WO2020211552A1

WO2020211552A1 - Commodity recognition and grabbing method and device, and storage medium

Info

Publication number: WO2020211552A1
Application number: PCT/CN2020/077452
Authority: WO
Inventors: 邓耀桓; 张�浩; 雒曹盼
Original assignee: 深圳蓝胖子机器人有限公司
Priority date: 2019-04-19
Filing date: 2020-03-02
Publication date: 2020-10-22
Also published as: CN109956251A

Abstract

Disclosed are a commodity recognition and grabbing method and device, and a storage medium. The commodity recognition and grabbing method is applied to the commodity recognition and grabbing device, and the method comprises: a sowing robot recognizing a commodity by means of a commodity recognition device to acquire information, such as the shape, size, position and orientation, of the commodity; and according to the recognized information, generating a grabbing posture, such that a mechanical arm reaches a pre-grabbing position to grab the commodity. By using the commodity recognition and grabbing method, intelligently recognizing and automatically grabbing the commodity by means of the sowing robot can efficiently and accurately realize sorting of commodities and greatly reduce the error rate in a commodity picking process, and can greatly improve working efficiency.

Description

Method, equipment and storage medium for identifying and grabbing goods

Technical field

The invention relates to the technical field of warehouse picking operations, in particular to a method, device and storage medium for identifying and grabbing goods.

Background technique

The prior art picking goods in warehouses is generally done manually. The specific process is: the warehouse management system generates an order. The order contains information such as shelf number, product model and weight. The picker walks or drives a forklift according to the information on the order. Pick up the goods on the designated shelf. When picking goods, two methods of picking fruit and seeding are widely used. Fruit picking refers to each order. Pickers or equipment patrol each shelf to take out the required goods. The characteristic is Only one order is processed at a time, and fruit picking is performed. Seeding refers to the re-sorting of goods mixed with multiple order items to the corresponding personal orders.

When there are a wide variety of goods and the number of orders is huge, the seed picking method can improve work efficiency, shorten the walking distance during picking, and increase the number of picks per unit time.

However, using purely manual seeding methods is prone to human error, and work efficiency varies from person to person.

Summary of the invention

In view of this, the purpose of the present invention is to provide a method, device and storage medium for identifying and grabbing goods, which can greatly improve work efficiency and reduce error rates.

The technical solutions adopted by the present invention to solve the above technical problems are as follows:

According to the first aspect of the present invention, there is provided a cargo identification and grabbing method, which is suitable for cargo identification and grabbing equipment, and the method includes the following steps:

The seeding robot recognizes the goods through the goods recognition device, and obtains the shape, size, position, orientation and other information of the goods;

According to the recognized information, a grasping posture is generated, and the robotic arm reaches a pre-grabbing position to grasp the goods.

According to a second aspect of the present invention, there is provided a device for identifying and grabbing goods, the device comprising: a memory, a processor, and a computer program stored in the memory and running on the processor, the computer When the program is executed by the processor, the steps described in the first aspect are implemented.

According to a third aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored on the computer-readable storage medium, and when the computer program is executed by the processor, the A step of.

The goods identification and grabbing method, equipment and storage medium of the embodiments of the present invention can intelligently identify and automatically grab the goods through the seeding robot, which can efficiently and accurately realize the sorting of goods, which greatly reduces the amount of time in the picking process. Error rate, and can greatly improve work efficiency.

Description of the drawings

Figure 1 is a schematic diagram of the overall working area of the warehouse picking system implementing the present invention;

2A is a schematic diagram of the robot processing area of the warehouse picking system of the present invention;

Figure 2B is a real view of the robot processing area that implements the warehouse picking system of the present invention;

Figure 3 is a schematic diagram of the position of the goods in the seed box;

4 is a flowchart of a method for identifying, grabbing, and picking goods according to an embodiment of the present invention;

Fig. 5 is a schematic diagram of a cargo identification grabbing device provided by an embodiment of the present invention.

The realization of the objectives, functional characteristics and advantages of the present invention will be further described in conjunction with the embodiments and with reference to the accompanying drawings.

detailed description

It should be understood that the specific embodiments described here are only used to explain the present invention, but not to limit the present invention.

In the following description, if a suffix such as “module”, “part” or “unit” used to indicate an element is used only to facilitate the description of the present invention, it has no specific meaning in itself. Therefore, "module", "part" or "unit" can be used in a mixed manner.

Please refer to Figure 1. Figure 1 is a schematic diagram of the overall working area of the warehouse picking system implementing the present invention.

As shown in Figure 1, the overall work of the warehouse picking system is divided into a robot processing area 10 and a manual processing area 11.

The robot processing area 10 includes a seed box 101 to be processed, a conveyor belt 102, a robot console 103, a seeding robot 104, and a seeding wall 105.

The seeding box 101 to be processed is used to place goods to be seeded.

The conveyor belt 102 conveys the seed box 101 to the robot operating table 103.

The robot console 103 is provided with a code scanning area on the shelf, and a barcode scanner or a camera is installed in the scanning area.

The seeding wall 105 includes a plurality of seeding frames. It is placed on both sides of the seeding robot 104.

The planting robot 104 recognizes the goods in the planting box 101 on the console 103 through the goods identification device, and obtains the goods information, including the shape, size, position and orientation of the goods in the planting box, and so on. The goods identification device may be a visual identification device, a three-dimensional scanner, a lidar, and so on.

The seeding robot 104 controls the mechanical arm to reach a pre-grabbing position according to the identified goods information, grabs the goods, and moves the grabbed goods to the code scanning device area to scan the codes. Warehouse goods are provided with corresponding barcodes.

The planting robot 104 puts the grabbed goods to be planted into the corresponding planting frame on the planting wall 105 according to the information obtained by scanning the code.

Adjacent to the robot processing area 10 is a manual processing area 11. The manual processing area 11 is used to manually seed the goods that the robot processing area 10 cannot process.

The manual processing area 11 includes a seeding box 111 to be processed, a manual operation table 112, a seeding wall 113, and the like.

The to-be-processed seed box 111 is used to place goods that the robot processing area 10 cannot process.

The seeding box 111 is placed on the manual operating table 112.

The seeding wall 113 includes a plurality of seeding frames, which are placed on both sides of the manual operating table 112.

When the manual console 112 receives the seeding box 111 with the goods that the robot processing area 10 cannot process, the picker will seed the goods that the robot cannot process and place them in the corresponding seeding frame of the seeding wall 113.

When the sowing is over, the sowing wall containing the sowed goods is transported to the packaging area for packaging.

The intelligent recognition and automatic grasping of the goods by the seeding robot can efficiently and accurately realize the sorting of the goods, which greatly reduces the error rate in the picking process, and can greatly improve the work efficiency.

Please refer to FIG. 2A and FIG. 2B, which are respectively a schematic diagram of a robot processing area and a real view of the robot processing area of the warehouse picking system of the present invention.

As shown in Figure 2, the robot processing area of the warehouse picking system includes a seed box to be processed 201, a conveyor belt 202, a container buffer 203, a robot console 204, a seeding robot 205, a seeding wall card 206, a seeding wall 207, and a seeding wall Storage area 208, protective fence 209, etc.

The seeding box 201 to be processed is used to place goods to be seeded.

In practical applications, warehouse management systems (WMS, warehouse management systems) send instructions to obtain the order goods to the goods-seeking robot. According to the order information in the instruction, the goods-seeking robot removes all the goods contained in all orders from the shelves and puts them into the seeding box 201. Then an automated guided vehicle (AGV, Automated Guided Vehicle) transports the seed box 201 containing the goods to be seeded to the conveyor belt 202 in the robot processing area.

The conveyor belt 202 conveys the seed box 201 to the robot operating table 204.

The container buffer zone 203 is used to store the seed box 201 to be transported and sorted.

The robot operating table 204 is provided with a code scanning area on the shelf, and a barcode scanner or a camera is installed in the code scanning area.

The seeding wall clamping positions 206 are located on both sides of the robot operating table 204 and the seeding robot 205, and are used to install a fixed seeding wall 207. In practical applications, the seeding wall clamping position 206 may be in the form of an electromagnet clamping position.

The seeding wall 207 is composed of multiple seeding frames.

The seeding wall storage area 208 is used to temporarily store empty seeding walls that have not yet been seeded. Before the robot sows, move and install the corresponding seeding wall 207 to the seeding wall stop 206.

Because the robot processing area is connected to the manual processing area, in order to protect the personal safety of workers, a protective device can also be installed in the robot processing area, for example, a protective fence 209 may be set around the periphery.

The planting robot 205 recognizes the goods to be planted in the planting box 201 on the robot console 204 through the identification device, and obtains the goods information including the shape, size, position and orientation of the goods in the planting box and so on. The identification device may be a visual device, an infrared scanning device, and so on.

The seeding robot 205 controls the mechanical arm to reach a pre-grabbing position according to the identified goods information, grabs the goods, and moves the grabbed goods to the code scanning area for code scanning.

The planting robot 205 puts the grabbed goods to be planted into the corresponding planting frame on the planting wall 207 according to the information obtained by scanning the code.

Wherein, the seeding robot 205 grabs the goods, including:

According to the cargo information obtained by the vision device, analyze the surface of the cargo to be grabbed, the position and orientation information of the seeding box 201;

According to the size of the surface of the grasping actuator used by the robot arm of the seeding robot 205, at least one preliminary grasping posture is generated in accordance with the grasped surface of the goods;

Sort the preliminary grasping postures, the closer the grasping posture is to the geometric center point of the grasping surface, the more priority it is used;

Combining the orientation of the goods and the position of the goods in the seeding box, generate at least one final grab posture on the sorted preliminary grab posture;

According to the final grasping posture, the robot arm of the seeding robot 205 grasps the goods.

Further, in combination with the orientation of the goods and the position of the goods in the seeding box 201, at least one final gripping posture is generated from the sorted preliminary gripping posture, the specific method is:

If the angle between the bottom of the cargo and the bottom of the seed box is less than the preset threshold, the sorted initial grasping posture will be used without any change;

If the angle between the bottom of the cargo and the bottom of the sowing box is greater than the preset threshold, and the cargo is close to the right wall of the sowing box 201, the maximum deflection to the left is α degrees according to the initial grasping posture, and the resolution of each deflection is β degrees. , Where the maximum deflection angle α is related to the elastic length and radius of the gripping actuator used by the robotic arm, and the size of β can be preset;

If the angle between the bottom of the cargo and the bottom of the sowing box is greater than the preset threshold, and the cargo is close to the left wall of the sowing box 201, the maximum deflection to the right is α degrees according to the initial grasping attitude, and the resolution of each deflection is β degrees. , Where the maximum deflection angle α is related to the elastic length and radius of the gripping actuator used by the robotic arm.

In order to enable the robot arm to grab the surface of the goods in the largest area, when the goods are tilted to a certain extent in the seeding box, the robot arm is controlled to deflect in a specific direction, so that the robot arm can better fit the surface of the grabbed goods. So as to achieve a firm and powerful grip.

The threshold can be preset, which can be 20 degrees, 30 degrees, 40 degrees, etc. acute angles. The size of α is determined by the structure of the mechanical arm and is related to the elastic length and radius of the gripping actuator used.

As shown in Fig. 3, the cargo 32 is placed near the left wall of the sowing box 31, and the angle θ between the cargo 32 and the bottom of the sowing box 31 is included. Assume that the preset included angle threshold is 30 degrees. The angle θ between the cargo 32 and the bottom of the seeding box 31 is 40 degrees, and the planting robot manipulator is deflected to the right by a degree.

Further, according to the information obtained by scanning the code, put the goods to be planted into the planting frame of the corresponding planting wall 207, including:

Scan the barcode information of the goods to be planted, query the warehouse management system WMS, and confirm the designated position in the planting box corresponding to the goods to be planted;

The planting robot 205 moves the mechanical arm to the top of the corresponding planting frame, and puts down the goods at the corresponding position.

The intelligent recognition and automatic grabbing of the goods by the seeding robot can efficiently and accurately realize the sorting of the goods, which greatly reduces the error rate in the picking process, and can greatly improve work efficiency.

Please continue to refer to FIG. 4, an embodiment of the present invention provides a method for identifying and capturing goods. The cargo identification and capture method of this embodiment is suitable for cargo identification and capture equipment. The method flow includes:

Step S401: The seeding robot recognizes the goods through the goods recognition device, and obtains information such as the shape, size, position, and orientation of the goods;

Step S402: Generate a grasping posture according to the recognized information, and the mechanical arm reaches a pre-grabbing position to grasp the goods.

The goods identification device is a visual identification device, a three-dimensional scanner, or a lidar or the like.

In a feasible solution, step S402, generating a grasping gesture according to the recognized information, specifically includes:

According to the obtained cargo information, analyze the surface of the cargo to be grabbed, the position and orientation of the seed box;

According to the surface size of the gripping actuator used by the robotic arm, at least one preliminary gripping posture is generated to fit the surface of the goods being gripped.

Further, according to the recognized information, a grasping posture is generated, and the seeding robot manipulator reaches a pre-grabbing position to grasp the goods, which also includes:

Sort the preliminary grasping postures, the closer the grasping posture is to the geometric center of the grasping surface, the more priority it will be used;

According to the final grasping posture, the robot arm of the seeding robot grasps the goods.

In a feasible solution, combining the orientation of the goods and the position of the goods in the seeding box, generate at least one final gripping posture from the sorted preliminary gripping postures. The specific methods that can be taken are as follows:

If the angle between the direction of the cargo and the ground is less than the angle θ, the sorted initial grasping posture will be used without any change;

If the angle between the direction of the goods and the ground is greater than the angle θ, and the goods are close to the right wall of the sowing box, according to the initial grasping posture, the maximum deflection to the left is α degrees, and the resolution of each deflection is β degrees, where the maximum deflection angle α Related to the elastic length and radius of the gripping actuator used by the robotic arm;

If the angle between the direction of the goods and the ground is greater than the angle θ, and the goods are close to the left wall of the seeding box, according to the initial grasping posture, the maximum deflection to the right is α degrees, and the resolution of each deflection is β degrees, where the maximum deflection angle α It is related to the elastic length and radius of the gripping actuator used by the robotic arm.

The threshold can be preset, which can be 20 degrees, 30 degrees, 40 degrees, etc. acute angles.

In addition, at a certain distance perpendicular to the grasping surface, the pre-grabbing posture can be set, from the pre-grabbing position to the pre-grabbing posture, and from the pre-grabbing posture to the grasping point. Crawl paths.

The goods identification and grabbing method of this embodiment uses robots to identify and grab the goods, and put them into the corresponding seeding frame, which can accurately and quickly sort the goods, which greatly reduces the error rate during the seeding process. , And can greatly improve work efficiency.

On the basis of the foregoing embodiment, the embodiment of the present invention also provides a cargo identification and grabbing device.

Referring to FIG. 6, the goods identification and grabbing device includes: a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302, and the computer program 303 is processed by the When executed, the device 302 implements the steps of identifying and capturing goods as in the first embodiment or the second embodiment.

The goods identification and grabbing equipment of this embodiment can efficiently and accurately realize the sorting of goods by intelligently identifying and grabbing the goods and accurately sorting the goods, greatly reducing the error rate in the picking process, and can greatly Provide work efficiency.

On the basis of the foregoing embodiment, another embodiment of the present invention also provides a computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and the computer program is executed by a processor When realizing the steps of identifying and capturing goods in the foregoing embodiment.

The computer-readable storage medium of this embodiment can accurately and quickly sort the goods by intelligently identifying and accurately grasping the goods and placing them in the corresponding seeding frame, which greatly reduces errors in the seeding process. Rate, and can greatly improve work efficiency.

It should be noted that in this article, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, It also includes other elements not explicitly listed, or elements inherent to the process, method, article, or device. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other identical elements in the process, method, article or device that includes the element.

The sequence numbers of the foregoing embodiments of the present invention are only for description, and do not represent the superiority of the embodiments.

Through the description of the above embodiments, those skilled in the art can clearly understand that the method of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better.的实施方式。 Based on this understanding, the technical solution of the present invention essentially or the part that contributes to the existing technology can be embodied in the form of a software product, and the computer software product is stored in a storage medium (such as ROM/RAM, magnetic disk, The optical disc) includes several instructions to enable a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the method described in each embodiment of the present invention.

The embodiments of the present invention are described above with reference to the accompanying drawings, but the present invention is not limited to the above-mentioned specific embodiments. The above-mentioned specific embodiments are only illustrative and not restrictive. Those of ordinary skill in the art are Under the enlightenment of the present invention, many forms can be made without departing from the purpose of the present invention and the protection scope of the claims, and these all fall within the protection of the present invention.

Claims

A method for identifying and grabbing goods, which is suitable for goods identifying and grabbing equipment, characterized in that the method includes the following steps:

The seeding robot recognizes the goods through the goods recognition device, and obtains the shape, size, position, orientation and other information of the goods;

According to the recognized information, a grasping posture is generated, and the robotic arm reaches a pre-grabbing position to grasp the goods.
The cargo identification and grab method according to claim 1, wherein generating a grab gesture according to the identified information comprises:

According to the obtained cargo information, analyze the surface of the cargo to be grabbed, the position and orientation of the seed box;

According to the surface size of the gripping actuator used by the robotic arm, at least one preliminary gripping posture is generated to fit the surface of the goods being gripped.
The cargo identification and grab method according to claim 2, characterized in that, according to the recognized information, generating a grab posture, and the robotic arm reaches a pre-grabbing position to grab the goods, further comprising:

Sort the preliminary grasping postures, the closer the grasping posture is to the geometric center of the grasping surface, the more priority it will be used;

Combining the orientation of the goods and the position of the goods in the seeding box, generate at least one final grab posture on the sorted preliminary grab posture;

According to the final grasping posture, the robot arm of the seeding robot grasps the goods.
The cargo identification and grab method according to claim 3, characterized in that, according to the final grab posture, the seeding robot manipulator arm grabs the goods, the specific method is:

Set the pre-grabbing posture at a certain distance perpendicular to the grasping surface;

From the pre-grabbing position to the pre-grabbing pose, and from the pre-grabbing pose to the grasping point, multiple grasping paths are planned in sections.
The method for identifying and grabbing goods according to claim 3, characterized in that, in combination with the orientation of the goods and the position of the goods in the seed box, at least one final grab posture is generated from the sorted preliminary grab postures, and the specific method is :

If the angle between the bottom of the cargo and the bottom of the seed box is greater than the preset threshold, the sorted initial grasping posture will be used without any change;

If the angle between the bottom of the cargo and the bottom of the seeding box is greater than the preset threshold, and the cargo is close to the right wall of the seeding box, the maximum deflection to the left is α degrees according to the initial grasping posture, where the maximum deflection angle α is the same as that of the robot arm. The elastic length and radius of the gripping actuator used are related;

If the angle between the bottom of the cargo and the bottom of the seeding box is greater than the preset threshold, and the cargo is close to the left wall of the seeding box, the maximum deflection to the right is α degrees according to the initial grasping posture, where the maximum deflection angle α is the same as that of the robot arm. The elastic length and radius of the gripping actuator used are related.
The cargo identification and grab method according to claim 5, characterized in that, according to the initial grab posture, the maximum deflection to the left by α degrees, or the maximum deflection to the right by α degrees, the resolution of each deflection is β degrees, The size of β can be preset.
5. The cargo identification and grab method according to claim 5, wherein the preset threshold is a preset acute angle.
The cargo identification and grab method according to claim 1, wherein the cargo identification device is a visual identification device, a three-dimensional scanner, or a lidar.
A device for identifying and grabbing goods, characterized in that the device includes a memory, a processor, and a computer program stored on the memory and running on the processor, and the computer program is executed by the processor. When executed, the steps described in any one of claims 1 to 8 are realized.
A computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the steps according to any one of claims 1 to 8 are realized.