC method. Combined with the internal and external constraints (such as vehicle speed) in the vehicle cruise control process. The opening degree of throttle pedal and brake pedal etc.) the state space model of vehicle cruise control system is established. The simulation results of the new algorithm are compared with the conventional weighted model predictive control (MPC) algorithm and the intelligent driving model (IDM). Experimental results show that the proposed method is practical and superior. (3) the demonstration software of vehicle adaptive cruise control system is designed and developed. The control effects of various adaptive cruise control algorithms are compared and analyzed. JAVA language is chosen as the programming language for software development because of its object-oriented programming characteristics. It can better show the graphical effect that we need, make the content visualization of the algorithm, and the language has the characteristics of cross-platform. The software can operate normally in any computer environment without the restriction of computer hardware and operating system.
【学位授予单位】：浙江工业大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：U463.6

【参考文献】