US-12619256-B2 - Method and apparatus for movable robot to adjust pose of goods rack

US12619256B2US 12619256 B2US12619256 B2US 12619256B2US-12619256-B2

Abstract

Disclosed are a method and apparatus for a movable robot to adjust the pose of a goods rack. A particular embodiment of the method comprises: acquiring a goods rack identifier image of a target goods rack; identifying a pose identifier of the target goods rack from the goods rack identifier image, and determining the current pose of the target goods rack based thereon; in response to the difference between the current pose of the target goods rack and a preset pose being greater than a preset deviation threshold value, determining that the target goods rack has shifted; generating a transport path on the basis of the difference between the current pose of the target goods rack and the preset pose; and sending the transport path to a movable robot, such that the movable robot adjusts the current pose of the target goods rack.

Inventors

Xiaoyan Fan
Jingren TANG

Assignees

Beijing Jingdong Qianshi Technology Co., Ltd.

Dates

Publication Date: 20260505
Application Date: 20220114
Priority Date: 20210129

Claims (18)

1 . A method for adjusting, by a movable robot, a current pose of a goods rack which has deviated from a preset pose, the method comprising: whilst a target goods rack is standing on a surface: acquiring a goods rack identifier image of the target goods rack, wherein the goods rack identifier image is an image obtained by photographing the goods rack identifier of the target goods rack by an image collection apparatus at a preset collection position with a preset collection attitude, the image collection apparatus is disposed on a vehicle body of the movable robot, and the goods rack identifier comprises a pose identifier of the target goods rack, wherein a center of a bottom surface of the target goods rack is provided with the goods rack identifier which is an imageable sign, wherein a projection of a center point of the goods rack identifier onto a horizontal plane coincides with a projection of a center point of the target goods rack onto the horizontal plane, wherein the pose identifier comprises: a length direction identifier of a goods rack, a width direction identifier of the goods rack, and a distance identifier, wherein the length direction identifier and the width direction identifier are elements of the imageable sign, and wherein the distance identifier comprises: a plurality of area identifiers evenly distributed within a region of the goods rack identifier, each area identifier represents a relative position of an area where the area identifier is located from the center point of the goods rack identifier; identifying the pose identifier of the target goods rack from the goods rack identifier image, and determining the current pose of the target goods rack based on the pose identifier of the target goods rack; determining, that the target goods rack is shifted, in response to determining that a difference value between the current pose of the target goods rack and the preset pose of the target goods rack is greater than a preset deviation threshold; generating, in response to determining that the target goods rack is shifted, a transport path based on the difference value between the current pose of the target goods rack and the preset pose of the target goods rack; controlling the movable robot by sending the transport path to the movable robot, wherein the transport path represents an instruction to cause the movable robot to adjust the current pose of the target goods rack to be consistent with the preset pose of the target goods rack; and then adjusting, by the movable robot, the current pose of the target goods rack to be consistent with the preset pose of the target goods rack according to the transport path; wherein the determining a current pose of the target goods rack based on the pose identifier of the target goods rack comprises: orientating the movable robot at the preset collection position into the preset collection attitude, by aligning a coordinate axis in an image coordinate system of the image collection apparatus with one of a first reference direction identifier and a second reference direction identifier which are provided by a preset collection position identifier on a ground surface: determining angles of a current length direction and a current width direction of the target goods rack relative to the image coordinate system of the image collection apparatus based on the length direction identifier and the width direction identifier of the target goods rack in the photographed goods rack identifier in the goods rack identifier image, to obtain a relative attitude of the target goods rack with respect to the image collection apparatus; identifying a relative position of the target goods rack relative to the image collection apparatus from the area identifiers in the goods rack identifier image, and determining a current position of the target goods rack based on the relative position of the target goods rack relative to the image collection apparatus and the preset collection position; and obtaining the current pose of the target goods rack based on the relative attitude and current position of the target goods rack.
2 . The method according to claim 1 , wherein the preset collection position is provided with the preset collection position identifier, the preset collection position identifier comprises the first reference direction identifier and the second reference direction identifier, and the method further comprises: acquiring an image of the preset collection position identifier; identifying the first reference direction identifier and the second reference direction identifier from the image of the preset collection position identifier; determining a direction indicated by the first reference direction identifier as a preset length direction of the target goods rack, and determining a direction indicated by the second reference direction identifier as a preset width direction of the target goods rack, to obtain a preset attitude of the target goods rack; determining a center point of the preset collection position as a preset position of the target goods rack; and determining the preset pose of the target goods rack based on the preset attitude and the preset position of the target goods rack.
3 . The method according to claim 1 , further comprising: sending alarm information, in response to determining that the goods rack identifier of the target goods rack does not exist in the goods rack identifier image.
4 . A non-transitory computer readable medium, storing a computer program, wherein the program, when executed by a processor, implements the method according to claim 1 .
5 . The method according to claim 1 , wherein the length direction indicator and the width direction indicator are each a directional sign which is displayed in the goods rack identifier which comprises a two-dimensional image.
6 . The method according to claim 1 , wherein the goods rack identifier comprises an image with grids arranged at intervals as the pose identifier, wherein a side length of each grid area is a calibration length, and an extension direction of the side length is parallel to the length direction or the width direction of the target goods rack.
7 . The method according to claim 1 , wherein a goods rack coordinate system is established with the center point of the goods rack identifier as an origin and the length direction and width direction of the target goods rack as coordinate axes, and a representation of the coordinate axes is displayed in the goods rack identifier.
8 . The method according to claim 7 , wherein the relative position information of each of the plurality of area identifiers in the goods rack identifier relative to the center point of the goods rack identifier is encoded in advance by embedding a coordinate value into each area identifier respectively.
9 . The method according to claim 1 , wherein a two-dimensional code image is used for each of the plurality of the area identifiers in the goods rack identifier, each two-dimensional code image is a square with equal side lengths, and an extension direction of the side length of the two-dimensional code image is parallel to the length direction or width direction of the target goods rack.
10 . An apparatus for adjusting, by a movable robot, a current pose of a goods rack which has deviated from a preset pose, the apparatus comprising: one or more processors; and a data storage device which stores one or more computer programs, wherein the one or more computer programs, when executed by the one or more processors, cause the one or more processors to implement operations, the operations comprising: whilst a target goods rack is standing on a surface: acquiring a goods rack identifier image of the target goods rack, wherein the goods rack identifier image is an image obtained by photographing the goods rack identifier of the target goods rack by an image collection apparatus at a preset collection position with a preset collection attitude, the image collection apparatus is disposed on a vehicle body of the movable robot, and the goods rack identifier comprises a pose identifier of the target goods rack, wherein a center of a bottom surface of the goods rack is provided with the goods rack identifier which is an imageable sign, wherein a projection of a center point of the goods rack identifier onto a horizontal plane coincides with a projection of a center point of the target goods rack onto the horizontal plane, wherein the pose identifier comprises: a length direction identifier of a goods rack, a width direction identifier of the goods rack, and a distance identifier, wherein the length direction identifier and the width direction identifier are elements of the imageable sign, and wherein the distance identifier comprises: a plurality of area identifiers evenly distributed within a region of the goods rack identifier, each area identifier represents a relative position of an area where the area identifier is located from the center point of the goods rack identifier; identifying the pose identifier of the target goods rack from the goods rack identifier image, and determining the current pose of the target goods rack based on the pose identifier of the target goods rack; determining, by a controller in communication with the image collection apparatus, that the target goods rack is shifted, in response to determining that a difference value between the current pose of the target goods rack and the preset pose is greater than a preset deviation threshold; generating, in response to determining that the target goods rack is shifted, a transport path based on the difference value between the current pose of the target goods rack and the preset pose of the target goods rack; controlling the movable robot by sending the transport path to the movable robot, wherein the transport path represents an instruction to cause the movable robot to adjust the current pose of the target goods rack to be consistent with the preset pose of the target goods rack; and then adjusting, by the movable robot, the current pose of the target goods rack to be consistent with the preset pose of the target goods rack according to the transport path; wherein the determining a current pose of the target goods rack based on the pose identifier of the target goods rack comprises: orientating the movable robot at the preset collection position into the preset collection attitude, by aligning a coordinate axis in an image coordinate system of the image collection apparatus with one of a first reference direction identifier and a second reference direction identifier which are provided by a preset collection position identifier on a ground surface; determining angles of a current length direction and a current width direction of the target goods rack relative to the image coordinate system of the image collection apparatus based on the length direction identifier and the width direction identifier of the target goods rack in the photographed goods rack identifier in the goods rack identifier image, to obtain a relative attitude of the target goods rack with respect to the image collection apparatus; identifying a relative position of the target goods rack relative to the image collection apparatus from the area identifiers in the goods rack identifier image, and determining a current position of the target goods rack based on the relative position of the target goods rack relative to the image collection apparatus and the preset collection position; and obtaining the current pose of the target goods rack based on the relative and current position of the target goods rack.
11 . The apparatus according to claim 10 , wherein the preset collection position is provided with the preset collection position identifier, the preset collection position identifier comprises the first reference direction identifier and the second reference direction identifier, and the operations further comprise: acquiring an image of the preset collection position identifier; identifying the first reference direction identifier and the second reference direction identifier from the image of the preset collection position identifier; determining a direction indicated by the first reference direction identifier as a preset length direction of the target goods rack, and determining a direction indicated by the second reference direction identifier as a preset width direction of the target goods rack to obtain a preset attitude of the target goods rack; determining a center point of the preset collection position identifier as a preset position of the target goods rack; and determining the preset pose of the target goods rack based on the preset attitude and the preset position of the target goods rack.
12 . The apparatus according to claim 11 , wherein the operations further comprise: sending alarm information in response to determining that the goods rack identifier of the target goods rack does not exist in the goods rack identifier image.
13 . A system for a warehouse comprising: a target goods rack, wherein a center of a bottom surface of the target goods rack is provided with a goods rack identifier which is an imageable sign, wherein a projection of a center point of the goods rack identifier onto a horizontal plane coincides with a projection of a center point of the target goods rack onto the horizontal plane, wherein the goods rack identifier comprises a pose identifier of a target goods rack, the pose identifier comprises: a length direction identifier of a goods rack, a width direction identifier of the goods rack, and a distance identifier, wherein the length direction identifier and the width direction identifier are elements of the imageable sign, and wherein the distance identifier comprises: a plurality of area identifiers evenly distributed within a region of the goods rack identifier, each area identifier represents a relative position of an area where the area identifier is located from the center point of the goods rack identifier; a movable robot for adjusting a current pose of the target goods rack, comprising a first image collection apparatus, a jacking-up apparatus for lifting the target goods rack off a ground surface and rotating the target goods rack, and a controller the dimensions of the movable robot are such that the movable robot can move under the target goods rack; wherein the first image collection apparatus is a camera which is fixedly disposed on a vehicle body of the movable robot, and is configured so that the field of view of the first image collection apparatus points vertically upward to collect, an image of the goods rack identifier provided on the bottom surface of the target goods rack; the jacking-up apparatus disposed on the movable robot comprises a vertical motion component and a rotary motion component, the vertical motion component is connected to the rotary motion component and is configured to cause the rotary motion component to move along a vertical direction to support and thereby lift the target goods rack from standing on the ground surface into a lifted position via a load-bearing surface of the rotary motion component and to lower the target goods rack from the lifted position to stand on the ground surface, and the rotary motion component is configured to cause the target goods rack to rotate around the vertical direction when in the lifted position; and the controller is in communication with the first image collection apparatus and the jacking-up apparatus, and is configured to perform operations, the operations comprising: whilst the target goods rack is standing on a surface: acquiring a goods rack identifier image of the target goods rack, wherein the goods rack identifier image is an image obtained by photographing the goods rack identifier of the target goods rack by the first image collection apparatus at a preset collection position with a preset collection attitude; identifying the pose identifier of the target goods rack from the goods rack identifier image, and determining the current pose of the target goods rack based on the pose identifier of the target goods rack; determining that the target goods rack is shifted, in response to determining that a difference value between the current pose of the target goods rack and a preset pose of the target goods rack is greater than a preset deviation threshold; generating, in response to determining that the target goods rack is shifted, a transport path based on the difference value between the current pose of the target goods rack and the preset pose of the target goods rack; controlling the movable robot by sending the transport path to the movable robot, wherein the transport path represents an instruction of a path along which the movable robot is to move to transport the target goods rack to adjust the current pose of the target goods rack to be consistent with the preset pose of the target goods rack; and then adjusting, by the movable robot, the current pose of the target goods rack to be consistent with the preset pose of the target goods rack according to the transport path; the determining a current pose of the target goods rack based on the pose identifier of the target goods rack comprises: orientating the movable robot at the preset collection position into the preset collection attitude, by aligning a coordinate axis in an image coordinate system of the first image collection apparatus with one of a first reference direction identifier and a second reference direction identifier which are provided by a preset collection position identifier on a ground surface: determining angles of a current length direction and a current width direction of the target goods rack relative to the image coordinate system of the first image collection apparatus based on the length direction identifier and the width direction identifier of the target goods rack in the photographed goods rack identifier in the goods rack identifier image, to obtain a relative attitude of the target goods rack with respect to the first image collection apparatus; identifying a relative position of the target goods rack relative to the first image collection apparatus from the area identifiers in the goods rack identifier image, and determining a current position of the target goods rack based on the relative position of the target goods rack relative to the first image collection apparatus and the preset collection position; and obtaining the current pose of the target goods rack based on the relative attitude and current position of the target goods rack.
14 . The system according to claim 13 , further comprising a second image collection apparatus which is a camera that is fixedly disposed on the vehicle body of the movable robot, wherein the second image collection apparatus is configured to collect by an image sensor an image of a preset collection position identifier provided on ground in a vertically downward direction, and the controller is further configured to identify the image of the preset collection position identifier, to control the movable robot to move to a preset collection position.
15 . The system according to claim 13 , wherein the rotary motion component comprises a horizontally disposed tray, a bottom surface of the tray is connected to the vertical motion component, a central area of the tray is provided with a vertical through-hole, a central axis of the vertical through-hole is collinear with a rotation axis of the tray, and the first image collection apparatus is located in an area of the through-hole of the tray projected onto a horizontal plane.
16 . The system according to claim 13 , further comprising: an automated guided vehicle, the automated guided vehicle is used to transport the goods rack, and the movable robot is configured to detect whether the goods rack is shifted, and adjust a pose of the shifted goods rack to a preset pose.
17 . The system according to claim 16 , comprising at least a first movable robot and a second movable robot, wherein the first movable robot and the second movable robot are configured to: the first movable robot and the second movable robot respectively detect whether the goods rack is shifted, and the first movable robot and the second movable robot simultaneously determine the shifted goods rack as a to-be-adjusted goods rack, the first movable robot is further configured to adjust a pose of the to-be-adjusted goods rack to a preset pose, and the second movable robot is further configured to detect whether the pose of the to-be-adjusted goods rack after the adjustment is shifted, and send alarm information in response to determining that the pose of the to-be-adjusted goods rack after the adjustment is shifted.
18 . The movable robot according to claim 13 or 14 , wherein the movable robot is an autonomous guided robot.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application is a national stage of International Application No. PCT/CN2022/072023, filed on Jan. 14, 2022, which claims the priority from Chinese Patent Application No. 202110134237.9, filed on Jan. 29, 2021. The contents of each of the aforementioned applications are hereby incorporated by reference in their entireties. TECHNICAL FIELD Embodiments of the present disclosure relate to the field of computer technology, specifically to the field of smart warehouse, and particularly to a method and apparatus for adjusting a goods rack pose by a movable robot. BACKGROUND OF THE INVENTION Movable robots are widely used in smart warehouses and usually used to transport goods racks. In the process of transporting the goods racks, the movable robots may come into contact with surrounding goods racks, making the surrounding goods racks shifted. In the related technology, the operating people usually observe the goods racks periodically, and manually correct the shifted goods racks. SUMMARY OF THE INVENTION Embodiments of the present disclosure propose a method and apparatus for adjusting a goods rack pose by a movable robot. In a first aspect, some embodiments of the present disclosure provide a method for adjusting a goods rack pose by a movable robot. The method comprises: acquiring a goods rack identifier image of a target goods rack, wherein the goods rack identifier image is an image obtained by photographing a goods rack identifier of the target goods rack by an image collection apparatus at a preset collection position with a preset collection attitude, and the goods rack identifier comprises a pose identifier of the target goods rack; identifying the pose identifier of the target goods rack from the goods rack identifier image, and determining a current pose of the target goods rack based on the pose identifier of the target goods rack; determining that the target goods rack is shifted, in response to determining that a difference value between the current pose of the target goods rack and a preset pose is greater than a preset deviation threshold; generating, in response to determining that the target goods rack is shifted, a transport path based on the difference value between the current pose of the target goods rack and the preset pose of the target goods rack; and sending the transport path to the movable robot, to cause the movable robot to adjust the current pose of the target goods rack to be consistent with the preset pose of the target goods rack according to the transport path. In a second aspect, some embodiments of the present disclosure provide an apparatus for adjusting a goods rack pose by a movable robot. The apparatus comprises: an image acquiring unit, configured to acquire a goods rack identifier image of a target goods rack, wherein the goods rack identifier image is an image obtained by photographing a goods rack identifier of the target goods rack by an image collection apparatus at a preset collection position with a preset collection attitude, and the goods rack identifier comprises a pose identifier of the target goods rack; an image identifying unit, configured to identify the pose identifier of the target goods rack from the goods rack identifier image, and determine a current pose of the target goods rack based on the pose identifier of the target goods rack; a pose determining unit, configured to determine that the target goods rack is shifted, in response to determining that a difference value between the current pose of the target goods rack and a preset pose is greater than a preset deviation threshold; a path generating unit, configured to generate, in response to determining that the target goods rack is shifted, a transport path based on the difference value between the current pose of the target goods rack and the preset pose of the target goods rack; and a path sending unit, configured to send the transport path to the movable robot, to cause the movable robot to adjust the current pose of the target goods rack to be consistent with the preset pose of the target goods rack according to the transport path. In a third aspect, some embodiments of the present disclosure provide a movable robot, comprising a first image collection apparatus, a jacking-up apparatus, and a controller, where the first image collection apparatus is fixedly disposed on a vehicle body of the movable robot, and is configured to collect, in a vertical upward direction, an image of a goods rack identifier provided on a bottom surface of a goods rack; the jacking-up apparatus comprises a vertical motion component and a rotary motion component, the vertical motion component is connected to the rotary motion component and is configured to cause the rotary motion component to move along a vertical direction, and the rotary motion component is configured to cause the goods rack to rotate around the vertical direction; and the controller is in communication with