公开(公告)号：US09014903B1

公开(公告)日：2015-04-21

申请号：US13477871

申请日：2012-05-22

Applicant: Jiajun Zhu ,  Dmitri A. Dolgov ,  Nathaniel Fairfield

Inventor： Jiajun Zhu ,  Dmitri A. Dolgov ,  Nathaniel Fairfield

IPC: G05D1/00 ,  G05D3/00 ,  G01C22/00 ,  G05D1/02 ,  G06G7/78 ,  G08G1/16 ,  G01S13/88 ,  G06K9/00 ,  G01C21/34

CPC classification number: G01C21/3446 ,  G01S13/865 ,  G01S13/867 ,  G01S13/931 ,  G01S17/875 ,  G01S17/89 ,  G01S17/936 ,  G01S2013/9321 ,  G01S2013/9342 ,  G01S2013/9346 ,  G05D1/0238 ,  G05D2201/0213

Abstract: An autonomous vehicle configured to determine the heading of an object-of-interest based on a point cloud. An example computer-implemented method involves: (a) receiving spatial-point data indicating a set of spatial points, each spatial point representing a point in three dimensions, where the set of spatial points corresponds to an object-of-interest; (b) determining, for each spatial point, an associated projected point, each projected point representing a point in two dimensions; (c) determining a set of line segments based on the determined projected points, where each respective line segment connects at least two determined projected points; (d) determining an orientation of at least one determined line segment from the set of line segments; and (e) determining a heading of the object-of-interest based on at least the determined orientation.

Abstract translation: 一种被配置为基于点云确定感兴趣对象的标题的自主车辆。 计算机实现的示例方法包括：（a）接收指示一组空间点的空间点数据，每个空间点表示三维点，其中该组空间点对应于感兴趣的对象; （b）为每个空间点确定相关联的投影点，每个投影点表示二维中的点; （c）基于所确定的投影点确定一组线段，其中每个相应线段连接至少两个确定的投影点; （d）从所述一组线段确定至少一个确定的线段的取向; 和（e）至少基于确定的方向来确定兴趣对象的标题。

公开(公告)号：US08874372B1

公开(公告)日：2014-10-28

申请号：US13440555

申请日：2012-04-05

Applicant: Jiajun Zhu ,  Michael Steven Montemerlo ,  Christopher Paul Urmson ,  Andrew Chatham

Inventor： Jiajun Zhu ,  Michael Steven Montemerlo ,  Christopher Paul Urmson ,  Andrew Chatham

IPC: G06K9/00

CPC classification number: G06K9/00805 ,  G01C21/26 ,  G08G1/09626 ,  G08G1/096725 ,  G08G1/096741 ,  G08G1/096775

Abstract: Aspects of the disclosure relate generally to safe and effective use of autonomous vehicles. More specifically, objects detected in a vehicle's surroundings may be detected by the vehicle's various sensors and identified based on their relative location in a roadgraph. The roadgraph may include a graph network of information such as roads, lanes, intersections, and the connections between these features. The roadgraph may also include the boundaries of areas, including for example, crosswalks or bicycle lanes. In one example, an object detected in a location corresponding to a crosswalk area of the roadgraph may be identified as a person. In another example, an object detected in a location corresponding to a bicycle area of the roadgraph and identified as a bicycle. By identifying the type of object in this way, an autonomous vehicle may be better prepared to react to or simply avoid the object.

Abstract translation: 本公开的方面一般涉及安全和有效地使用自主车辆。 更具体地，可以由车辆的各种传感器检测在车辆周围检测到的物体，并且基于它们在路标图中的相对位置来识别。 道路图可以包括信息的图形网络，例如道路，车道，交叉路口以及这些特征之间的连接。 道路图也可以包括区域的边界，包括例如人行横道或自行车道。 在一个示例中，在与路灯图的人行横道区域相对应的位置中检测到的对象可以被识别为人。 在另一示例中，在与路灯图的自行车区域相对应的位置中检测到的物体，并被识别为自行车。 通过以这种方式识别对象的类型，自主车辆可以更好地准备反应或者简单地避开物体。

公开(公告)号：US08825265B1

公开(公告)日：2014-09-02

申请号：US13422688

申请日：2012-03-16

Applicant: David I. Ferguson ,  Jiajun Zhu

Inventor： David I. Ferguson ,  Jiajun Zhu

IPC: G05D1/02

CPC classification number: G05D1/0212 ,  B60W40/04 ,  B62D15/025 ,  B62D15/0255 ,  G05D1/0088 ,  G05D1/0214 ,  G05D1/0223 ,  G05D1/0246 ,  G05D1/0287 ,  G08G1/096725

Abstract: A method and apparatus is provided for controlling the operation of an autonomous vehicle. According to one aspect, the autonomous vehicle may track the trajectories of other vehicles on a road. Based on the other vehicle's trajectories, the autonomous vehicle may generate a pool of combined trajectories. Subsequently, the autonomous vehicle may select one of the combined trajectories as a representative trajectory. The representative trajectory may be used to change at least one of the speed or direction of the autonomous vehicle.

Abstract translation: 提供了一种用于控制自主车辆的操作的方法和装置。 根据一个方面，自主车辆可以跟踪道路上的其他车辆的轨迹。 基于其他车辆的轨迹，自主车辆可以产生组合轨迹池。 随后，自主车辆可以选择组合轨迹中的一个作为代表轨迹。 代表性轨迹可以用于改变自主车辆的速度或方向中的至少一个。

公开(公告)号：US08781721B2

公开(公告)日：2014-07-15

申请号：US13489794

申请日：2012-06-06

Applicant: Jonathan Baldwin Dowdall ,  Jiajun Zhu

Inventor： Jonathan Baldwin Dowdall ,  Jiajun Zhu

IPC: G08G1/16 ,  B60R21/00 ,  B60R21/013 ,  B60W30/08 ,  B60W30/095

CPC classification number: G08G1/166 ,  B60R21/00 ,  B60R21/013 ,  B60W30/08 ,  B60W30/095 ,  G06K9/00805

Abstract: A vehicle configured to operate in an autonomous mode may engage in an obstacle evaluation technique that includes employing a sensor system to collect data relating to a plurality of obstacles, identifying from the plurality of obstacles an obstacle pair including a first obstacle and a second obstacle, engaging in an evaluation process by comparing the data collected for the first obstacle to the data collected for the second obstacle, and in response to engaging in the evaluation process, making a determination of whether the first obstacle and the second obstacle are two separate obstacles.

Abstract translation: 构造成以自主模式操作的车辆可以参与障碍物评估技术，其包括采用传感器系统来收集与多个障碍物相关的数据，从多个障碍物识别包括第一障碍物和第二障碍物的障碍物， 通过将针对第一障碍物收集的数据与针对第二障碍物收集的数据进行比较，并且响应于参与评估处理，确定第一障碍物和第二障碍物是否是两个单独的障碍物来进行评估过程。

公开(公告)号：US08630806B1

公开(公告)日：2014-01-14

申请号：US13277580

申请日：2011-10-20

Applicant: Jiajun Zhu

Inventor： Jiajun Zhu

IPC: G01C21/26

CPC classification number: G01C21/26 ,  B60W30/16 ,  B60W2420/42 ,  B60W2520/16 ,  B60W2550/142 ,  B60W2550/308 ,  G01S11/12 ,  G01S17/936 ,  G05D1/0251 ,  G05D2201/0213 ,  G06K9/00805 ,  G08G1/165 ,  G08G1/166 ,  G08G9/02

Abstract: A method is provided for processing an image in which only parts of the image that appear above a point on a horizon line are analyzed to identify an object. In one embodiment, the distance between the object and a vehicle is determined, and at least one of the speed and direction of the vehicle is changed when it is determined that the distance is less than the range of a sensor. The method for processing images is not limited to vehicular applications only and it may be used in all applications where computer vision is used to identify objects in an image.

Abstract translation: 提供了一种处理图像的方法，其中仅分析出现在水平线上的点上方的图像的一部分以识别对象。 在一个实施例中，确定物体与车辆之间的距离，并且当确定距离小于传感器的范围时，车辆的速度和方向中的至少一个被改变。 用于处理图像的方法不仅限于车辆应用，并且可以在计算机视觉用于识别图像中的对象的所有应用中使用。

公开(公告)号：US20130197736A1

公开(公告)日：2013-08-01

申请号：US13361083

申请日：2012-01-30

Applicant: Jiajun Zhu ,  Dmitri A. Dolgov ,  David I. Ferguson

Inventor： Jiajun Zhu ,  Dmitri A. Dolgov ,  David I. Ferguson

IPC: G05D1/00

CPC classification number: G05D1/0088 ,  G05D2201/0213

Abstract: Aspects of the disclosure relate generally to maneuvering autonomous vehicles. Specifically, the vehicle may determine the uncertainty in its perception system and use this uncertainty value to make decisions about how to maneuver the vehicle. For example, the perception system may include sensors, object type models, and object motion models, each associated with uncertainties. The sensors may be associated with uncertainties based on the sensor's range, speed, and /or shape of the sensor field. The object type models may be associated with uncertainties, for example, in whether a perceived object is of one type (such as a small car) or another type (such as a bicycle). The object motion models may also be associated with uncertainties, for example, not all objects will move exactly as they are predicted to move. These uncertainties may be used to maneuver the vehicle.

Abstract translation: 本公开的方面通常涉及操纵自主车辆。 具体来说，车辆可以确定其感知系统的不确定性，并使用该不确定性值来决定如何操纵车辆。 例如，感知系统可以包括传感器，对象类型模型和对象运动模型，每个与不确定性相关联。 传感器可能会根据传感器范围，速度和/或传感器场的形状与不确定性相关联。 对象类型模型可能与不确定性相关联，例如，感知对象是否是一种类型（例如小型轿厢）或另一类型（例如自行车）。 对象运动模型也可能与不确定性相关联，例如，并不是所有的对象将按照预测的移动精确地移动。 这些不确定性可用于操纵车辆。

公开(公告)号：US20120327184A1

公开(公告)日：2012-12-27

申请号：US13605635

申请日：2012-09-06

Applicant: Jiajun ZHU ,  Daniel Filip ,  Luc Vincent

Inventor： Jiajun ZHU ,  Daniel Filip ,  Luc Vincent

IPC: H04N7/00

CPC classification number: G06T19/003 ,  G06F3/04815 ,  G06T3/4038 ,  G06T15/06 ,  G06T17/00 ,  G06T19/00 ,  G06T2200/32

Abstract: The present invention relates to using image content to facilitate navigation in panoramic image data. In an embodiment, a computer-implemented method for navigating in panoramic image data includes: (1) determining an intersection of a ray and a virtual model, wherein the ray extends from a camera viewport of an image and the virtual model comprises a plurality of facade planes; (2) retrieving a panoramic image; (3) orienting the panoramic image to the intersection; and (4) displaying the oriented panoramic image.

Abstract translation: 本发明涉及使用图像内容以促进全景图像数据中的导航。 在一个实施例中，用于在全景图像数据中导航的计算机实现的方法包括：（1）确定射线和虚拟模型的交集，其中所述射线从图像的相机视口延伸，并且所述虚拟模型包括多个 立面飞机 （2）检索全景图像; （3）将全景图像定向到十字路口; 和（4）显示定向全景图像。

公开(公告)号：US20090179895A1

公开(公告)日：2009-07-16

申请号：US12014513

申请日：2008-01-15

Applicant: Jiajun Zhu ,  Daniel Filip ,  Luc Vincent

Inventor： Jiajun Zhu ,  Daniel Filip ,  Luc Vincent

IPC: G06T17/00

CPC classification number: G06T11/60 ,  G06F17/241 ,  G06F17/30241 ,  G06T7/75 ,  G06T19/00 ,  G06T19/20 ,  G06T2207/10028 ,  G06T2207/20101 ,  G06T2210/12 ,  G06T2219/004

Abstract: The present invention relates to annotating images. In an embodiment, the present invention enables users to create annotations corresponding to three-dimensional objects while viewing two-dimensional images. In one embodiment, this is achieved by projecting a selecting object onto a three-dimensional model created from a plurality of two-dimensional images. The selecting object is input by a user while viewing a first image corresponding to a portion of the three-dimensional model. A location corresponding to the projection on the three-dimensional model is determined, and content entered by the user while viewing the first image is associated with the location. The content is stored together with the location information to form an annotation. The annotation can be retrieved and displayed together with other images corresponding to the location.

Abstract translation: 本发明涉及注释图像。 在一个实施例中，本发明使用户能够在观看二维图像的同时创建与三维对象相对应的注释。 在一个实施例中，这是通过将选择对象投影到由多个二维图像创建的三维模型上而实现的。 选择对象由用户输入，同时观看对应于三维模型的一部分的第一图像。 确定与三维模型上的投影对应的位置，并且在查看第一图像时由用户输入的内容与位置相关联。 内容与位置信息一起存储以形成注释。 注释可以与对应于该位置的其他图像一起检索和显示。

公开(公告)号：US09381916B1

公开(公告)日：2016-07-05

申请号：US13366828

申请日：2012-02-06

Applicant: Jiajun Zhu ,  David I. Ferguson

Inventor： Jiajun Zhu ,  David I. Ferguson

IPC: G08G1/16 ,  B60W30/095 ,  G01S13/93 ,  G05D1/00

CPC classification number: G05D1/0088 ,  B60W30/0956 ,  G01S13/862 ,  G01S13/865 ,  G01S13/867 ,  G01S13/931 ,  G01S15/931 ,  G01S17/936 ,  G01S2013/9353 ,  G01S2013/9357 ,  G01S2013/9375 ,  G01S2013/9378 ,  G01S2013/9382 ,  G01S2013/9385 ,  G05D1/0248 ,  G05D1/0257 ,  G05D2201/0212 ,  G08G1/16 ,  G08G1/166

Abstract: Aspects of the invention relate generally to autonomous vehicles. The features described improve the safety, use, driver experience, and performance of these vehicles by performing a behavior analysis on mobile objects in the vicinity of an autonomous vehicle. Specifically, the autonomous vehicle is capable of detecting nearby objects, such as vehicles and pedestrians, and is able to determine how the detected vehicles and pedestrians perceive their surroundings. The autonomous vehicle may then use this information to safely maneuver around all nearby objects.

Abstract translation: 本发明的方面一般涉及自主车辆。 所描述的特征通过对自主车辆附近的移动物体进行行为分析来改善这些车辆的安全性，使用性，驾驶者体验和性能。 具体来说，自主车辆能够检测车辆和行人等附近物体，并能够确定检测到的车辆和行人如何察觉周围环境。 然后，自主车辆可以使用该信息来安全地围绕所有附近的物体操纵。

公开(公告)号：US08611696B1

公开(公告)日：2013-12-17

申请号：US13441588

申请日：2012-04-06

Applicant: Jiajun Zhu ,  Abhijit Ogale

Inventor： Jiajun Zhu ,  Abhijit Ogale

IPC: G06K9/36

CPC classification number: G06T5/005 ,  G06T3/4038

Abstract: Systems and methods for filling panoramic images having valid and invalid pixel regions are provided. An invalid region is identified in an initial panoramic image. Pixel data of invalid pixels in the initial panoramic image are replaced with pixel data of pixels from a valid region in at least one nearby panoramic image to obtain a valid fill region.

Abstract translation: 提供了用于填充具有有效和无效像素区域的全景图像的系统和方法。 在初始全景图像中识别无效区域。 初始全景图像中的无效像素的像素数据被替换为来自至少一个附近全景图像中的有效区域的像素的像素数据，以获得有效的填充区域。