考虑道路几何特征的车速自适应控制方法研究

发布时间：2018-05-27 00:42

本文选题：智能车辆 + 道路几何特征　；参考：《武汉理工大学》2014年博士论文

【摘要】：近年来，随着我国汽车数量的急剧增多，道路交通安全问题日趋严重。由于车辆具有高速移动性，且道路存在陡坡、急弯、不规则路面等复杂的几何特征，行驶在相关道路上的车辆经常会出现追尾、侧翻等交通事故。智能车辆作为智能交通系统的关键载体，广泛涵盖了以主动安全为导向的先进车辆辅助驾驶与自动驾驶功能，可以提高道路通行能力，提升交通安全性和快捷性，并在此基础上节约能源、减少污染等。在不同道路几何特征条件下的车速自适应控制一直是智能车辆关键技术的研究重点和难点。半实物仿真技术在系统设计与测试的便捷性、复验性、适应性以及安全性方面具有实车道路实验无可比拟的优势。本文依托武汉理工大学智能交通系统研究中心构建的道路交通仿真系统，采用半实物仿真技术，针对不同道路几何特征的车速自适应控制问题，提出了一些新的评价函数，以提高控制律的适应性。首先，引入相似理论，对理论模型与仿真缩微车之间的几何相似、运动相似以及动力相似三个方面进行分析，构建了缩微车模型与智能调控模拟实验平台参数标定方法。在道路几何建模方面，采用回旋线函数建立弯道和坡道几何模型。在车辆运动学建模方面，采用拉格朗日法建立基于3自由度的车辆纵横向耦合模型，包括纵向运动、横向运动与横摆运动的耦合。基于道路交通仿真平台和车辆硬件在环系统，建立了车速自适应控制方法实验环境。接着，分别引入坡道和弯道几何线形特征，提出考虑道路几何特征的评价函数，设计了车速自适应控制方法。在坡道条件下，，推导了坡度角及其变化率的计算公式。根据系统耗散性所反映的能量损耗特性，将车速自适应控制转化成以能量存储函数为优化目标的H∞控制问题。建立关于车速调控的γ耗散性能准则，采用HJI方法(Hamilton-Jacobi-Issacs)，将控制律的设计转化为构造包含坡道几何特征激励费用和车速、纵向加速度变化补偿费用的能量存储函数。采用Backstepping方法，沿着车速控制系统的积分器链信号传递的正向，通过逐步逼近γ耗散不等式的方式，设计车速自适应控制律。在MATLAB环境下开展坡道自主驾驶的智能车速调控仿真实验，对整个上、下坡道的牵引力变化分析，结果表明考虑坡道几何特征的车速自适应控制律不仅可以根据上下坡的坡度角变化自动调节车速，且具有车辆行驶安全性高，能量消耗低的优势。在弯道条件下，对弯道曲率及其变化率的计算公式进行推导，建立车速调控γ耗散性能准则，将横向偏差自调节转化成以偏差平方和最小的最优控制问题。构造包含弯道几何特征激励费用和车速、横向加速度变化补偿费用的评价函数，采用LMI方法将控制律设计转换为满足横向偏差控制系统Hurwitz稳定的正定矩阵的求解。实现弯道自主驾驶的纵横向最优调控MATLAB仿真，考虑弯道几何特征的纵横向最优控制律不仅可以根据弯道曲率变化自动调节车速和横向偏差，并且整个弯道过程中的前轮转角的范围较小，能量消耗较低，更能增强车辆弯道行驶的安全性。最后，开展不同道路条件下的缩微车智能调控模拟实验，包括坡道的智能车速调控和弯道纵横向最优调控。验证了车速自适应控制与横向偏差自调节方法在道路交通仿真平台上的的可行性和有效性。
[Abstract]:In recent years, with the rapid increase of the number of cars in our country, the problem of road traffic safety is becoming more and more serious. Because the vehicle has high-speed mobility, and the road has the complicated geometric features such as steep slope, sharp bend, irregular pavement and so on, vehicles running on the related roads often appear tailing, rollover and other traffic accidents. Intelligent vehicles are used as intelligent traffic. The key carrier of the system covers an active safety oriented advanced vehicle driving and autopilot function, which can improve road traffic capacity, improve traffic safety and shortcut, and save energy and reduce pollution on this basis. Adaptive speed control of vehicle speed under different road geometric characteristics has always been intelligent. The hardware in the loop simulation technology has an unparalleled advantage in the convenience of the system design and testing, the retesting, the adaptability and the safety of the vehicle road experiment. This paper relies on the simulation system built by the research center of the intelligent transportation system of Wuhan University of Technology, which adopts the semi physical imitation. In order to improve the adaptability of the control law, some new evaluation functions are proposed for vehicle speed adaptive control of different road geometry characteristics.
First, the similarity theory is introduced, and the geometric similarity between the theoretical model and the simulated micro vehicle, motion similarity and dynamic similarity are analyzed in three aspects. The calibration method of the parameters of the model and the intelligent control simulation experiment platform is constructed. In the aspect of road geometric modeling, the curve and the geometric model of the ramp are used to establish the curve and the geometric model of the ramp. In the vehicle kinematics modeling, the vehicle longitudinal and transverse coupling model based on 3 degrees of freedom is established by Lagrange method, which includes the longitudinal motion, the coupling of transverse motion and the yaw motion. Based on the road traffic simulation platform and the vehicle hardware in the ring system, the experimental environment of the speed adaptive control method is established.
Then, the geometric linear features of the ramp and bend are introduced, and the evaluation function of the geometric characteristics of the road is put forward, and the adaptive control method of the speed is designed. The calculation formula of the slope angle and the rate of change is derived under the slope condition. The adaptive control of vehicle speed is converted into energy according to the energy loss characteristics reflected by the dissipative system of the system. The storage function is the H infinity control problem for the optimization target. A gamma dissipation criterion for speed regulation is established. The HJI method (Hamilton-Jacobi-Issacs) is used to transform the design of the control law into a energy storage function for constructing the compensation cost of the excitation cost and speed of the ramp and the variation of the longitudinal acceleration. The Backstepping method is used. Along the forward speed control system of the integrator chain signal, the speed adaptive control law is designed by gradually approaching the gamma dissipation inequality. In the MATLAB environment, the intelligent speed control simulation experiment of the autonomous driving on the slope is carried out. The change of the traction force of the lower ramp is analyzed. The results show that the geometric characteristics of the ramp are taken into consideration. The speed adaptive control law can not only automatically adjust the speed according to the gradient angle of the downhill slope, but also has the advantages of high driving safety and low energy consumption.
The calculation formula of curve curvature and its rate of change is deduced under the condition of bend, and the criterion of speed regulation is set up. The optimal control problem is transformed into the optimal control problem with the minimum square sum of deviation. The LMI method is used to convert the control law design into the solution of the positive definite matrix which satisfies the Hurwitz stability of the lateral deviation control system. To realize the optimal control of the longitudinal and transverse direction of the autonomous driving of the bend, the optimal control law of the longitudinal and transverse direction of the curve is not only to automatically adjust the speed and lateral deviation according to the curvature of the bend, but also to adjust the optimal control law of the MATLAB. During the curve, the angle of the front wheel is smaller and the energy consumption is lower, which can enhance the safety of the vehicle running in the curve.
Finally, the intelligent control simulation experiment of the micro vehicle under different road conditions is carried out, including the intelligent speed control of the ramp and the optimal control of the longitudinal and transverse direction of the bend. The feasibility and effectiveness of the adaptive speed control and the lateral deviation self adjustment method on the road traffic simulation platform are verified.
【学位授予单位】：武汉理工大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：U495;U463.6

【参考文献】