【作者】 张佳旺；

【导师】 王云泽； 崔莹；

【作者基本信息】 石家庄铁道大学 ， 交通运输工程（专业学位）， 2022， 硕士

【摘要】 近年来,我国的汽车拥有量显著增加,汽车行业飞速发展的同时也带来一系列的问题,例如交通拥堵、交通事故等问题。无人驾驶技术的发展为解决这些问提供了一些新的思路。无人驾驶技术主要包括获取环境信息、路径规划以及轨迹跟踪三个部分,其中路径规划和轨迹跟踪是无人驾驶技术的核心问题。本文针对无人驾驶的局部路径规划进行研究并验证其可行性,构建基于改进势场法的局部路径轨迹,并采用模型预测控制对轨迹进行跟踪仿真验证。具体研究内容如下:首先,介绍了局部路径的相关算法,分别构建了车辆的运动学模型和车辆的动力学模型,并对车辆的运动学模型以及动力学模型进行了详细阐述与公式推理,为路径规划和轨迹跟踪提供了理论基础。其次,对传统势场法理论进行研究分析,详细阐述了传统势场法的优缺点,并对传统势场法做如下改进:根据行驶车辆自身的特性以及道路边界的约束,搭建满足车辆安全行驶的道路势场函数,建立边界斥力。研究障碍物车辆形状特性,构建了更加符合真实车辆的障碍物边界斥力。针对以往的研究存在道路曲率过小无法满足车辆正常行驶问题,引入了B样条曲线对曲率进行约束,以满足车辆的正常行驶。再次,对模型预测进行详细阐述,建立线性化误差模型并对目标函数以及约束条件进行优化。搭建基于车辆运动学模型和车辆动力学模型的轨迹跟踪控制器,并且在Matlab中进行仿真验证,对不同速度下的控制器进行跟踪效果检验。最后,搭建Carsim/Simulink仿真平台,在仿真平台上对设计的非线性模型预测控制器进行建模和仿真,并验证改进势场法在路径规划中的可行性。结果表明,本文设计的改进势场法所规划的局部路径满足跟踪车辆的正常行驶。更多还原

【Abstract】 In recent years,China’s car ownership has increased significantly,the rapid development of the automobile industry has also brought a series of problems,such as traffic congestion,traffic accidents and other problems.The development of unmanned driving technology provides some new ideas to solve these questions.The unmanned driving technology mainly includes obtaining environmental information,path planning and trajectory tracking,among which path planning and trajectory tracking are the core issues of the unmanned driving technology.In this thesis,the feasibility of unmanned local path planning is studied and verified,the local path trajectory based on improved potential field method is constructed,and the trajectory tracking simulation is verified by model predictive control.The specific research contents are as follows:Firstly,the relevant algorithms of local path are introduced,the vehicle kinematics model and vehicle dynamics model are constructed respectively,and the vehicle kinematics model and dynamics model are described in detail and formula reasoning,which provides a theoretical basis for path planning and trajectory tracking.Secondly,the traditional potential field method theory is studied and analyzed,the advantages and disadvantages of the traditional potential field method are described in detail,and the traditional potential field method is improved as follows:according to the characteristics of driving vehicles and the constraints of road boundary,the road potential field function meeting the safe driving of vehicles is built,and the boundary repulsion force is established.The shape characteristics of obstacle vehicle are studied,and the obstacle boundary repulsion force more in line with the real vehicle is constructed.In view of the problem that the road curvature is too small to meet the normal driving of vehicles in previous studies,B-spline curve is introduced to restrict the curvature to meet the normal driving of vehicles.Thirdly,the model prediction is described in detail,the linearized error model is established and the objective function and constraint conditions are optimized.The trajectory tracking controller based on vehicle kinematics model and vehicle dynamics model was built,and the simulation verification was carried out in Matlab,and the tracking effect of the controller at different speeds was tested.Finally,the Carsim/Simulink simulation platform is built,and the nonlinear model predictive controller is modeled and simulated on the simulation platform,and the feasibility of the improved potential field method in path planning is verified.The results show that the local paths planned by the improved potential field method can satisfy the normal driving of the tracking vehicle.更多还原