A robot includes a grasping unit and performs an action based on: first imaging information of the grasping unit which does not grasp an object to be grasped in a first point; second imaging information of the grasping unit which does not grasp the object to be grasped in a second point which is dif
A robot includes a grasping unit and performs an action based on: first imaging information of the grasping unit which does not grasp an object to be grasped in a first point; second imaging information of the grasping unit which does not grasp the object to be grasped in a second point which is different from the first point; and third imaging information of the object to be grasped which is grasped by the grasping unit in the first point.
대표청구항▼
1. A robot comprising: a hand that is configured to grasp an object;a camera that is configured to take first, second, and third actual images of the hand, the camera being configured to output first, second, and third actual image data corresponding to the first, second, and third images, the camer
1. A robot comprising: a hand that is configured to grasp an object;a camera that is configured to take first, second, and third actual images of the hand, the camera being configured to output first, second, and third actual image data corresponding to the first, second, and third images, the camera being configured to: take the first actual image of the hand without grasping the object when the hand is located at a first actual position;take the second actual image of the hand without grasping the object when the hand is located at a second actual position that is different from the first position; andtake the second actual image of the hand while grasping the object when the hand is located at the first actual position;a memory that stores: first ideal image data corresponding to a first state in which the hand without grasping the object is located at a first ideal position;second ideal image data corresponding to a second state in which the hand without grasping the object is located at a second ideal position; andthird ideal image data corresponding to a third state in which the hand while grasping the object is located at the first position; anda processor that is configured to: correct a position of the hand by comparing the first actual image data with the first ideal data when the hand while grasping the object is located at the first actual position so as to move the hand to a first corrected position, and correct a relative location of the hand and the object by comparing the third actual image data with the third ideal image data;move the hand from the first corrected position to the second position; andcorrect a position of the hand by comparing the second actual image data with the second ideal image data when the hand while grasping the object is located at the second actual position so as to move the hand to a second corrected position. 2. The robot according to claim 1, wherein the first and second actual image data correspond to position and posture information of the hand, andthe third actual image data and the first, second, and third ideal image data correspond to position and posture of the hand and object. 3. The robot according to claim 1, wherein the hand includes a plurality of claws which extend from a base of the hand, each of the plurality of claws including a distal portion which grasps the object, and each of the plurality of claws including a proximal portion which is positioned closer to the base of the hand than the distal portion, andthe camera is configured to take the first, second, and third actual images when the distal portion is located closer to the camera than the proximal portion. 4. The robot according to claim 3, wherein a plurality of the cameras are provided. 5. The robot according to claim 3, wherein the camera is a stereo camera. 6. The robot according to claim 1, wherein the second actual image includes a marker provided on the hand. 7. A control method of a robot, the robot includes: a hand that is configured to grasp an object;a camera that is configured to take first, second, and third actual images of the hand, the camera being configured to output first, second, and third actual image data corresponding to the first, second, and third images, the camera being configured to: take the first actual images of the hand without grasping the object when the hand is located at a first actual position;take the second actual image of the hand without grasping the object when the hand is located at a second actual position that is different from the first position; andtake the second actual image of the hand while grasping the object when the hand is located at the first actual position;a memory that stores: first ideal image data corresponding to a first state in which the hand without grasping the object is located at a first ideal position;second ideal image data corresponding to a second state in which the hand without grasping the object is located at a second ideal position; andthird ideal image data corresponding to a third state in which the hand while grasping the object is located at the first position; anda processor, the control method comprising:correcting a position of the hand by comparing the first actual image data with the first ideal image data by the processor when the hand while grasping the object is located at the first actual position so as to move the hand to a first corrected position by the processor, and correcting a relative location of the hand and the object by comparing the third actual image data with the third ideal image data by the processor;moving the hand from the first corrected position to the second position by the processor; andcorrecting a position of the hand by comparing the second actual image data with the second ideal image data by the processor when the hand while grasping the object is located at the second actual position so as to move the hand to a second corrected position by the processor. 8. A control device of a robot which controls the robot, the robot includes: a hand that is configured to grasp an object; anda camera that is configured to take first, second, and third actual images of the hand, the camera being configured to output first, second, and third actual image data corresponding to the first, second, and third images, the camera being configured to: take the first actual image of the hand without grasping the object when the hand is located at a first actual position;take the second actual image of the hand without grasping the object when the hand is located at a second actual position that is different from the first position; andtake the second actual image of the hand while grasping the object when the hand is located at the first actual position, the control device comprising:a memory that stores: first ideal image data corresponding to a first state in which the hand without grasping the object is located at a first ideal position;second ideal image data corresponding to a second state in which the hand without grasping the object is located at a second ideal position; andthird ideal image data corresponding to a third state in which the hand while grasping the object is located at the first position; anda processor, the control method comprising: correcting a position of the hand by comparing the first actual image data with the first ideal image data by the processor when the hand while grasping the object is located at the first actual position so as to move the hand to a first corrected position by the processor, and correcting a relative location of the hand and the object by comparing the third actual image data with the third ideal image data by the processor;moving the hand from the first corrected position to the second position by the processor; andcorrecting a position of the hand by comparing the second actual image data with the second ideal image data by the processor when the hand while grasping the object is located at the second actual position so as to move the hand to a second corrected position by the processor.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (7)
Valpola, Harri; Lukka, Tuomas, Method for the selection of physical objects in a robot system.
Kirsch Jerry (36 Beacon Hill Grosse Pointe Farms MI 48236) Kirsch Kerry F. (36 Beacon Hill Grosse Pointe Farms MI 48236), Rotary video article centering, orienting and transfer device for computerized electronic operating systems.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.