中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

56 records (took 0.03 seconds)

    System and Method to Generate Augmented Training Data for Neural Network
    1.
    发明申请
    System and Method to Generate Augmented Training Data for Neural Network 有权

    公开(公告)号：US20250014322A1

    公开(公告)日：2025-01-09

    申请号：US18570721

    申请日：2021-06-17

    Applicant: ABB Schweiz AG

    Inventor： Qilin Zhang ,  Yinwei Zhang ,  Biao Zhang ,  Jorge Vidal-Ribas

    IPC: G06V10/82

    Abstract: A robotic system capable of being trained with a plurality of images that are synthetically augmented from an initial image data set includes a training system toward that end. An image augmentation system includes in one form a neural network trained to generate synthetic images using a generative adversarial network which includes the ability to synthesize images having various poses with adjustments to image parameters such as light and color among potential others. In another form the image augmentation system includes a set of images projected or transformed from its original pose to a number of different poses using an affine transform, and the ability to progress across an entire dimensional space of anticipated robot movements which produce various potential poses.

    System and Method for Robotic Calibration and Tuning
    2.
    发明申请
    System and Method for Robotic Calibration and Tuning 有权

    公开(公告)号：US20220410397A1

    公开(公告)日：2022-12-29

    申请号：US17772370

    申请日：2019-10-29

    Applicant: Biao Zhang ,  ABB Schweiz AG

    Inventor： Biao Zhang ,  Saumya Sharma ,  Yixin Liu ,  Jianjun Wang ,  William J. Eakins ,  Andrew M. Salm ,  Yun Hsuan Su ,  Jorge Vidal-Ribas ,  Jordi Artigas ,  Ramon Casanelles

    IPC: B25J9/16

    Abstract: A system and method for use of artificial and nature racking features to calibrate sensors and tune a robotic control system of a sensor fusion guided robotic assembly. The artificial tracking features can have a configuration, or be at a location, that may be less susceptible to noise and error. Thus, the sensors can at least initially be calibrated, and the control system initially tuned, using the first tracking features until the sensors and control system satisfy operation performance criteria. Second tracking features, which may correspond to features on a workpiece that will be utilized in an assembly operation performed by the robot. By pre-calibrating the sensors, and pre-tuning the control system prior to calibration using the second tracking features, sensor calibration and system tuning based on the second tracking features can be attained faster and with less complexity.

    METHOD AND SYSTEM FOR OBJECT TRACKING IN ROBOTIC VISION GUIDANCE
    4.
    发明申请
    METHOD AND SYSTEM FOR OBJECT TRACKING IN ROBOTIC VISION GUIDANCE 有权

    公开(公告)号：US20210331322A1

    公开(公告)日：2021-10-28

    申请号：US16856508

    申请日：2020-04-23

    Applicant: ABB Schweiz AG

    Inventor： Biao Zhang ,  Jianjun Wang ,  Yixin Liu ,  Saumya Sharma ,  Jorge Vidal-Ribas ,  Jordi Artigas ,  Ramon Casanelles

    IPC: B25J9/16 ,  B25J13/08

    Abstract: A system and method for predicting the location at which a feature that is being tracked during a robotic assembly operation will be located within one or more images captured by a vision device. A vision device can be mounted to a robot such that the location of the vision device as the robot moves can be known or determined. In the event of an interruption of the tracking of the feature by the vision device as the corresponding workpiece is moving, the location of the feature relative to a vision device can be predicted, such as, via use of current or past historical movement information for the feature and/or the associated workpiece. Using the predicted location of the feature and the known location of the vision device, the location at which the feature will be located in an image(s) captured by the vision device can be predicted.

    Method for operating a robot
    5.
    发明授权
    Method for operating a robot 有权

    公开(公告)号：US10828778B2

    公开(公告)日：2020-11-10

    申请号：US15827334

    申请日：2017-11-30

    Applicant: ABB Schweiz AG

    Inventor： Biao Zhang ,  Xiongzi Li ,  George Q. Zhang

    IPC: B25J9/16

    Abstract: A method for operating a robot includes: creating a production robot program for execution on a robotic controller, wherein the robot program defines a robot path; performing an offline simulation of robot motion along the robot path using the production robot program; analyzing loads between a robot end effector and an object along the robot path, based on the offline simulation, to identify a maximum load experienced during the simulation; tuning production robot program parameters to reduce the maximum load if the maximum load is not within a predefined limit; generating a test robot program to test the end effector and the object with the maximum load within the predefined limit; testing the end effector with the object online using the test robot program; repeating the tuning and testing until no objects are dropped during the testing; and operating the robot during production using tuned robot program parameters.

    Robotic Systems and Methods Used with Installation of Component Parts
    10.
    发明公开
    Robotic Systems and Methods Used with Installation of Component Parts 审中-公开

    公开(公告)号：US20240278434A1

    公开(公告)日：2024-08-22

    申请号：US18570156

    申请日：2021-06-17

    Applicant: ABB Schweiz AG

    Inventor： Yinwei Zhang ,  Qilin Zhang ,  Biao Zhang ,  Jorge Vidal-Ribas

    IPC: B25J9/16 ,  B25J11/00

    CPC classification number: B25J9/1697 ,  B25J9/1664 ,  B25J11/0075

    Abstract: A robotic system for use in installing final trim and assembly part includes an auto-labeling system that combines images of a primary component, such as a vehicle, with those of computer based model, where feature based object tracking methods are used to compare the two. In some forms a camera can be mounted to a moveable robot, while in other the camera can be fixed in position relative to the robot. An artificial marker can be used in some forms. Robot movement tracking can also be used. A runtime operation can utilize a deep learning network to augment feature-based object tracking to aid in initializing a pose of the vehicle as well as an aid in restoring tracking if lost.

Patent Agency Ranking