公开(公告)号：US09248569B2

公开(公告)日：2016-02-02

申请号：US14088258

申请日：2013-11-22

Applicant: BRAIN CORPORATION

Inventor： Patryk Laurent ,  Jean-Baptiste Passot ,  Filip Ponulak ,  Eugene Izhikevich

IPC: B25J9/16

CPC classification number: B25J9/163 ,  G05B13/026 ,  G05B2219/40512 ,  G05B2219/42058 ,  G05D1/0088 ,  G05D2201/02 ,  G06N3/049 ,  G06N99/005

Abstract: A robotic device may comprise an adaptive controller configured to learn to predict consequences of robotic device's actions. During training, the controller may receive a copy of the planned and/or executed motor command and sensory information obtained based on the robot's response to the command. The controller may predict sensory outcome based on the command and one or more prior sensory inputs. The predicted sensory outcome may be compared to the actual outcome. Based on a determination that the prediction matches the actual outcome, the training may stop. Upon detecting a discrepancy between the prediction and the actual outcome, the controller may provide a continuation signal configured to indicate that additional training may be utilized. In some classification implementations, the discrepancy signal may be used to indicate occurrence of novel (not yet learned) objects in the sensory input and/or indicate continuation of training to recognize said objects.

Abstract translation: 机器人设备可以包括被配置为学习预测机器人设备的动作的后果的自适应控制器。 在训练期间，控制器可以接收基于机器人对命令的响应获得的计划和/或执行的电动机命令和感觉信息的副本。 控制器可以基于命令和一个或多个现有的感觉输入来预测感觉结果。 预测的感觉结果可能与实际结果进行比较。 根据预测与实际结果的匹配，培训可能会停止。 在检测到预测和实际结果之间的差异时，控制器可以提供被配置为指示可以利用附加训练的连续信号。 在一些分类实现中，差异信号可以用于指示感觉输入中的新颖（尚未学习）的对象的发生和/或指示用于识别所述对象的训练的继续。

公开(公告)号：US20150283703A1

公开(公告)日：2015-10-08

申请号：US14244890

申请日：2014-04-03

Applicant: BRAIN CORPORATION

Inventor： Eugene M. Izhikevich ,  Patryk Laurent ,  Csaba Petre

IPC: B25J9/16 ,  G06N99/00

CPC classification number: B25J9/163 ,  G05B13/00 ,  G05D1/0022 ,  G05D1/0088 ,  G05D1/0246 ,  G05D2201/02 ,  G06N3/008 ,  G06N3/049 ,  G06N20/00 ,  Y10S901/47

Abstract: Robotic devices may be operated by users remotely. A learning controller apparatus may detect remote transmissions comprising user control instructions. The learning apparatus may receive sensory input conveying information about robot's state and environment (context). The learning apparatus may monitor one or more wavelength (infrared light, radio channel) and detect transmissions from user remote control device to the robot during its operation by the user. The learning apparatus may be configured to develop associations between the detected user remote control instructions and actions of the robot for given context. When a given sensory context occurs, the learning controller may automatically provide control instructions to the robot that may be associates with the given context. The provision of control instructions to the robot by the learning controller may obviate the need for user remote control of the robot thereby enabling autonomous operation by the robot.

Abstract translation: 机器人设备可以由用户远程操作。 学习控制器装置可以检测包括用户控制指令的远程传输。 学习装置可以接收关于机器人的状态和环境（上下文）的感官输入。 学习装置可以监视一个或多个波长（红外光，无线电信道），并且在用户操作期间检测从用户遥控装置到机器人的传输。 学习装置可以被配置为在给定的上下文之间发展检测到的用户远程控制指令和机器人的动作之间的关联。 当发生给定的感觉上下文时，学习控制器可以自动向可能与给定上下文相关联的机器人提供控制指令。 通过学习控制器向机器人提供控制指令可以避免用户对机器人的远程控制的需要，从而实现机器人的自主操作。

公开(公告)号：US09142063B2

公开(公告)日：2015-09-22

申请号：US13768471

申请日：2013-02-15

Applicant: Caterpillar Inc.

Inventor： Qi Chen ,  Michael D. Braunstein ,  Satish Yadav ,  Eric Alan Reiners ,  Kenneth Lee Stratton

IPC: G07C5/00 ,  G01S1/24 ,  G01S13/86 ,  G05D1/02 ,  G08G1/16 ,  G01S13/93 ,  G07C5/08

CPC classification number: G07C5/008 ,  B60W2300/17 ,  G01S13/865 ,  G01S13/867 ,  G01S2013/9332 ,  G05D1/024 ,  G05D1/0246 ,  G05D1/0257 ,  G05D1/0278 ,  G05D2201/02 ,  G05D2201/0202 ,  G05D2201/021 ,  G07C5/085 ,  G08G1/16

Abstract: A positioning system and method for determining a position of a machine on a worksite are disclosed. The method may store a map of the worksite which includes one or more known objects in the worksite. The method may determine whether a locating device associated with the machine is accurately providing the position of the machine. The method may also include detecting one or more objects in the worksite. The method may further determine an unmatched object from among the detected objects that does not match the one or more known objects stored in the map. The method may also store the unmatched object in the map as a known object of the worksite.

Abstract translation: 公开了一种用于确定机器在工地上的位置的定位系统和方法。 该方法可以存储在工地中包括一个或多个已知对象的工地的地图。 该方法可以确定与机器相关联的定位装置是否准确地提供机器的位置。 该方法还可以包括检测工地中的一个或多个对象。 该方法还可以从检测到的对象中确定与映射中存储的一个或多个已知对象不匹配的不匹配对象。 该方法还可以将不匹配的对象存储在地图中作为工地的已知对象。

公开(公告)号：US20150175202A1

公开(公告)日：2015-06-25

申请号：US14137954

申请日：2013-12-20

Applicant: Orbotix, Inc.

Inventor： Ross MacGregor ,  Ian H. Bernstein ,  Adam Wilson

IPC: B62D11/00 ,  G05D1/00 ,  B62D11/04

CPC classification number: G05D1/0011 ,  B60K1/04 ,  B60K7/0007 ,  B60K17/043 ,  B60K2001/0438 ,  B60Y2200/81 ,  B62D61/00 ,  B62K11/007 ,  G05D2201/02

Abstract: A self-propelled device is disclosed that includes a center of mass drive system. The self-propelled device includes a substantially cylindrical body and wheels, with each wheel having a diameter substantially equivalent to the body. The self-propelled device may further include an internal drive system with a center of mass below a rotational axis of the wheels. Operation and maneuvering of the self-propelled device may be performed via active displacement of the center of mass.

Abstract translation: 公开了一种包括质心驱动系统的自推进装置。 自推进装置包括基本上圆柱形的主体和轮子，每个车轮的直径基本上等于车身。 自推进装置还可以包括内部驱动系统，其质心小于车轮的旋转轴线。 自推进装置的操作和操纵可以通过主动位移的质心进行。

公开(公告)号：US08812156B2

公开(公告)日：2014-08-19

申请号：US12775101

申请日：2010-05-06

Applicant: Chang-Eun Lee ,  Beom-Su Seo ,  Joong-Bae Kim ,  Sung-Hoon Kim ,  Myung-Chan Roh ,  Jae-Min Byun ,  Kyeong-Ho Lee

Inventor： Chang-Eun Lee ,  Beom-Su Seo ,  Joong-Bae Kim ,  Sung-Hoon Kim ,  Myung-Chan Roh ,  Jae-Min Byun ,  Kyeong-Ho Lee

IPC: G05D1/00 ,  A01M29/06 ,  G06K9/32

CPC classification number: A01M31/002 ,  A01M29/06 ,  A01M29/10 ,  A01M29/16 ,  G05D1/0027 ,  G05D1/0038 ,  G05D2201/02

Abstract: Provided is a method including receiving information on a surrounding situation detected by the mobile robot; detecting birds from the received surrounding situation information; allocating a birds control mission to the mobile robot by extracting a birds control pattern corresponding to the surrounding situation; and verifying a result in accordance with performing the allocated birds control mission from the mobile robot. By controlling the birds so as to, in advance, prevent a loss of lives and economical loss which may be caused when the birds collide with airplanes at the airport, it is possible to improve productivity and efficiency of a birds repelling job in an airport and provide construction of a new type of aviation maintenance business model by activating an air traffic control industry through providing a safer airplane operating model while saving operating personnel costs for preventing collision of birds.

Abstract translation: 提供了一种包括接收由移动机器人检测到的周围情况的信息的方法; 从收到的周边情况信息中检测鸟类; 通过提取与周围情况相对应的鸟类控制模式，为移动机器人分配鸟类控制任务; 以及根据从移动机器人执行所分配的鸟类控制任务来验证结果。 通过控制鸟类，预先防止鸟类在机场与飞机相撞时可能造成的生命损失和经济损失，可以提高机场的排斥工作的生产力和效率， 通过提供更安全的飞机运行模式，启动空中交通管制行业，同时节省运营人员防止鸟类碰撞的成本，提供新型航空维修业务模式的建设。

公开(公告)号：US20120041592A1

公开(公告)日：2012-02-16

申请号：US13084350

申请日：2011-04-11

Applicant: Thomas J. Murray, IV ,  Baoquoc N. Pham ,  Paolo Pirjanian

Inventor： Thomas J. Murray, IV ,  Baoquoc N. Pham ,  Paolo Pirjanian

IPC: B25J9/16

CPC classification number: G06N3/008 ,  B25J9/161 ,  B25J9/1661 ,  B25J9/1666 ,  G05B2219/40389 ,  G05B2219/40397 ,  G05D1/0088 ,  G05D2201/02 ,  G06F9/455 ,  Y10T436/11

Abstract: Methods and apparatus that provide a hardware abstraction layer (HAL) for a robot are disclosed. A HAL can reside as a software layer or as a firmware layer residing between robot control software and underlying robot hardware and/or an operating system for the hardware. The HAL provides a relatively uniform abstract for aggregates of underlying hardware such that the underlying robotic hardware is transparent to perception and control software, i.e., robot control software. This advantageously permits robot control software to be written in a robot-independent manner. Developers of robot control software are then freed from tedious lower level tasks. Portability is another advantage. For example, the HAL efficiently permits robot control software developed for one robot to be ported to another. In one example, the HAL permits the same navigation algorithm to be ported from a wheeled robot and used on a humanoid legged robot.

Abstract translation: 公开了一种为机器人提供硬件抽象层（HAL）的方法和装置。 HAL可以作为软件层或驻留在机器人控制软件和底层机器人硬件和/或硬件的操作系统之间的固件层存在。 HAL为底层硬件的聚合提供了相对均匀的抽象，使得潜在的机器人硬件对感知和控制软件即机器人控制软件是透明的。 这有利地允许以机器人独立的方式写入机器人控制软件。 机器人控制软件的开发者然后摆脱繁琐的低级任务。 便携性是另一个优点。 例如，HAL有效地允许为一个机器人开发的机器人控制软件移植到另一个机器人。 在一个示例中，HAL允许从轮式机器人移植相同的导航算法，并在人形腿式机器人上使用。

公开(公告)号：US20110054691A1

公开(公告)日：2011-03-03

申请号：US12775101

申请日：2010-05-06

Applicant: Chang-Eun LEE ,  Beom-Su Seo ,  Joong-Bae Kim ,  Sung-Hoon Kim ,  Myung-Chen Roh ,  Jae-Min Byun ,  Kyeong-Ho Lee

Inventor： Chang-Eun LEE ,  Beom-Su Seo ,  Joong-Bae Kim ,  Sung-Hoon Kim ,  Myung-Chen Roh ,  Jae-Min Byun ,  Kyeong-Ho Lee

IPC: G05B15/00

CPC classification number: A01M31/002 ,  A01M29/06 ,  A01M29/10 ,  A01M29/16 ,  G05D1/0027 ,  G05D1/0038 ,  G05D2201/02

Abstract: Provides is a method including receiving information on a surrounding situation detected by the mobile robot; detecting birds from the received surrounding situation information; allocating a birds control mission to the mobile robot by extracting a birds control pattern corresponding to the surrounding situation; and verifying a result in accordance with performing the allocated birds control mission from the mobile robot. By controlling the birds so as to, in advance, prevent a loss of lives and economical loss which may be caused when the birds collide with airplanes at the airport, it is possible to improve productivity and efficiency of a birds repelling job in an airport and provide construction of a new type of aviation maintenance business model by activating an air traffic control industry through providing a safer airplane operating model while saving operating personnel costs for preventing collision of birds.

Abstract translation: 提供是一种包括接收由移动机器人检测到的周围情况的信息的方法; 从收到的周边情况信息中检测鸟类; 通过提取与周围情况相对应的鸟类控制模式，为移动机器人分配鸟类控制任务; 以及根据从移动机器人执行所分配的鸟类控制任务来验证结果。 通过控制鸟类，预先防止鸟类在机场与飞机相撞时可能造成的生命损失和经济损失，可以提高机场的排斥工作的生产力和效率， 通过提供更安全的飞机运行模式，启动空中交通管制行业，同时节省运营人员防止鸟类碰撞的成本，提供新型航空维修业务模式的建设。

公开(公告)号：US20080147691A1

公开(公告)日：2008-06-19

申请号：US12025540

申请日：2008-02-04

Applicant: Richard Alan Peters

Inventor： Richard Alan Peters

IPC: G06F17/30 ,  G05B19/042

CPC classification number: G06N3/008 ,  B25J9/161 ,  B25J9/163 ,  G05B2219/39254 ,  G05D1/0088 ,  G05D1/0274 ,  G05D2201/02 ,  G06N99/005

Abstract: An architecture for robot intelligence enables a robot to learn new behaviors and create new behavior sequences autonomously and interact with a dynamically changing environment. Sensory information is mapped onto a Sensory Ego-Sphere (SES) that rapidly identifies important changes in the environment and functions much like short term memory. Behaviors are stored in a DBAM that creates an active map from the robot's current state to a goal state and functions much like long term memory. A dream state converts recent activities stored in the SES and creates or modifies behaviors in the DBAM.

Abstract translation: 机器人智能的架构使机器人能够自主学习新行为并创建新的行为序列，并与动态变化的环境进行交互。 感官信息被映射到感觉自由球（SES）上，其快速识别环境中的重要变化，并且功能非常类似于短期记忆。 行为存储在DBAM中，它创建一个从机器人当前状态到目标状态的活动映射，并且功能非常像长期内存。 梦境状态转换存储在SES中的最新活动，并创建或修改DBAM中的行为。

公开(公告)号：US20080071423A1

公开(公告)日：2008-03-20

申请号：US11945893

申请日：2007-11-27

Applicant: Thomas Murray, IV ,  Baoquoc Pham ,  Paolo Pirjanian

Inventor： Thomas Murray, IV ,  Baoquoc Pham ,  Paolo Pirjanian

IPC: G05B19/00

CPC classification number: G06N3/008 ,  B25J9/161 ,  B25J9/1661 ,  B25J9/1666 ,  G05B2219/40389 ,  G05B2219/40397 ,  G05D1/0088 ,  G05D2201/02 ,  G06F9/455 ,  Y10T436/11

Abstract: Methods and apparatus that provide a hardware abstraction layer (HAL) for a robot are disclosed. A HAL can reside as a software layer or as a firmware layer residing between robot control software and underlying robot hardware and/or an operating system for the hardware. The HAL provides a relatively uniform abstract for aggregates of underlying hardware such that the underlying robotic hardware is transparent to perception and control software, i.e., robot control software. This advantageously permits robot control software to be written in a robot-independent manner. Developers of robot control software are then freed from tedious lower level tasks. Portability is another advantage. For example, the HAL efficiently permits robot control software developed for one robot to be ported to another. In one example, the HAL permits the same navigation algorithm to be ported from a wheeled robot and used on a humanoid legged robot.

Abstract translation: 公开了一种为机器人提供硬件抽象层（HAL）的方法和装置。 HAL可以作为软件层或驻留在机器人控制软件和底层机器人硬件和/或硬件的操作系统之间的固件层存在。 HAL为底层硬件的聚合提供了相对均匀的抽象，使得潜在的机器人硬件对感知和控制软件即机器人控制软件是透明的。 这有利地允许以机器人独立的方式写入机器人控制软件。 机器人控制软件的开发者然后摆脱繁琐的低级任务。 便携性是另一个优点。 例如，HAL有效地允许为一个机器人开发的机器人控制软件移植到另一个机器人。 在一个示例中，HAL允许从轮式机器人移植相同的导航算法，并在人形腿式机器人上使用。

公开(公告)号：US07328196B2

公开(公告)日：2008-02-05

申请号：US10749326

申请日：2003-12-31

Applicant: Richard Alan Peters, II

Inventor： Richard Alan Peters, II

IPC: G06N5/00

CPC classification number: G06N3/008 ,  B25J9/161 ,  B25J9/163 ,  G05B2219/39254 ,  G05D1/0088 ,  G05D1/0274 ,  G05D2201/02 ,  G06N99/005

Abstract: An architecture for robot intelligence enables a robot to learn new behaviors and create new behavior sequences autonomously and interact with a dynamically changing environment. Sensory information is mapped onto a Sensory Ego-Sphere (SES) that rapidly identifies important changes in the environment and functions much like short term memory. Behaviors are stored in a DBAM that creates an active map from the robot's current state to a goal state and functions much like long term memory. A dream state converts recent activities stored in the SES and creates or modifies behaviors in the DBAM.

Abstract translation: 机器人智能的架构使机器人能够自主学习新行为并创建新的行为序列，并与动态变化的环境进行交互。 感官信息被映射到感觉自由球（SES）上，其快速识别环境中的重要变化，并且功能非常类似于短期记忆。 行为存储在DBAM中，它创建一个从机器人当前状态到目标状态的活动映射，并且功能非常像长期内存。 梦境状态转换存储在SES中的最新活动，并创建或修改DBAM中的行为。