单目变焦摄像机自运动的标定测量法

为了在单目摄像机变焦情况下测量其自运动参数,提出一种单目变焦摄像机自运动的参数标定测量法。在飞行平台着陆过程中,固连其上的单目俯视摄像机对包含已知世界坐标的特征点的静态着陆平面进行连续拍摄,该方法利用单帧图像可解算得到摄像机拍摄当时的等效焦距及其相对于着陆平面的6自由度位置,结合多帧信息即可对摄像机自运动的运动速度进行估计,进而转换为飞行平台的着陆运动参数。实验结果证明该方法可行有效。     

单目视觉定位方法研究综述

根据单目视觉定位所用图像帧数不同把定位方法分为基于单帧图像的定位和基于双帧或多帧图像的定位两类。单帧图像定位常利用已知的点特征、直线特征或曲线特征与其在图像上的投影关系进行定位,其中用点特征和直线特征的定位方法简单有效而应用较多;基于双帧或多帧图像的定位方法因其操作复杂、精度不高而研究的还较少。通过对各方法的介绍和评述,为单目视觉定位问题的研究提供参考。     

目标位姿测量中的三维视觉方法

要测量出一组特征点分别在两个空间坐标系下的坐标，就可以求解两个空间目标间的位姿关系，实现上述目标位姿测量方法的前提条件是要保证该组特征点在不同坐标系下，其位置关系相同，但计算误差的存在却破坏了这种固定的位置关系，为此，提出了两种基于模型的三维视觉方法－基于模型的单目视觉和基于模型的双目视觉，前者从视觉计算的物理意义入手，通过简单的约束迭代求解实现模型约束，后者则将简单的约束最小二乘法和基于模型的单目视觉方法融合在一起来实现模型约束，引入模型约束后，单目视觉方法可以达到很高的测量精度，而基于模型的双目视觉较传统的无模型立体视觉方法位移精度提高有限，但姿态精度提高很多。     

基于单目视觉的电动公交车自动更换电池系统

目前电动公交车所用电池沉重而体积大,人工更换时存在挪动调整费力、精确定位困难等问题。针对这些问题,提出一种基于单目视觉的电动公交车自动更换电池系统。该系统使用单目摄像机获取图像,对蓄电池箱的空间位姿进行测量,再通过两级导轨和旋转机构将蓄电池调整到指定位置。位姿测量时采用基于点特征的单目视觉测量算法,提取蓄电池箱的四个角点作为特征点,根据特征点在物体坐标系下以及在CCD成像面上的坐标值,在空间投影模型的基础上求解出物体坐标系相对于摄像机坐标系的变换矩阵,从而求得蓄电池箱的空间位姿。实验证明该方法具有较高的可靠性,并且满足实际问题的精度要求。     

一种基于点线不变量的视觉SLAM算法

点线特征结合的视觉SLAM（同步定位与地图构建）算法中,线特征匹配准确度差会引入新的误差,点线特征误差的累积加剧了数据关联失败情况的发生．针对这一问题,本文设计了一种基于点线不变量的线特征匹配方法,该点线不变量对线段与相邻2个特征点的局部几何关系进行编码,直接在现有特征点的基础上完成线匹配,可有效提高线段匹配的速度和准确度;此外,在点线特征的融合过程引入加权思想,根据场景特征丰富程度,在构造误差函数时对点线特征的权重进行合理分配．在TUM室内数据集和KITTI道路数据集上的实验表明,与现有的点线SLAM系统相比,本文提出的点线SLAM系统有效地提高了视觉SLAM中线特征匹配的准确度,提高了特征匹配环节的运行效率,使线特征在SLAM过程中发挥了积极有效的作用,提高了系统数据关联的稳定性．     

基于点线特征融合的低纹理单目视觉同时定位与地图构建算法

当图像因相机快速运动造成模糊或者处在低纹理场景时,仅使用点特征的同步定位与地图构建（SLAM）算法难以跟踪提取足够多的特征点,导致定位精度和匹配鲁棒性较差。而如果造成误匹配,甚至系统都无法工作。针对上述问题,提出了一种基于点线特征融合的低纹理单目SLAM算法。首先,加入了线特征来加强系统稳定性,并解决了点特征算法在低纹理场景中提取不足的问题;然后,对点、线特征提取数量的选择引入了加权的思想,根据场景的丰富程度,对点线特征的权重进行了合理分配。所提算法是在低纹理场景下运行的,因而设置以线特征为主、点特征为辅。在TUM室内数据集上的实验结果表明,与现有的点线特征算法相比,所提算法有效地提高了线特征的匹配精度,使得轨迹误差减小了大约9个百分点,也使得特征提取时间减少了30个百分点,使加入的线特征在低纹理场景中发挥出积极有效的作用,提高了数据整体的准确度和可信度。     

基于SIFT特征提取的单目视觉里程计在导航系统中的实现

设计一个基于SIFT(Scale Inviriant Feature Transform)特征提取的单目视觉里程计用于导航系统中,有效克服传统里程计车轮打滑等情况下的偏差问题,并提供传统里程计无法测量的航向信息.实现过程采用单个像机,利用SIFT特征点尺度及旋转不变的性质精确匹配图像序列中的投影点,并根据坐标变换来测定车辆的移动距离和旋转角度. 实验中安装要求和实现都相对简单,其结果表明可达到较高的测量精度.     

基于机器视觉的机器人测距研究

机器人测距技术是机器人导航、路径规划以及协作编队关键技术之一。基于透视投影几何原理,提出了单目视觉测距模型,实现物体测距。单目视觉测距模型实现了在摄像机与目标平面平行或者有夹角情况下,对目标物体距离的测量。实验结果表明,所提出的测距模型和和所用的算法,不仅避免了复杂的摄像机标定,而且测量速度快,能够实现对目标物体空间距离的精确测量。     

物体位姿单目视觉传感测量系统

物体位置与姿态的测量在航空航天、汽车车轮定位等领域都有着重要的应用价值.基于单目红外图像传感测量系统,通过对靶标图像特征的有效提取以及数据最优化方法,实现物体位姿测量.设计了专用于物体姿态测量的平面靶标,并介绍了单目传感器测量基本原理、测量模型及数据处理方法.系统具有结构简单、速度快,稳定性好等特点.实验结果表明,转角测量的重复性误差小于8″;平移测量的重复性误差小于0.02 mm.     

基于深度学习的多旋翼无人机单目视觉目标定位追踪方法

针对无人机对目标的识别定位与跟踪,本文提出了一种基于深度学习的多旋翼无人机单目视觉目标识别跟踪方法,解决了传统的基于双目摄像机成本过高以及在复杂环境下识别准确率较低的问题。该方法基于深度学习卷积神经网络的目标检测算法,使用该算法对目标进行模型训练,将训练好的模型加载到搭载ROS的机载电脑。机载电脑外接单目摄像机,单目摄像头检测目标后,自动检测出目标在图像中的位置,通过采用一种基于坐标求差的优化算法进行目标位置准确获取,然后将目标位置信息转化为控制无人机飞行的期望速度和高度发送给飞控板,飞控板接收到机载电脑发送的跟踪指令,实现对目标物体的跟踪。试验结果验证了该方法可以很好的进行目标识别并实现目标追踪     

Planning Stable and Efficient Paths for Reconfigurable Robots On Uneven Terrain

This paper proposes a novel path planning method for improving the feasibility of a forced landing. When an aircraft completely loses its thrust, the only measure it can take is to make a forced landing at an adjacent airport as soon as possible. In such a situation, the flight path to the landing point must be safe and viable. This paper details a method which enables safer and easier landing by transferring the benefits of excess altitude to the final approach length. Moreover, by planning the descent angle of final approach to be in the middle of a non-spoiler and a full-spoiler glide angle, this method enables a change in descent angle to correct any tracking errors, without using thrust. To verify the effectiveness of the proposed method, six degrees-of-freedom nonlinear simulations were performed and the results are compared with comparable methods. From the simulation results, it was confirmed that the proposed method could plan a safe path in a sufficiently short time and the aircraft could reach the landing point safely.     

基于LSD检测的无人机视觉着陆定位算法

针对复杂电磁干扰和拒止环境下固定翼无人机自主着陆的应用场景,提出了一种基于LSD的无人机视觉着陆定位算法,通过检测跑道的左右边线以及起始线对无人机进行定位;根据机场跑道的形态学特征,构建机场跑道数据模型,并对实验所用相机进行标定,采用灰度化和高斯滤波对采集到的图像进行预处理,采用LSD直线检测算法以提取跑道的直线特征,设计几何滤波策略从直线特征中提取出跑道的三条边线,采用蒙版技术以提高检测算法的抗干扰能力;根据相机成像原理推导出基于线检测的PNP定位算法,通过检测得到的机场跑道线在像素坐标系下的位置信息求出无人机相对于跑道的三维位置;分别在视景仿真环境和真实机场环境进行检测和定位解算,结果满足无人机着陆定位实时性和准确性的要求,从而验证了视觉着陆定位算法的有效性。     

求解机场终端区飞机着陆调度问题的遗传算法

空中管制员需为到达的飞机安排跑道并计算着陆时间,以飞机空中延误最小为出发点研究了多跑道的飞机着陆调度问题,约束条件为每架飞机的着陆时间应落在规定的时间窗内及相邻两架飞机应满足最小时间间隔。针对该问题设计了一种遗传算法对问题进行求解,其中染色体由飞机排序链表和跑道链表组成,相应的交叉和变异算子也做了改进设计。仿真实验用数据库OR-Library中的实例验证了该算法的有效性。     

Aircraft landing problems with aircraft classes

This article focuses on the aircraft landing problem that is to assign landing times to aircraft approaching the airport under consideration. Each aircraft’s landing time must be in a time interval encompassing a target landing time. If the actual landing time deviates from the target landing time additional costs occur which depend on the amount of earliness and lateness, respectively. The objective is to minimize overall cost. We consider the set of aircraft being partitioned into aircraft classes such that two aircraft of the same class are equal with respect to wake turbulence. We develop algorithms to solve the corresponding problem. Analyzing the worst case run-time behavior, we show that our algorithms run in polynomial time for fairly general cases of the problem. Moreover, we present integer programming models. We show by means of a computational study how optimality properties can be used to increase efficiency of standard solvers.     

An efficient hybrid particle swarm optimization algorithm in a rolling horizon framework for the aircraft landing problem

This paper proposes a hybrid particle swarm optimization algorithm in a rolling horizon framework to solve the aircraft landing problem (ALP). ALP is an important optimization problem in air traffic control and is well known as NP-hard. The problem consists of allocating the arriving aircrafts to runways at an airport and assigning a landing time to each aircraft. Each aircraft has an optimum target landing time determined based on its most fuel-efficient airspeed and a deviation from it incurs a penalty which is proportional to the amount of deviation. The landing time of each aircraft is constrained within a specified time window and must satisfy minimum separation time requirement with its preceding aircrafts. The objective is to minimize the total penalty cost due to deviation of landing times of aircrafts from the respective target landing times. The performance of the proposed algorithm is evaluated on a set of benchmark instances involving upto 500 aircrafts and 5 runways. Computational results reveal that the proposed algorithm is effective in solving the problem in short computational time.     

Post disaster adaptation management in airport: A coordination of runway and hangar resources for relief cargo transports

Air traffic congestions for processing relief cargos under post-disaster relief scenarios are common, due to high transport demands within a short time. To enhance the resilience of relief operations at airport, an optimization problem of relief air cargo transportations involving aircraft sequencing and loading/unloading within a designated hangar is studied in this paper. The objective is minimizing the tardiness in fulfilling inbound and outbound relief cargos. A mixed-integer linear programming model is formulated, which incorporates aircraft sequencing and hangar parking planning. To resolve the practical problem efficiently, we propose a two-stage optimization approach, which reduces complexity in solving the original model by coordinating the decisions of aircraft landing and take-off schedule and cargo hangar parking arrangement through iterations. The efficiency of the proposed method is examined through the computational results. High-quality solutions can be obtained by the two-stage optimization method within a reasonable time for practical implementation, which enhances the responsiveness and resource utilization of airport operations management under disaster relief situations.     

一种无人机自主着陆视觉跟踪方法

视觉导航是利用图像信息进行飞机定位的,在无人机着陆过程中,为使机载传感器能始终追踪到机场跑道,提出了一种基于摄像机姿态与变焦控制的视觉跟踪方法;该方法通过调节摄像机姿态来追踪目标特征点,使目标特征点尽量位于成像平面的中心,然后根据无人机与机场跑道的距离,适时的调节焦距,以保证图像特征点完全位于图像平面内;实验结果表明,该法能很好地控制摄像机姿态角,使目标特征点位于图像中心附近,达到很好的跟踪效果;且该法操作简单,不需要增加图像处理的难度,实时响应速度快,可以满足工程需要。     

基于某型雷达的着陆航迹纠偏监控研究

为了减少飞机着陆阶段的飞行事故,基于某型精密进场雷达设计了飞机着陆航迹纠偏监控设备;通过研究某型雷达的回波数据格式和传输要求,在Windows环境下,采用Microsoft Visual Studio 10.0开发工具,用C#语言开发了着陆航迹纠偏监控软件;该软件能实时、自动监控加入着陆航线的多批次飞机的着陆航迹数据,对超出安全着陆要求的飞机按不同着陆等级设置三类告警门限,在超限事件引发严重事故之前,实时提醒地面人员指挥引导飞机切入安全合适的下滑道进行着陆,解决了某型雷达完全依赖人工监视飞机着陆状态的缺陷;同时,软件可对雷达数据包进行实时采集和误差分析,以作为日常飞行训练质量的辅助分析手段和飞行事故的调查依据。     

Reachability Analysis of Landing Sites for Forced Landing of a UAS

This paper details a method to ascertain the reachability of known emergency landing sites for any fixed wing aircraft in a forced landing situation. With a knowledge of the aircraft’s state and parameters, as well as a known wind profile, the area of maximum glide range can be calculated using aircraft equations of motion for gliding flight. A landing descent circuit technique used by human pilots carrying out forced landings called high key low key is employed to account for the extra glide distance required for an approach and landing. By combining maximum glide range analysis with the descent circuit, all the reachable landing sites can be determined. X-Plane flight simulator is used to demonstrate and validate the techniques presented.     

积雪对仪表着陆系统下滑道的影响分析

机场积雪会造成仪表着陆系统信号畸变,影响着陆安全,在分析仪表着陆系统下滑道形成原理的基础上,根据几何光学理论建立机场积雪覆盖模型,并依据经验问题的抽象模型,通过仿真分析两类典型的地面积雪覆盖情况下仪表着陆系统下滑道性能的变化。