公开(公告)号：US20180290302A1

公开(公告)日：2018-10-11

申请号：US15947281

申请日：2018-04-06

Applicant: FANUC CORPORATION

Inventor： Kaimeng Wang ,  Wenjie Chen ,  Kouichirou Hayashi

IPC: B25J9/16

Abstract: A device that can prevent a decrease in an efficiency of a manufacturing line. The device includes a shape acquisition section for acquiring a shape of a workpiece; a motion pattern acquisition section for acquiring basic motion patterns including a reference workpiece shape, a reference working position in the reference workpiece shape, and a type of an operation carried out on the reference working position; a similarity determination section for determining whether a shape of the workpiece is similar to the reference work piece shape; a position determination section for, based on a shape of the workpiece and the reference workpiece shape, determining the working position on the workpiece that corresponds to the reference working position; and an motion-path generation section for, by changing the reference working position to the determined working position, generating a motion path.

公开(公告)号：US20240201677A1

公开(公告)日：2024-06-20

申请号：US18068760

申请日：2022-12-20

Applicant: FANUC CORPORATION

Inventor： Kaimeng Wang ,  Yu Zhao

IPC: G05B19/423 ,  B25J9/16 ,  G06N3/092

CPC classification number: G05B19/423 ,  B25J9/1697 ,  G06N3/092

Abstract: A method for teaching a robot to perform an operation including human demonstration using inverse reinforcement learning and a reinforcement learning reward function. A demonstrator performs an operation with contact force and workpiece motion data recorded. The demonstration data is used to train an encoder neural network which captures the human skill, defining a Gaussian distribution of probabilities for a set of states and actions. Encoder and decoder neural networks are then used in live robotic operations, where the decoder is used by a robot controller to compute actions based on force and motion state data from the robot. After each operation, the reward function is computed, with a Kullback-Leibler divergence term which rewards a small difference between human demonstration and robot operation probability curves, and a completion term which rewards a successful operation by the robot. The decoder is trained using reinforcement learning to maximize the reward function.

公开(公告)号：US11207788B2

公开(公告)日：2021-12-28

申请号：US16266231

申请日：2019-02-04

Applicant: FANUC CORPORATION

Inventor： Wenjie Chen ,  Tetsuaki Kato ,  Kaimeng Wang

IPC: B25J15/10 ,  B25J9/16 ,  B25J13/08

Abstract: A hand control apparatus including an extracting unit extracting a grip pattern of an object having a shape closest to that of the object acquired by a shape acquiring unit from a storage unit storing and associating shapes of plural types of objects and grip patterns, a position and posture calculating unit calculating a gripping position and posture of the hand in accordance with the extracted grip pattern, a hand driving unit causing the hand to grip the object based on the calculated gripping position and posture, a determining unit determining if a gripped state of the object is appropriate based on information acquired by at least one of the shape acquiring unit, a force sensor and a tactile sensor, and a gripped state correcting unit correcting at least one of the gripping position and the posture when it is determined that the gripped state of the object is inappropriate.

公开(公告)号：US11130236B2

公开(公告)日：2021-09-28

申请号：US16293965

申请日：2019-03-06

Applicant: FANUC CORPORATION

Inventor： Wenjie Chen ,  Kaimeng Wang ,  Tetsuaki Kato

IPC: B25J9/16 ,  B25J9/00

Abstract: A robot movement teaching apparatus including a movement path extraction unit configured to process time-varying images of a first workpiece and fingers or arms of a human working on the first workpiece, and thereby extract a movement path of the fingers or arms of the human; a mapping generation unit configured to generate a transform function for transformation from the first workpiece to a second workpiece worked on by a robot, based on feature points of the first workpiece and feature points of the second workpiece; and a movement path generation unit configured to generate a movement path of the robot based on the movement path of the fingers or arms of the human extracted by the movement path extraction unit and based on the transform function generated by the mapping generation unit.

公开(公告)号：US10994422B2

公开(公告)日：2021-05-04

申请号：US16201550

申请日：2018-11-27

Applicant: FANUC CORPORATION

Inventor： Hajime Suzuki ,  Shuusuke Watanabe ,  Kaimeng Wang

IPC: B25J13/06 ,  G05B19/4097 ,  B25J9/16

Abstract: A robot system that performs desired processing on a processing target object using a processing tool. The robot system includes a robot having an arm tip that holds the processing tool, a position detector that detects a position of the arm tip, and a robot controller that controls an operation of the robot based on a position command and a position feedback detected by the position detector. The robot controller includes an adjustment operation creating unit that, during adjustment of operation parameters for controlling the operation of the robot, acquires an application and an operation area of the robot and automatically creates an adjustment operation corresponding to the acquired application and the operation area and a parameter adjustment unit that automatically adjusts the operation parameters during execution of the adjustment operation created by the adjustment operation creating unit so that a performance required for the application is satisfied.

公开(公告)号：US10737384B2

公开(公告)日：2020-08-11

申请号：US16102058

申请日：2018-08-13

Applicant: FANUC CORPORATION

Inventor： Nobuaki Yamaoka ,  Hajime Suzuki ,  Kaimeng Wang

IPC: B25J9/16 ,  H04N5/225 ,  B25J19/02

Abstract: A robot system includes a light source, an image capture device, a robot mechanism unit having a target site of position control where the light source is provided, and a robot controller that controls the position of the robot mechanism unit based on a position command, a position feedback, and a position compensation value. The robot controller includes a path acquisition unit that makes the image capture device capture an image of light from the light source continuously during the predetermined operation to acquire a path of the light source from the image capture device, a positional error estimation unit that estimates positional error of the path of the light source from the position command based on the acquired path of the light source and the position command, and a compensation value generation unit that generates the position compensation value based on the estimated positional error.

公开(公告)号：US09937620B2

公开(公告)日：2018-04-10

申请号：US15281084

申请日：2016-09-30

Applicant: FANUC CORPORATION

Inventor： Nobuaki Yamaoka ,  Wenjie Chen ,  Kaimeng Wang

IPC: B25J9/00 ,  B25J9/16 ,  B25J13/08

CPC classification number: B25J9/1651 ,  B25J9/1692 ,  B25J13/088 ,  B25J13/089 ,  G05B2219/39017 ,  G05B2219/39058

Abstract: A robot control device which controls a robot includes a sensor coordinate system calculation unit which calculates a position and an orientation of a sensor by making the robot perform a predetermined operation. The sensor coordinate system calculation unit comprises an operation parameter optimization unit which is configured to obtain a combination most suitable for calculating the position and orientation of the sensor, from a plurality of combinations of modified values of a predetermined type of operation parameters, by making the robot perform the predetermined operation, successively using each of the combinations.

公开(公告)号：US20170067728A1

公开(公告)日：2017-03-09

申请号：US15259118

申请日：2016-09-08

Applicant: FANUC CORPORATION

Inventor： Takahiro Iwatake ,  Kaimeng Wang ,  Bunketsu Chin

IPC: G01B5/012

CPC classification number: G01B5/004 ,  G01B21/04

Abstract: A system which calculates the relative posture or the relative position and posture between a first object and a second object. This invention provides an object posture calculation system including a first object and a second object contactable at three contact points, a driving unit which brings the first object and the second object into contact with each other, a force measuring unit which measures a force acting between the first object and the second object, and an object posture calculation unit which calculates the relative posture or the relative position and posture between the first object and the second object on the basis of the force measured by the force measuring unit when the first object and the second object come into contact with each other at the three contact points.

Abstract translation: 一种计算第一对象和第二对象之间的相对姿势或相对位置和姿势的系统。 本发明提供一种物体姿势计算系统，包括：第一物体和第三物体，其在三个接触点处可接触;驱动单元，其使所述第一物体和所述第二物体彼此接触;力测量单元， 第一物体和第二物体，以及物体姿势计算单元，其基于当第一物体和第二物体的力测量单元测量的力时，计算第一物体和第二物体之间的相对姿势或相对位置和姿势 第二个物体在三个接触点处相互接触。

公开(公告)号：US12179350B2

公开(公告)日：2024-12-31

申请号：US17502636

申请日：2021-10-15

Applicant: FANUC CORPORATION

Inventor： Kaimeng Wang ,  Tetsuaki Kato

IPC: B25J9/00 ,  B25J9/16 ,  B25J19/02 ,  G06N3/02 ,  G06T5/70 ,  G06T7/13 ,  G06T7/70 ,  G06V40/10 ,  G06V40/20

Abstract: A method for dual arm robot teaching from dual hand detection in human demonstration. A camera image of the demonstrator's hands and workpieces is provided to a first neural network which determines the identity of the left and right hand from the image, and also provides cropped sub-images of the identified hands. The cropped sub-images are provided to a second neural network which detects the poses of both the left and right hand from the images. The dual hand pose data for an entire operation is converted to robot gripper pose data and used for teaching two robot arms to perform the operation on the workpieces, where each hand's motion is assigned to one robot arm. Edge detection from camera images may be used to refine robot motions in order to improve part localization for tasks requiring precision, such as inserting a part into an aperture.

公开(公告)号：US12030187B2

公开(公告)日：2024-07-09

申请号：US16503865

申请日：2019-07-05

Applicant: FANUC CORPORATION

Inventor： Kaimeng Wang ,  Wenjie Chen ,  Tetsuaki Kato

IPC: B25J9/16

CPC classification number: B25J9/1666 ,  B25J9/1653

Abstract: A robot system includes: a feature point position detection unit configured to detect, at a constant cycle, a position of a feature point of an obstacle that moves or deforms within a motion range of a robot; a movement path calculation unit configured to calculate a movement path of the robot before a motion of the robot; a mapping function derivation unit configured to derive a mapping function based on a position of the feature point that is detected at a time interval; and a path adjustment unit configured to dynamically adjust the movement path of the robot using the derived mapping function.