KR-102963285-B1 - SENSOR-BASED CORRECTION OF ROBOT-HELD OBJECT

KR102963285B1KR 102963285 B1KR102963285 B1KR 102963285B1KR-102963285-B1

Abstract

A robot object handling system comprises a robot arm, a non-contact sensor, a first station, and a computing device. The computing device causes the robot arm to pick up an object on an end effector, causes the robot arm to position the object within the detection area of a non-contact sensor, causes the non-contact sensor to generate sensor data of the object, determines at least one of a rotation error of the object relative to a target orientation or a position error of the object relative to a target position based on the sensor data, induces adjustment of the robot arm to almost eliminate at least one of the rotation error or position error from the object, and causes the robot arm to place the object at the first station, wherein the placed object has no rotation error or position error.

Inventors

코펙, 니콜라스 마이클
콕스, 데이먼 케이.
볼포브스키, 레온

Assignees

어플라이드 머티어리얼스, 인코포레이티드

Dates

Publication Date: 20260508
Application Date: 20200624
Priority Date: 20190625

Claims (20)

As a method, A step of picking up an object on the end effector of a robot arm; A step of generating an image of the object on the end effector using an image sensor while the object is being held on the end effector of the robot arm; (i) determining at least one of a rotational error of the object based on the orientation of a registration feature reflected by an image of the object, or (ii) a positional error of the object based on the position of a registration feature reflected by an image of the object; A step of adjusting the robot arm to approximately eliminate at least one of the rotation error or the position error; and A step comprising placing the object at the first station using the robot arm without at least one of the rotation error or the position error. method.
In Article 1, The robot arm includes one or more joints that enable rotation of the end effector up to a critical amount around the end effector axis, and The above method further includes the step of determining an angle correction for the object to almost eliminate the rotation error of the object, and The step of adjusting the robot arm to almost eliminate the rotational error of the object includes the step of rotating the end effector of the robot arm to achieve the angle correction, and the step of positioning is performed using the end effector rotated to achieve the angle correction. method.
In Article 2, The above angle correction further includes a step of determining that the angle correction is less than or equal to a rotation threshold amount, method.
In Article 1, The robot arm includes one or more joints that enable rotation of the end effector up to a critical amount around the end effector axis, and The above method is, A step of determining an angle correction for the object to almost eliminate the rotation error of the object; and The method further includes a step of determining that the above angle correction exceeds a rotation threshold amount, The step of adjusting the robot arm to almost eliminate the rotational error from the object comprises: a) until the angle correction for the object is achieved, or b) until the residual angle error is less than the rotational threshold of the end effector. A step of placing the above object at the second station; A step of repositioning the above end effector; and A step of picking up the object from the second station using the repositioned end effector. It includes a step of performing at least once, The above object has a smaller rotational error after being picked up from the second station, method.
In Paragraph 4, The above object has a smaller position error after being picked up from the second station, method.
In Article 1, The robot arm includes one or more joints that enable rotation of the end effector up to a critical amount around the end effector axis, and the robot arm can correct a position error up to a critical amount without using a second station, The above method is, A step of determining an angle correction for the object to almost eliminate the rotation error of the object; A step of determining a position correction to almost eliminate the position error of the above object; and a) determining at least one of the angle correction exceeding a rotation threshold or b) the position error exceeding a position error threshold, and further including the step of determining at least one of the above angle correction exceeding a rotation threshold or b) the position error exceeding a position error threshold. The step of adjusting the robot arm to almost eliminate at least one of the rotation error or the position error from the object is, A step of placing the above object at the second station; A step of repositioning the above end effector; and A step of picking up the object from the second station using the repositioned end effector. It includes a step of performing at least once, The above object has at least one of a smaller rotational error or a smaller positional error after being picked up from the second station, method.
In Article 1, The above robot arm does not have the ability to rotate the end effector around the end effector axis, and The above method further includes the step of determining an angle correction for the object to almost eliminate the rotation error of the object, and The step of adjusting the robot arm to almost eliminate the rotational error from the object is until the angle correction for the object is achieved. A step of placing the above object at the second station; A step of repositioning the above end effector; and A step of picking up the object from the second station using the repositioned end effector. It includes a step of performing at least once, The above object has a smaller rotation error, method.
In Article 7, The above object has a smaller position error after being picked up from the second station, method.
In Article 1, The step of generating an image of the object includes the step of generating one or more images using the image sensor, and the step of determining the rotation error of the object includes the step of performing image processing on the one or more images to identify at least one orientation among flat, notch, or fiducial of the object. method.
In Article 1, The above object is attached to a carrier, the step of picking up the object includes the step of picking up the carrier attached to the object, and the step of placing the object includes the step of placing the carrier attached to the object. The above method is, A step of generating sensor data of the carrier using the image sensor; and A method further comprising the step of determining a rotation error of the carrier based on the sensor data of the carrier. method.
In Article 10, A step of comparing the rotation error of the object and the rotation error of the carrier; A step of determining the difference between the rotation error of the object and the rotation error of the carrier based on the above comparison step; A step of determining whether the above difference exceeds a threshold; and In response to determining that the above difference does not exceed the above threshold, the method further comprises the step of placing the object at the first station. method.
In Article 1, A step of picking up a carrier attached to a second object on the end effector of the robot arm from a first position; A step of generating new sensor data of the object and the carrier using the image sensor while the carrier is held on the end effector of the robot arm; A step of determining the orientation of the carrier based on the new sensor data above; A step of determining the orientation of the second object based on the new sensor data; A step of determining the difference between the orientation of the object and the orientation of the carrier; A step of determining whether the above difference exceeds a threshold; and In response to determining that the above difference exceeds the above threshold, the method further comprises the step of placing the carrier and the object back at the first position. method.
In Article 1, The above object includes consumable parts for a processing chamber, method.
In Article 1, The image sensor above includes a laser emitter that generates a laser beam and a laser receiver that receives the laser beam, and The step of generating an image of the above object is, Repeatedly (a) extending the end effector until the object is between the laser emitter and the laser receiver and interrupts the laser beam so that the laser beam is no longer received by the laser receiver, and (b) recording one or more parameters of the robot arm when the object interrupts the laser beam — the one or more parameters include the rotation of the end effector and the position of the end effector —; A step of generating an array of measurements — each measurement corresponds to a different rotation of the end effector —; and A method comprising the step of determining the rotation angle of the object based on the plurality of measurements above. method.
In Article 1, After the step of adjusting the robot arm and before the step of placing the object at the first station, A step of generating an additional image of the object using the image sensor; A step of determining at least one of the residual rotation error of the object or the residual position error of the object based on the above image; and Step of further adjusting the robot arm to almost eliminate at least one of the residual rotation error or the residual position error from the object including additional steps to perform, method.
As a robot object handling system, Robot arm including an end effector; Image sensor; 1st station; and It includes the image sensor and a computing device operably coupled to the robot arm, The above computing device is, The robot arm is made to pick up an object on the end effector; While the object is held on the end effector of the robot arm, the image sensor is made to generate an image of the object on the end effector; (i) determining at least one of the rotation error of the object based on the orientation of the matching feature reflected by the image of the object or (ii) the position error of the object based on the position of the matching feature reflected by the image of the object; Inducing adjustment to the robot arm to almost eliminate at least one of the rotation error or the position error; and The robot arm is configured to place the object at the first station without at least one of the rotation error or the position error. Robot object handling system.
In Article 16, The robot arm includes one or more joints that enable rotation of the end effector up to a critical amount around the end effector axis, and The computing device further determines an angle correction for the object to almost eliminate the rotation error of the object, and Inducing adjustment to the robot arm to almost eliminate the rotational error from the object comprises rotating the end effector of the robot arm to achieve the angle correction, and the positioning is performed using the end effector rotated to achieve the angle correction. Robot object handling system.
In Article 17, The computing device further determines that the angle correction is less than or equal to a rotation threshold amount. Robot object handling system.
In Article 16, The robot arm includes one or more joints that enable rotation of the end effector up to a critical amount around the end effector axis, and The above computing device additionally, Determining an angle correction for the object to almost eliminate the rotation error of the object; and It is determined that the above angle correction exceeds the rotation threshold, Inducing adjustment of the robot arm to almost eliminate the rotational error from the object is a) until the angle correction for the object is achieved or b) until the residual angle error is less than the rotation threshold of the end effector. Placing the above object at the second station — the above object has an initial position and orientation on the second station —; Repositioning the above end effector; and Picking up the object from the second station using the repositioned end effector Includes performing at least once, The above object has a smaller rotational error after being picked up from the second station, Robot object handling system.
In Article 16, Generating an image of the object comprises generating one or more images using the image sensor, and determining the rotation error of the object comprises performing image processing on the one or more images to identify at least one orientation among flat, notch, or origin of the object. Robot object handling system.

Description

Sensor-based correction of robot-held object [0001] Embodiments of the present disclosure relate to the detection and correction of misalignment (e.g., rotational misalignment) of objects picked up by a robot arm. [0002] In semiconductor processing and other electronic processing, platforms using robotic arms are commonly used to transport objects, such as wafers, between process chambers, from storage areas (e.g., FOUPs (front opening unified pods)) to process chambers and from process chambers to storage areas. In many cases, objects are out of alignment when they are picked up by the end effector of the robotic arm. Objects may contain rotational errors and/or positional errors when they are picked up by the end effector. Typically, such misalignment is corrected by placing the objects on an alignment station that rotates the objects until they are in correct alignment. Then, the robotic arm picks the aligned objects up again from the alignment station before moving them to their destination. The use of an alignment station to correct the alignment of objects increases the amount of time used to transport the objects between the starting position and the destination, adds additional handoffs (between the alignment station and the robotic arm), which can introduce additional errors. Additionally, the alignment station consumes real estate on the platform and the fabrication facility where it is located, and adds additional costs to the platform. Furthermore, for some objects, the end effector, adapter for the end effector, and/or the design of the object itself is made to accommodate the alignment station. However, by designing these components to accommodate the alignment station, the end effector, adapter, and/or object design may be less effective for other purposes. [0003] Some of the embodiments described herein cover a method for aligning an object. The method may include the step of picking up an object on an end effector of a robot arm. Subsequently, the object may be positioned within the detection area of a non-contact sensor using the robot arm. While the object is being held on the end effector of the robot arm, sensor data of the object may be generated using the non-contact sensor. Subsequently, based on the sensor data, at least one of the rotational error of the object relative to the target orientation or the positional error of the object relative to the target position may be determined. The robot arm may be adjusted to approximately eliminate at least one of the rotational error or positional error from the object. Subsequently, the object may be placed at a first station using the robot arm, and the placed object lacks at least one of the rotational error or positional error. In some embodiments, the robot arm comprises one or more joints that enable rotation of the end effector around the end effector axis up to a threshold amount, and the method further comprises the step of determining an angle correction for the object to eliminate rotational error of the object, the step of adjusting the robot arm to eliminate rotational error of the object comprises the step of rotating the end effector of the robot arm to achieve angle correction, and the step of positioning is performed using the end effector rotated to achieve angle correction. [0004] In some embodiments, a robot object handling system comprises a robot arm including an end effector, a non-contact sensor having a detection area within the reach of the robot arm, a first station within the reach of the robot arm, and a computing device operably coupled to the non-contact sensor and the robot arm. The computing device may execute commands to cause the robot arm to pick up an object on the end effector. The computing device may additionally cause the robot arm to position the object within the detection area of the non-contact sensor. The computing device may additionally cause the non-contact sensor to generate sensor data of the object while the object is held on the end effector of the robot arm. The computing device may additionally determine at least one of the rotation error of the object relative to the target orientation or the position error of the object relative to the target position based on the sensor data. The computing device may additionally cause adjustment to the robot arm to almost eliminate at least one of the rotation error or position error from the object. The computing device can additionally enable the robot arm to place an object at the first station, and the placed object has no rotation error or at least one position error. [0005] In some embodiments, the robot handling system comprises a robot arm including an end effector, a first station within the reach of the robot arm, a second station within the reach of the robot arm, a non-contact sensor having a detection area in the first station, and a computing device operably coupled to the non-contact sensor and the robot arm. The computing device may