小车位环境下的自动泊车路径规划与跟踪控制

针对小车位泊车路径设计不连续导致位姿调整不平滑的问题,提出一种回旋曲线、圆弧、直线组合的路径规划方法,并基于非时间参考量的跟踪偏差,设计滑模控制器。根据泊车要求和限制条件,规划出泊入路径与调整路径：泊入路径引导车辆以极限位姿进入车位;调整路径通过平滑的路径组合实现车位内的位姿调整。选用指数趋近率设计滑模控制器,利用连续函数代替趋近率中的符号函数,降低抖振影响。仿真结果表明,车辆能够连续、无突变地完成泊车操作,跟踪误差较小,抖振抑制效果较好。     

自动泊车运动规划方法综述

自动泊车是自动驾驶车辆在低速行驶场景中的核心技术之一,已被广泛应用于各类车型中。运动规划确保所规划出的轨迹能指引车辆从起始位姿无碰撞、符合运动学规律地行驶至目标位姿,是体现自动泊车智能水准的关键技术模块。随着当前自动驾驶技术从初步的技术原型研究逐渐向产业化应用转型,低速自动驾驶领域中的自动泊车技术成为了这一转型过程中的关键突破点。文章回顾了现有的应用于自动泊车领域的运动规划方法,将其归纳为几何方法、采样方法、搜索方法、数值优化方法以及机器学习方法等类别,并对各类方法的发展历程与技术特点进行了回顾;探讨了自动泊车运动规划方法的未来研究方向,具体包括构建泛化性更强的模型、提升规划方法的稳定性、关注在线实时高质量重规划的能力与防御性驾驶的能力,以及考虑与其他交通参与者的交互等。     

一种自动泊车的仿人智能控制方法

为简化平行泊车,通过总结模拟熟练驾驶员的泊车经验,提出一种自动泊车的仿人智能控制方法。该方法以后轮驱动、前轮转向的四轮汽车为对象,建立车辆运动学模型;通过分析熟练驾驶员泊车流程,将泊车过程分为4个阶段;在相切圆弧加公切线的规划路径上,选取泊车阶段转换时车辆姿态调整的关键点作为跟踪目标;根据泊车过程车辆的位姿信息,提取12种泊车的特征状态,作为描述车辆泊车动态行为的特征模型;根据熟练驾驶员的泊车策略,构建控制模态集。泊车开始后,依据特征模型先验知识和当前车辆位姿与泊车目标的偏差,对当前车辆特征状态进行模式识别,由辨识出的特征状态驱动相应控制模态,控制汽车按规划路径泊车入位。建立了车辆运动学和仿人智能控制器Simulink模型,并进行了仿真实验。仿真结果表明,该方法能有效控制车辆泊入车位。     

一种应用于自动泊车系统的车位检测方法

自动泊车系统是为了解决倒车入位的复杂性问题而研发的一项车辆智能化的新技术。车位检测是实现自动泊车系统的一个重要问题。本文首先对车位检测问题进行总体描述;然后提出一种基于超声波传感器和增量式编码器的车位检测方法,并刘此方法的测量误差进行分析;接着介绍实验平台的组成;最后给出此方法在买验平台上的测试结果,并通过对结果的分析,说明此方法是可行的。     

基于平行泊车路径规划的智能泊车系统设计

针对智能泊车系统对泊车准确度及计费实时性要求,提出一种基于距离和避障信息融合的自动泊车路径规划方法和泊车计费方法,解决当前泊车与计费不能有效融合,不利于进行智能管控的现实问题。系统采用STM32单片机作为系统主控芯片,电机驱动模块RZ7889D驱动直流电机,转向舵机采用MG996,超声波测距传感器HC-SR04用于测量车身与障碍物物理距离并结合LM393实现红外避障,实现智能泊车。系统采用LabVIEW设计智能泊车系统上位机面板,实时显示泊车车位、数量、时长和费用,并具有历史数据查询和车辆数据波形显示功能。经仿真测试和系统微缩物理模型测试,该系统能够实现平行泊车路径规划,可实现平稳、顺滑的良好泊车,平均泊车时长为41s,能够根据停车时长进行计费,满足简单、可靠且准确的智能泊车计费系统设计要求。     

基于STM32和CoppeliaSim Edu的自动泊车系统设计与仿真

本项目设计并制作一个自动泊车系统智能小车，以STM32单片机为控制核心，并由单片机主控模块、电机驱动模块、超声波模块、红外避障模块等主要模块构成；通过LUA语言编程，利用Coppelia Sim Edu进行模拟仿真，能够完成自动循迹、速度控制以及传感器测距等多项功能。项目采用可视图法对小车路径进行规划，将小车、库位点和障碍物的各顶点进行组合连接，通过改善路径规划，提高小车自动泊车效率。通过优化自动倒车入库和侧方位倒车入库智能小车算法，根据停车的不同场景，对环境信息进行不同处理，综合确定停车空间，并作出不同的控制电信号，进而规划停车入库的路径，完成自动泊车功能。     

自动代客泊车背景下的共享停车供需匹配模型及对应禁忌搜索算法

为了减少自动代客泊车车辆在停车场或指定停车区域内的车辆移位次数和距离,从而降低相关的成本和潜在事故风险,在满足共享停车需求的条件下构建了相应的车辆和泊位匹配优化模型.考虑到无人驾驶车辆停车中可自由移位的特征,将共享停车的需求和供给在时间上加以细分,与决策变量和可行解对应定义了匹配、匹配条和匹配图的概念;通过概念转换将求解匹配图的有效邻居转换为经典指派问题,并利用匈牙利算法加以求解;针对匹配模型的NP-hard特征,设计了对应的禁忌搜索算法.数值分析不仅验证了模型的合理性和求解算法的有效性,也证实模型与方法可处理有人驾驶的共享停车匹配问题.结果表明,利用自动代客泊车可以进一步提升共享泊位利用率,增加可停放的共享车辆数.     

基于A*算法的停车场路径引导设计

为提高停车场停车效率和选择最佳停车位,根据停车场的车位布局及车位使用状况,设计了泊车路径规划模型。把空闲车位抽象为二维坐标系的坐标点,分别计算其距离用户位置的曼哈顿距离,选取曼哈顿距离最小值的停车位为最优车位。选用改进的A*算法为停车路径规划算法,为进一步优化算法遍历过程,使之更加适用于停车场车位分布较分散的情况,以停车位对应路段为遍历节点进行路径搜索,并完成路径规划,结果表明:该方法有效地去除了冗余节点,提高了算法遍历速度。     

基于A*优化算法的停车场动态泊车研究

针对目前大型停车场停车效率、车位利用率低等问题,将停车效率问题转化为动态泊车概率问题,结合实际的地下停车场车位模型,以泊车用户的泊车概率、算法搜索效率为主要评价指标,并以较大的概率快速寻找到可用空车位为目的,提出了一种带约束条件的A*优化算法,然后基于VC++6.0环境对停车场实例进行算法的应用仿真。实验结果表明,优化后的A*算法大大增加了泊车用户的泊车概率,A*算法的搜索效率也得到显著提高,此A*算法在停车场中的应用,很大程度上提高了泊车用户的停车效率以及停车位的利用率,减少了泊车用户盲目重复寻找车位所花费的时间,在大型停车场中具有一定的应用研究价值。     

基于改进A*算法的AGV智能泊车算法

随着社会的发展和文明的进步，人类对车辆智能化水平、便捷性和安全性的要求越来越高，针对停车场车位少、停车难的问题，在A*算法的基础上，增加了时间因素，将等待时间加入启发函数，综合路径距离和等待时间两个因素进而规划出入库和出库任务的最佳路径，设计了三维A*智能泊车算法，在结构化环境下，根据给定的停车场地图，预留出最优车位，规划出多辆同时工作的自动导引车（Automated Guided Vehicle，AGV）最优路径，安排AGV运载车辆到指定车位或完成出库过程，尽量减少客户在停车和取车中的等待时间，并使总成本最小，呈现真正"互联网化"的智能停车体验.     

基于二段多项式的窄空间平行泊车路径规划方法

针对窄空间平行泊车路径规划问题, 提出一种基于二段五阶多项式的方法分别计算车辆前进路线与倒车路线, 其中前进路线用于调整车辆位姿, 实现倒车入库路径曲率最优. 根据常规停车位大小与家庭轿车的尺寸建立仿真环境, 结合车辆运动模型, 搜索自由空间内无碰撞路径. 仿真结果表明, 二段五阶多项式法可以实现在有限空间范围内的智能泊车, 且在由两段路径连接得到的路径上方向盘转向角连续变化.     

Kinematics of automatic guided vehicles with an inclined steering column and an offset distance: criterion for existence of inverse kinematic solution

This article presents the kinematic model of an automatic guided vehicle (AGV) with an inclined steering column and an offset distance. The vehicle configuration considered here employs the front wheel as the driving and steering wheel. From the kinematic model, a criterion is derived to test the existence of inverse kinematic solution. The criterion is a function of the geometric parameters of the AGV and the steering angle of the wheel. From the criterion defined, conditions for the existence of inverse kinematic solution for steering column configurations with and without inclination and offset distance are obtained. The criterion also presents the range of steering angles for which the inverse kinematic solution exists. The criterion signifies the presence of a unique solution for the motion parameters of the wheel, i.e., the wheel speed, steering angle, and steering rate for desired values of the motion parameters of the AGV, i.e., the linear and angular velocities of the AGV. This information is needed for the guidance control of AGVs based upon a kinematic control scheme (dead-reckoning control) without using the absolute position referencing system to follow a predefined path at prescribed linear and angular velocities. © 1992 John Wiley & Sons, Inc.     

超声波测距的模糊控制泊车系统模型设计

设计了一种基于超声波测距的自动泊车系统模型。首先对小车运动模型进行分析,结合实际设计垂直、平行、斜行三种泊车模糊控制器,并且在Matlab的Simulink中分别进行仿真验证。采用超声波测距技术、视频采集技术等来获取周围环境数据,按照实际车辆和车库的大小,通过等比例缩放设计小车模型和车库模型,并把验证通过的模糊控制器转换为C代码,嵌入到飞思卡尔K60主控制板中进行自动泊车可行性测试。经测试,当模型车的航向角在-32°~+32°范围内时,实现了模型车在0.92倍标准大小模型车库内的自主泊车。     

Postural support strategies of disabled drivers and the effectiveness of postural support aids

The paper discusses a series of driving trials that were conducted to investigate postural stability of disabled drivers and to assess the effectiveness of a representative sample of support aids. Twenty-three disabled drivers with varying levels of physical disability and seven non disabled drivers participated in the study. The test car was equipped with transducers to measure vehicle velocity and acceleration (longitudinal and lateral), steering wheel movement and torque, and the bracing forces exerted by the driver on the steering wheel. Video cameras were installed to record postural support strategies and displacement of the driver and to record deviation of the car from the specified path. Subjective data regarding driver attitudes and acceptance were also collected through the administration of questionnaires. Findings from the study showed that support aids significantly improved driving performance and reduced physical exertion to maintain an upright driving position for disabled drivers. However, ergonomics design aspects regarding the ease of use and acceptance of the support aids by the end users were identified as obstacles to their sustained use in everyday driving.     

辅助实现机器人驾驶车辆的研究与仿真

针对机器人驾驶汽车在未知复杂环境下行驶的问题,提出一种基于图像的辅助实现机器人驾驶车辆的方法.根据机器人驾驶过程中所拍摄的路面图像,在图像中取一点作为车辆行驶的目标位置点;根据此点的位置,计算出车辆行驶时的转弯半径;对方向盘转角与转弯半径之间的关系进行曲线拟合,计算得出方向盘的转角;根据转角信息在图像上画出可供参考的车辆可行驶辅助线.本文算法利用matlab进行了仿真研究,实验结果表明,采用这种方法获取辅助线能够较为准确的作为车辆行驶的参考,为辅助机器人驾驶车辆行驶带来了很大的便利.     

The prediction of car driver size and position to enhance safety in crashes

The positions which car drivers adopt when driving will depend on their anthropometric characteristics, the range and type of adjustment available from the vehicle package and their preferred driving posture. The design and testing of systems to protect occupants in car crashes assumes that the size and position of the driver is ‘normal’ or ‘average’, although there is some accommodation for adjustability. If, however, the occupant protection system had information on the driver's chosen seat position, on whether the driver was particularly large or small and on whether the driver was sitting close to or further from the steering wheel, in a crash the system could tailor its performance and enhance the protection offered. This study investigated whether it was possible to predict the physical characteristics of the driver and the driver's position in relation to the steering wheel, from data that could be collected by sensors in the seat and seat mounting. In order to do this, anthropometric characteristics of drivers and their usual seated position in their own vehicle were measured and analyses were undertaken to identify whether there were any relationships between the driver-related and the vehicle-related measures. The results showed that it was possible to predict drivers' head and chest positions relative to injury-producing features of the vehicle such as the steering wheel (and hence the airbag) and to predict some physical dimensions of drivers.     

适用于车头泊入的路径规划和跟踪控制新方法

自动泊车作为智能汽车自动驾驶技术的一个关键组成部分,已成为业界关注和研究的热点。现有的自动泊车系统都是基于倒车实现,而大多数新型电动汽车充电口安装在车头,因此设计了车头泊入的泊车新方法。该方法首先对车辆进行运动学建模,然后根据汽车与车库的相对位置基于几何推导进行路径规划,最后利用PID对泊车过程进行跟踪控制。采用PID路径跟踪控制算法与现有泊车跟踪控制算法相比较,有效简化了算法结构,克服了“计算爆炸”的问题,具有较强的鲁棒性,符合现有新型电动汽车车头泊入的需求。使用Matlab和CarSim联合仿真实车参数模型车的运动学模型,完善了传统泊车过程模型的运动仿真分析,仿真结果验证了所设计的跟踪控制方法的有效性。     

Autonomous car driving by a humanoid robot

Enabling a humanoid robot to drive a car requires the development of a set of basic primitive actions. These include walking to the vehicle, manually controlling its commands (e.g., ignition, gas pedal, and steering) and moving with the whole body to ingress/egress the car. We present a sensor‐based reactive framework for realizing the central part of the complete task, consisting of driving the car along unknown roads. The proposed framework provides three driving strategies by which a human supervisor can teleoperate the car or give the robot full or partial control of the car. A visual servoing scheme uses features of the road image to provide the reference angle for the steering wheel to drive the car at the center of the road. Simultaneously, a Kalman filter merges optical flow and accelerometer measurements to estimate the car linear velocity and correspondingly compute the gas pedal command for driving at a desired speed. The steering wheel and gas pedal reference are sent to the robot control to achieve the driving task with the humanoid. We present results from a driving experience with a real car and the humanoid robot HRP‐2Kai. Part of the framework has been used to perform the driving task at the DARPA Robotics Challenge.     

一种用于AFS的半实物仿真平台

为了给自适应前照灯系统（Adaptive Front-Lighting System,AFS）提供一个控制算法验证平台,开发了一套基于汽车驾驶模拟器的AFS半实物硬件仿真平台。在该仿真平台中,PC机采集驾驶模拟器的档位、油门、离合、刹车、方向盘等驾驶信息,由车辆动力学模型模拟实际车辆行为,采用AFS控制模型计算车灯转角控制量,并发送至电机驱动模块进而控制车灯转动,同时将车灯转角信息反馈至PC机。实验表明,该系统能实时记录和显示AFS工作过程中的多种参数,便于进行实验观察和数据分析,从而为AFS系统的控制算法提供一个检验和修正的平台。