公开(公告)号：US20220212342A1

公开(公告)日：2022-07-07

申请号：US17574760

申请日：2022-01-13

Applicant: Brain Corporation

Inventor： Patryk Laurent ,  Jean-Baptiste Passot ,  Oleg Sinyavskiy ,  Filip Ponulak ,  Borja Ibarz Gabardos ,  Eugene Izhikevich

IPC: B25J9/16 ,  G06N3/00 ,  G06N3/04

Abstract: Robotic devices may be trained by a user guiding the robot along target action trajectory using an input signal. A robotic device may comprise an adaptive controller configured to generate control signal based on one or more of the user guidance, sensory input, performance measure, and/or other information. Training may comprise a plurality of trials, wherein for a given context the user and the robot's controller may collaborate to develop an association between the context and the target action. Upon developing the association, the adaptive controller may be capable of generating the control signal and/or an action indication prior and/or in lieu of user input. The predictive control functionality attained by the controller may enable autonomous operation of robotic devices obviating a need for continuing user guidance.

公开(公告)号：US10688657B2

公开(公告)日：2020-06-23

申请号：US16171635

申请日：2018-10-26

Applicant: Brain Corporation

Inventor： Eugene Izhikevich ,  Oleg Sinyavskiy ,  Jean-Baptiste Passot

IPC: G06F19/00 ,  B25J9/16 ,  G06N3/00 ,  G06N3/04

Abstract: Apparatus and methods for training and operating of robotic devices. Robotic controller may comprise a predictor apparatus configured to generate motor control output. The predictor may be operable in accordance with a learning process based on a teaching signal comprising the control output. An adaptive controller block may provide control output that may be combined with the predicted control output. The predictor learning process may be configured to learn the combined control signal. Predictor training may comprise a plurality of trials. During initial trial, the control output may be capable of causing a robot to perform a task. During intermediate trials, individual contributions from the controller block and the predictor may be inadequate for the task. Upon learning, the control knowledge may be transferred to the predictor so as to enable task execution in absence of subsequent inputs from the controller. Control output and/or predictor output may comprise multi-channel signals.

公开(公告)号：US20190302791A1

公开(公告)日：2019-10-03

申请号：US15942123

申请日：2018-03-30

Applicant: Brain Corporation

Inventor： Jayram Moorkanikara Nageswaran ,  Oleg Sinyavskiy ,  Borja Ibarz Gabardos

IPC: G05D1/02 ,  G05D1/00

Abstract: Systems and methods for robotic mapping are disclosed. In some example implementations, an automated device can travel in an environment. From travelling in the environment, the automated device can create a graph comprising a plurality of nodes, wherein each node corresponds to a scan taken by one or more sensors of the automated device at a location in the environment. In some example embodiments, the automated device can reevaluate its travel along a desired path if it encounters objects or obstructions along its path, whether those objects or obstructions are present in the front, rare or side of the automated device. In some example embodiments, the automated device uses a timestamp methodology to maneuver around its environment that provides faster processing and less usage of memory space.

公开(公告)号：US20190299410A1

公开(公告)日：2019-10-03

申请号：US16376237

申请日：2019-04-05

Applicant: BRAIN CORPORATION

Inventor： Oleg Sinyavskiy ,  Jean-Baptiste Passot ,  Borja Ibarz Gabardos ,  Diana Vu Le

IPC: B25J9/16 ,  G05D1/02 ,  A47L11/40 ,  A47L11/283 ,  G01C21/00 ,  B25J11/00 ,  G01C21/34 ,  G01C21/32

Abstract: Systems and methods for robotic path planning are disclosed. In some implementations of the present disclosure, a robot can generate a cost map associated with an environment of the robot. The cost map can comprise a plurality of pixels each corresponding to a location in the environment, where each pixel can have an associated cost. The robot can further generate a plurality of masks having projected path portions for the travel of the robot within the environment, where each mask comprises a plurality of mask pixels that correspond to locations in the environment. The robot can then determine a mask cost associated with each mask based at least in part on the cost map and select a mask based at least in part on the mask cost. Based on the projected path portions within the selected mask, the robot can navigate a space.

公开(公告)号：US10377040B2

公开(公告)日：2019-08-13

申请号：US15423442

申请日：2017-02-02

Applicant: BRAIN CORPORATION

Inventor： Oleg Sinyavskiy ,  Jean-Baptiste Passot ,  Borja Ibarz Gabardos ,  Diana Vu Le

IPC: B25J9/16 ,  G05D1/00 ,  G05D1/02

Abstract: Systems and methods assisting a robotic apparatus are disclosed. In some exemplary implementations, a robot can encounter situations where the robot cannot proceed and/or does not know with a high degree of certainty it can proceed. Accordingly, the robot can determine that it has encountered an error and/or assist event. In some exemplary implementations, the robot can receive assistance from an operator and/or attempt to resolve the issue itself. In some cases, the robot can be configured to delay actions in order to allow resolution of the error and/or assist event.

公开(公告)号：US20190009408A1

公开(公告)日：2019-01-10

申请号：US16131128

申请日：2018-09-14

Applicant: Brain Corporation

Inventor： Botond Szatmary ,  Oyvind Grotmol ,  Eugene Izhikevich ,  Oleg Sinyavskiy

IPC: B25J9/16 ,  B25J9/00 ,  G06N99/00

Abstract: Apparatus and methods for training and operating of robotic devices. Robotic controller may comprise a plurality of predictor apparatus configured to generate motor control output. One predictor may be operable in accordance with a pre-configured process; another predictor may be operable in accordance with a learning process configured based on a teaching signal. An adaptive combiner component may be configured to determine a combined control output controller block may provide control output that may be combined with the predicted control output. The pre-programmed predictor may be configured to operate a robot to perform a task. Based on detection of a context, the controller may adaptively switch to use control output of the learning process to perform the given or another task. User feedback may be utilized during learning.

公开(公告)号：US20180243903A1

公开(公告)日：2018-08-30

申请号：US15967240

申请日：2018-04-30

Applicant: Brain Corporation

Inventor： Jean-Baptiste Passot ,  Oleg Sinyavskiy ,  Eugene Izhikevich

IPC: B25J9/16 ,  G06N99/00

CPC classification number: B25J9/163 ,  B25J9/161 ,  G05B2219/33034 ,  G05B2219/39289 ,  G05B2219/39298 ,  G06N3/008 ,  G06N3/049 ,  G06N3/063 ,  G06N3/08 ,  G06N20/00 ,  Y10S901/03

Abstract: Apparatus and methods for training and controlling of, for instance, robotic devices. In one implementation, a robot may be trained by a user using supervised learning. The user may be unable to control all degrees of freedom of the robot simultaneously. The user may interface to the robot via a control apparatus configured to select and operate a subset of the robot's complement of actuators. The robot may comprise an adaptive controller comprising a neuron network. The adaptive controller may be configured to generate actuator control commands based on the user input and output of the learning process. Training of the adaptive controller may comprise partial set training. The user may train the adaptive controller to operate first actuator subset. Subsequent to learning to operate the first subset, the adaptive controller may be trained to operate another subset of degrees of freedom based on user input via the control apparatus.

公开(公告)号：US20180001474A1

公开(公告)日：2018-01-04

申请号：US15199224

申请日：2016-06-30

Applicant: BRAIN CORPORATION

Inventor： Oleg Sinyavskiy ,  Borja Ibarz Gabardos ,  Jean-Baptiste Passot

IPC: B25J9/16 ,  B25J5/00

CPC classification number: B25J9/1676 ,  B25J5/007 ,  B25J9/1697 ,  G05B2219/40442 ,  G05B2219/49143 ,  G05B2219/49157 ,  Y10S901/01

Abstract: Systems and methods for detection of people are disclosed. In some exemplary implementations, a robot can have a plurality of sensor units. Each sensor unit can be configured to generate sensor data indicative of a portion of a moving body at a plurality of times. Based on at least the sensor data, the robot can determine that the moving body is a person by at least detecting the motion of the moving body and determining that the moving body has characteristics of a person. The robot can then perform an action based at least in part on the determination that the moving body is a person.

公开(公告)号：US09597797B2

公开(公告)日：2017-03-21

申请号：US14102410

申请日：2013-12-10

Applicant: BRAIN CORPORATION

Inventor： Filip Ponulak ,  Moslem Kazemi ,  Patryk Laurent ,  Oleg Sinyavskiy ,  Eugene Izhikevich

IPC: G05B19/18 ,  B25J9/16 ,  G05D1/00 ,  G05D1/02 ,  G06N3/00 ,  G06N3/04 ,  G06N99/00

CPC classification number: B25J9/163 ,  B25J9/161 ,  G05B2219/36418 ,  G05B2219/36425 ,  G05B2219/40499 ,  G05D1/005 ,  G05D1/0088 ,  G05D1/0221 ,  G05D2201/02 ,  G06N3/008 ,  G06N3/049 ,  G06N99/005

Abstract: Robotic devices may be trained by a trainer guiding the robot along a target trajectory using physical contact with the robot. The robot may comprise an adaptive controller configured to generate control commands based on one or more of the trainer input, sensory input, and/or performance measure. The trainer may observe task execution by the robot. Responsive to observing a discrepancy between the target behavior and the actual behavior, the trainer may provide a teaching input via a haptic action. The robot may execute the action based on a combination of the internal control signal produced by a learning process of the robot and the training input. The robot may infer the teaching input based on a comparison of a predicted state and actual state of the robot. The robot's learning process may be adjusted in accordance with the teaching input so as to reduce the discrepancy during a subsequent trial.

公开(公告)号：US09489623B1

公开(公告)日：2016-11-08

申请号：US14054366

申请日：2013-10-15

Applicant: Brain Corporation

Inventor： Oleg Sinyavskiy ,  Vadim Polonichko

IPC: G06N3/08 ,  B25J9/16

CPC classification number: G06N3/084 ,  B25J9/163 ,  G06N3/008 ,  G06N3/0445 ,  G06N3/049 ,  G06N3/063 ,  Y10S901/50

Abstract: Apparatus and methods for developing robotic controllers comprising parallel networks. In some implementations, a parallel network may comprise at least first and second neuron layers. The second layer may be configured to determine a measure of discrepancy (error) between a target network output and actual network output. The network output may comprise control signal configured to cause a task execution by the robot. The error may be communicated back to the first neuron layer in order to adjust efficacy of input connections into the first layer. The error may be encoded into spike latency using linear or nonlinear encoding. Error communication and control signal provision may be time multiplexed so as to enable target action execution. Efficacy associated with forward and backward/reverse connections may be stored in individual arrays. A synchronization mechanism may be employed to match forward/reverse efficacy in order to implement plasticity.

Abstract translation: 用于开发包括并联网络的机器人控制器的装置和方法。 在一些实现中，并行网络可以包括至少第一和第二神经元层。 第二层可以被配置为确定目标网络输出和实际网络输出之间的差异（误差）的度量。 网络输出可以包括被配置为导致机器人执行任务的控制信号。 该误差可以被传送回第一神经元层，以便调节输入连接到第一层的功效。 可以使用线性或非线性编码将误差编码为尖峰等待时间。 错误通信和控制信号提供可以被时分复用以便能够执行目标动作执行。 与前向和后向/反向连接相关联的功效可以存储在单独的阵列中。 为了实现可塑性，可以采用同步机制来匹配正向/反向效能。