DISTRIBUTED ROBOTIC DEMONSTRATION LEARNING

    公开(公告)号:US20210362327A1

    公开(公告)日:2021-11-25

    申请号:US16880724

    申请日:2020-05-21

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for distributed robotic demonstration learning. One of the methods includes receiving a skill template to be trained to cause a robot to perform a particular skill having a plurality of subtasks. One or more demonstration subtasks defined by the skill template are identified, wherein each demonstration subtask is an action to be refined using local demonstration data. On online execution system uploads sets of local demonstration data to a cloud-based training system. The cloud-based training system generates respective trained model parameters for each set of local demonstration data. The skill template is executed on the robot using the trained model parameters generated by the cloud-based training system.

    System and method for determining tool offsets

    公开(公告)号:US10016892B2

    公开(公告)日:2018-07-10

    申请号:US14807434

    申请日:2015-07-23

    Abstract: Systems and methods are disclosed for determining tool offset data for a tool attached to a robot at an attachment point. In an embodiment, a method includes controlling the robot to contact a reference object with the tool. The reference object is a rigid object with a known location. A force feedback sensor of the robot may indicates when the tool has contacted the reference object. Once contact is made, data indicating robot position during tool contact is received. Additionally, the robot may temporarily stops movement of the tool to prevent damage to the tool or the reference object. Next, tool offset data s determined based on the position of the reference object relative to the robot and the received robot position data. The tool offset data describes the distance between at least one point on the tool and the attachment point.

    SKILL TEMPLATES FOR ROBOTIC DEMONSTRATION LEARNING

    公开(公告)号:US20210362331A1

    公开(公告)日:2021-11-25

    申请号:US16880862

    申请日:2020-05-21

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using skill templates for robotic demonstration learning. One of the methods includes receiving a skill template for a task to be performed by a robot, wherein the skill template defines a state machine having a plurality of subtasks and one or more respective transition conditions between one or more of the subtasks. Local demonstration data for a demonstration subtask of the skill template is received, where the local demonstration data is generated from a user demonstrating how to perform the demonstration subtask with the robot. A machine learning model is refined for the demonstration subtask and the skill template is executed on the robot, causing the robot to transition through the state machine defined by the skill template to perform the task.

    INTEGRATING SENSOR STREAMS FOR ROBOTIC DEMONSTRATION LEARNING

    公开(公告)号:US20210362329A1

    公开(公告)日:2021-11-25

    申请号:US16880857

    申请日:2020-05-21

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for integrating sensor streams for robotic demonstration learning. One of the methods includes selecting, by a learning system for a robot, a base update rate for combining multiple sensor streams into a task state representation. The learning system repeatedly generates the task state representation at the base update rate, including combining, during each time period defined by the update rate, the task state representation from most recently updated sensor data processed by the plurality of neural networks. The learning system repeatedly uses the task state representations to generate commands for the robot at the base update rate.

    System and method for determining tool offsets

    公开(公告)号:US10583555B2

    公开(公告)日:2020-03-10

    申请号:US16002205

    申请日:2018-06-07

    Abstract: Systems and methods are disclosed for determining tool offset data for a tool attached to a robot at an attachment point. In an embodiment, a method includes controlling the robot to contact a reference object with the tool. The reference object is a rigid object with a known location. A force feedback sensor of the robot indicates when the tool has contacted the reference object. Once contact is made, data indicating robot position during tool contact is received. Additionally, the robot temporarily stops movement of the tool to prevent damage to the tool or the reference object. Next, tool offset data is determined based on the position of the reference object relative to the robot and the received robot position data. The tool offset data describes the distance between at least one point on the tool and the attachment point.

    System and method for determining a work offset

    公开(公告)号:US10456914B2

    公开(公告)日:2019-10-29

    申请号:US15729953

    申请日:2017-10-11

    Abstract: Example systems and methods are disclosed for determining work offset data for a robot in a work environment. A robot operating in a work environment may receive an indication to determine a work offset. The work offset may describe the location and angular orientation of a working plane of the work environment relative to a base plane of the robot. In response to the indication, the robot may identify the working plane. The robot may be controlled to contact one or more points of the working plane. The robot may determine respective point locations of the contacted points relative to the base plane based on the respective positions of the robot at respective times of contact. The robot may determine the location and angular orientation of the working plane relative to the base plane based on the determined respective point locations of the contacted points.

    System And Method For Determining Tool Offsets

    公开(公告)号:US20180281175A1

    公开(公告)日:2018-10-04

    申请号:US16002205

    申请日:2018-06-07

    Abstract: Example systems and methods are disclosed for determining tool offset data for a tool attached to a robot at an attachment point. The method may include controlling the robot to contact a reference object with the tool. The reference object may be a rigid object with a known location. A force feedback sensor of the robot may indicate when the tool has contacted the reference object. Once contact is made, data indicating robot position during tool contact may be received. Additionally, the robot may temporarily stop movement of the tool to prevent damage to the tool or the reference object. Next, tool offset data may be determined based on the position of the reference object relative to the robot and the received robot position data. The tool offset data may describe the distance between at least one point on the tool and the attachment point.

    SIMULATED LOCAL DEMONSTRATION DATA FOR ROBOTIC DEMONSTRATION LEARNING

    公开(公告)号:US20210362333A1

    公开(公告)日:2021-11-25

    申请号:US16880869

    申请日:2020-05-21

    Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for using simulated local demonstration data for robotic demonstration learning. One of the methods includes receiving perceptual data of a workcell of a robot to be configured to execute a task according to a skill template, wherein the skill template specifies one or more subtasks required to perform the skill, wherein at least one of the subtasks is a demonstration subtask that relies on learning visual characteristics of the workcell. A virtual model is generated of a portion of the workcell. A training system generates simulated local demonstration data from the virtual model of the portion of the workcell and tunes a base control policy for the demonstration subtask using the simulated local demonstration data generated from the virtual model of the portion of the workcell.

    Adapting programming of a robot and/or control of the robot based on one or more parameters of an end effector of the robot

    公开(公告)号:US10166674B1

    公开(公告)日:2019-01-01

    申请号:US15599221

    申请日:2017-05-18

    Abstract: Methods and apparatus that adapt programming and/or control of a robot. The robot has at least one attachment area that can removably couple with any one of a plurality of end effectors. The programming and/or control is adapted based on received parameters of an end effector attached (or to be attached) to the attachment area of the robot. Some implementations are directed to adapting graphical user interface output of a robot programming application based on one or more parameters of an end effector for a robot being programmed via the robot programming application. Some implementations are directed to adapting control of a robot based on one or more parameters of an end effector attached (or to be attached) to the robot.

Patent Agency Ranking