WO2023061501A1 - 基于货架标识的导航系统、方法 - Google Patents

基于货架标识的导航系统、方法 Download PDF

Info

Publication number
WO2023061501A1
WO2023061501A1 PCT/CN2022/125539 CN2022125539W WO2023061501A1 WO 2023061501 A1 WO2023061501 A1 WO 2023061501A1 CN 2022125539 W CN2022125539 W CN 2022125539W WO 2023061501 A1 WO2023061501 A1 WO 2023061501A1
Authority
WO
WIPO (PCT)
Prior art keywords
shelf
identification
target
image
access device
Prior art date
Application number
PCT/CN2022/125539
Other languages
English (en)
French (fr)
Inventor
张凯博
王华培
Original Assignee
北京极智嘉科技股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from CN202111205706.8A external-priority patent/CN113936010A/zh
Priority claimed from CN202111274173.9A external-priority patent/CN114132679A/zh
Application filed by 北京极智嘉科技股份有限公司 filed Critical 北京极智嘉科技股份有限公司
Publication of WO2023061501A1 publication Critical patent/WO2023061501A1/zh

Links

Images

Classifications

    • BPERFORMING OPERATIONS; TRANSPORTING
    • B65CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
    • B65GTRANSPORT OR STORAGE DEVICES, e.g. CONVEYORS FOR LOADING OR TIPPING, SHOP CONVEYOR SYSTEMS OR PNEUMATIC TUBE CONVEYORS
    • B65G1/00Storing articles, individually or in orderly arrangement, in warehouses or magazines
    • B65G1/02Storage devices
    • B65G1/04Storage devices mechanical
    • B65G1/137Storage devices mechanical with arrangements or automatic control means for selecting which articles are to be removed

Definitions

  • the present application relates to the technical field of warehousing, in particular to a navigation system and method based on shelf identification.
  • the container robot completes the warehousing and outbound work of the container.
  • the robot uses the QR code pasted on the ground in the positioning of the warehouse map.
  • the robot position is accurately corrected twice.
  • the center of the container and the center of the QR code on the robot are aligned to achieve accurate container position.
  • the two-dimensional codes are all pasted on the ground.
  • the points where the two-dimensional codes are pasted are not easy to locate.
  • the two-dimensional codes on the ground are relatively dense, and the workload is relatively large. The accuracy and firmness of pasting the two-dimensional codes are relatively poor. In practical application, it will lead to deviation of navigation.
  • the embodiment of the present application provides a navigation system and method based on shelf identification, so as to solve the technical defects existing in the prior art.
  • a navigation system based on shelf identification includes a control server, an inventory area, an access device, and a shelf, wherein shelves are set in the inventory area, and the space between the shelves A roadway is formed for the movement of access equipment, and shelves are provided with shelf signs extending along the roadway;
  • the control server is configured to determine a target storage position and a target access device on the target shelf, and send a movement instruction to the target access device;
  • the target access device is configured to, in response to the movement instruction, move to the target storage position according to the shelf identifier in the target aisle corresponding to the target shelf.
  • an identification tape is provided on the shelf, and the shelf identification is provided on the identification tape.
  • a plurality of the shelf labels are arranged on a continuous label strip at intervals according to preset rules.
  • a plurality of the shelf marks are equidistantly arranged on one continuous mark strip.
  • the shelf identification includes a two-dimensional code or other identifications other than the two-dimensional code.
  • the system also includes a high-speed passage, the high-speed passage is adjacent to the inventory area, and the high-speed passage is provided with a ground sign;
  • the target access device is further configured to move to the entrance of the target roadway in the expressway according to the ground marking.
  • the shelf includes at least one layer of shelf crossbeams, and the shelf mark is arranged on each shelf crossbeam.
  • the target access device is further configured to move to the target storage position according to the shelf identification on the beam of any shelf.
  • the target access device is further configured to move to the target storage position according to the shelf mark whose distance from the ground is not greater than a preset height threshold.
  • the target access device is further configured to take out the container from the target storage position or place the container in the target storage position according to the shelf ID.
  • the target access device includes a sensor
  • the target access device is further configured to adjust the target access device according to the position deviation between the target access device and the corresponding shelf mark determined by the sensor before taking and placing the container from the target storage position. position to eliminate the positional deviation.
  • the sensor includes an image sensor
  • the target access device is further configured to acquire the target image collected by the image sensor for the target shelf and the calibration parameters of the image sensor, and the target shelf is arranged with an identification tape
  • the identification band is composed of visual identification; extracting the identification image and the preset identification points of the visual identification in the identification image from the target image; based on the calibration parameters, the identification image and the preset identification points , calculating pose information of the identification tape relative to the image sensor; determining a positioning result of the target shelf according to the pose information.
  • the position of the target access device is adjusted to eliminate the position deviation.
  • the target access device is further configured to: perform binarization processing on the target image to obtain a binarized image; identify the coverage area of the visual mark in the binarized image; determine the The identification image composed of the covered area, and the preset identification points of the visual identification are extracted from the covered area.
  • the target access device is further configured to: perform boundary detection on the binary image, determine boundary information of the binary image; identify the binary image based on the boundary information The coverage area of the visual identity in the image.
  • the identification band is composed of multiple visual identifications; the calibration parameters include internal references;
  • the target access device is further configured to: use the internal reference to perform normalization processing on the preset marker points of the visual markers in the marker image, and obtain the preset marker points of the visual markers in the The first coordinates on the logo image; the second coordinates of the preset logo points of the visual logos in the world coordinate system are obtained; according to the first coordinates, the center points of the visual logos are calculated on the logo The third coordinate on the image; according to the second coordinate, calculate the fourth coordinate of the center point of each visual sign in the world coordinate system; according to the third coordinate and the fourth coordinate, use the The center point of the visual mark satisfies the preset arrangement condition as a constraint condition, and the pose information of the mark band relative to the image sensor is calculated.
  • the calibration parameters include internal references
  • the target access device is further configured to: use the internal reference to perform normalization processing on the preset marker points to obtain the first coordinates of the preset marker points on the marker image; obtain the The second coordinates of the preset marker points in the world coordinate system; according to the first coordinates and the second coordinates, calculate the pose information of the marker band relative to the image sensor; acquire the The fifth coordinate of any point in the world coordinate system and the sixth coordinate of any point on the logo image; according to the fifth coordinate and the pose information, calculate the seventh coordinate of any point ; making a difference between the sixth coordinate and the seventh coordinate to obtain a re-projection error, and adjusting the pose information with the re-projection error being smaller than a preset threshold as a constraint condition.
  • the calibration parameters also include distortion parameters
  • the target access device is further configured to: use the distortion parameters to perform de-distortion processing on the identification image to obtain the processed identification image.
  • the calibration parameters also include external parameters; multiple identification tapes are arranged on the target shelf;
  • the target access device is further configured to: calculate the rotation of each identification band around the rotation axis in the world coordinate system according to the pose information of each identification band relative to the image sensor and the preset rotation matrix angle; according to the position and orientation information of each identification band relative to the image sensor and the external parameters, calculate the position and orientation information of each identification band in the world coordinate system; according to the position and orientation information of any two identification bands in the world According to the rotation angle and pose information in the coordinate system, the curve shape between any two identification strips is fitted; according to the curve shape between any two identification strips, the current shape of the target shelf is obtained.
  • the identification tape has a rectangular shape, and the identification tape is arranged on a beam of the target shelf; the target shelf includes a plurality of beams, and the identification tape is arranged on each beam.
  • the shelf includes shelf legs, and the target access device includes a distance measuring device;
  • the target access device is further configured to calculate the running distance from the rack legs on both sides of the target aisle through the distance measuring device, and adjust its position in the target aisle based on the running distance.
  • a shielding mark is provided at the bottom of the shelf, and the target access device includes a distance measuring device;
  • the target access device is further configured to use the distance measuring device to calculate the running distance from the occlusion marks on both sides of the target roadway, and adjust its position in the target roadway based on the running distance.
  • the shielding sign includes a shielding baffle.
  • the shielding mark is arranged around the bottom of the shelf.
  • the ranging device includes a ranging radar or a vision sensor.
  • a navigation system based on shelf identification includes an inventory area, an access device, and a shelf, wherein shelves are set in the inventory area, and a supply chain is formed between the shelves.
  • the aisle where the access equipment moves, and the shelf is provided with a blocking sign, and the access equipment includes a distance measuring device;
  • the access device is configured to use the distance measuring device to calculate the running distance from the shielding signs on both sides of the aisle or the distance from the shelf legs of the shelves on both sides of the aisle, and adjust its distance between the two sides of the aisle based on the running distance. location in the roadway.
  • the shielding mark is arranged around the bottom of the shelf.
  • the shielding sign includes a shielding baffle.
  • a navigation method based on shelf identification is provided, the method is applied to a storage system, and the system includes a control server, an inventory area, an access device, and a shelf, wherein the inventory There are shelves in the area, and the lanes for the movement of access equipment are formed between the shelves, and the shelf signs extending along the lanes are set on the shelves;
  • the methods include:
  • the control server determines the target storage position and the target access device on the target shelf, and sends a movement instruction to the target access device;
  • the target access device moves to the target storage position according to the shelf identifier in the target aisle corresponding to the target shelf.
  • the system also includes a high-speed passage, the high-speed passage is adjacent to the inventory area, and the high-speed passage is provided with a ground sign;
  • the method also includes:
  • the target access equipment moves to a roadway entrance of the target roadway according to the ground marking in the expressway.
  • the shelf includes at least one layer of shelf crossbeams, and the shelf mark is set on each shelf crossbeam;
  • the target access device moves to the target storage position according to the shelf identification, including:
  • the target access device moves to the target storage position according to the shelf identification on the beam of any shelf.
  • the target access device moves to the target storage location according to the shelf identification on the beam of any shelf, including:
  • the target access device moves to the target storage location according to the shelf mark whose distance from the ground is not greater than a preset height threshold.
  • the method also includes:
  • the target access device takes out the container from the target storage position or places the container in the target storage position according to the shelf ID.
  • the target access device includes a sensor
  • the method also includes:
  • the target access device picks and places the container from the target storage position, according to the position deviation of the target access device determined by the sensor and the corresponding shelf mark, adjust the position of the target access device to eliminate the The above position deviation.
  • the shelf includes shelf legs, and the target access device includes a distance measuring device;
  • the method also includes:
  • the target access device calculates the running distance from the rack legs on both sides of the target aisle through the distance measuring device, and adjusts its position in the target aisle based on the running distance.
  • the shelf includes a shielding mark
  • the target access device includes a distance measuring device
  • the target access device calculates the running distance from the occlusion marks on both sides of the target roadway through the distance measuring device, and adjusts its position in the target roadway based on the running distance.
  • the senor includes an image sensor, and determining the position deviation between the target access device and the corresponding shelf mark according to the sensor, adjusting the position of the target access device, and eliminating the position deviation includes:
  • the target shelf is provided with an identification tape, and the identification tape is composed of visual identification;
  • the logo image and the preset logo points calculate the pose information of the logo band relative to the image sensor
  • a positioning result of the target shelf is determined according to the pose information, and a position of the target access device is adjusted according to the positioning result to eliminate the position deviation.
  • the step of extracting the logo image and the preset logo points visually identified in the logo image from the target image includes:
  • An identification image composed of the coverage area is determined, and preset identification points of the visual identification are extracted from the coverage area.
  • the step of identifying the coverage area of the visual identification in the binarized image includes:
  • a visually identified coverage area in the binarized image is identified.
  • the identification band is composed of multiple visual identifications; the calibration parameters include internal references;
  • the step of calculating the pose information of the identification tape relative to the image sensor based on the calibration parameters, the identification image and the preset identification points includes:
  • the calibration parameters include internal references
  • the step of calculating the pose information of the identification tape relative to the image sensor based on the calibration parameters, the identification image and the preset identification points includes:
  • a difference between the sixth coordinate and the seventh coordinate is obtained to obtain a re-projection error, and the pose information is adjusted with the re-projection error being smaller than a preset threshold as a constraint condition.
  • de-distortion processing is performed on the logo image to obtain the processed logo image.
  • the calibration parameters also include external parameters; multiple identification tapes are arranged on the target shelf;
  • the step of determining the positioning result of the target shelf according to the pose information it also includes:
  • the current form of the target shelf is obtained.
  • the identification tape has a rectangular shape, and the identification tape is arranged on a beam of the target shelf; the target shelf includes a plurality of beams, and the identification tape is arranged on each beam.
  • a computer-readable storage medium stores a computer program, and it is characterized in that, when the computer program is executed by a processor, the above-mentioned shelf-based identification The steps of the navigation method.
  • the navigation system based on the shelf identification includes a control server, an inventory area, an access device, and a shelf, wherein shelves are set in the inventory area, and a shelf for the access device to move is formed between the shelves.
  • the shelf In the laneway, the shelf is provided with a shelf mark extending along the laneway;
  • the control server is configured to determine the target storage position and the target access device on the target shelf, and send a movement instruction to the target access device;
  • the target The access device is configured to, in response to the movement instruction, move to the target storage position according to the shelf identifier in the target aisle corresponding to the target shelf.
  • Fig. 1 is a schematic diagram of the system structure of the storage system provided by the embodiment of the present application.
  • Fig. 2 is the front view of the shelf provided by the embodiment of the present application.
  • Fig. 3 is a flow chart of a shelf positioning method provided by an embodiment of the present application.
  • Fig. 4 is a flowchart of another shelf positioning method provided by an embodiment of the present application.
  • Fig. 5 is a flowchart of another shelf positioning method provided by an embodiment of the present application.
  • Fig. 6 is a schematic diagram of a shelf provided by another embodiment of the present application.
  • Fig. 7 is a flow chart of a navigation method based on shelf identification provided by an embodiment of the present application.
  • Fig. 8 is a schematic structural diagram of a shelf positioning device provided by an embodiment of the present application.
  • Fig. 9 is a structural block diagram of a shelf handling device provided by an embodiment of the present application.
  • 110-control server 111-processor, 112-memory, 113-order pool, 114-operating table, 120-inventory area, 1201-lane, 130-shelf, 1301-shelf identification, 1302-container, 1303-blocking identification , 140-access equipment, 150-high-speed channel.
  • first, second, etc. may be used to describe various information in one or more embodiments of the present application, the information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second, and similarly, second may also be referred to as first, without departing from the scope of one or more embodiments of the present application. Depending on the context, the word “if” as used herein may be interpreted as “at” or “when” or “in response to a determination.”
  • Rack handling equipment is an automated device capable of handling shelves.
  • Binarization processing the technology of setting the grayscale value of the pixels on the image to 0 or 255, that is, to present the entire image with an obvious visual effect of only black and white.
  • CCD is the abbreviation of Charge Coupled Device (Charge Coupled Device), which can convert light into charge and store and transfer the charge, and can also take out the stored charge to change the voltage.
  • CCD cameras are widely used due to their small size, light weight, unaffected by magnetic fields, and anti-vibration and shock resistance.
  • Calibration of image sensors In the process of image measurement and machine vision applications, in order to determine the relationship between the three-dimensional geometric position of a point on the surface of a space object and its corresponding point in the image, it is necessary to establish a geometric model of image sensor imaging. These geometric models The parameters are the parameters of the image sensor. Under most conditions, these parameters must be obtained through experiments and calculations. This process of solving parameters is called calibration. Calibration methods mainly include: traditional calibration method, active vision calibration method, self-calibration method, and zero-distortion calibration method.
  • Visual identification patterns with special identification features, including QR codes, special geometric patterns, special logos or nameplates, etc.
  • the container access equipment completes the storage and output work of the container.
  • the access equipment is located in the storage area.
  • the positioning of the location is usually to paste the two-dimensional code on the ground.
  • the position of the access device is accurately corrected twice.
  • the center of the container is aligned with the center of the two-dimensional code on the shelf to realize the
  • the two-dimensional codes pasted on the ground are relatively dense, and the workload is relatively large.
  • the workload of pasting the two-dimensional codes on the shelves is also large, and the pasting accuracy and firmness are usually relatively poor.
  • Fig. 1 shows a schematic diagram of the system structure of the storage system provided by the embodiment of the present application, the system includes a control server 110, an inventory area 120, a shelf 130, and an access device 140, wherein the inventory area 120 is set There are shelves 130 on which shelf signs are arranged, and the shelves 130 include aisles 1201 for moving access equipment.
  • the control server 110 is connected to the access device 140 by wireless communication, and the staff operates the control server 110 through the operation console 114.
  • the control server 110 is a software system running on the server, which has data storage and information processing capabilities, and can be accessed via wireless
  • the control server 110 may include one or more servers, and may be a centralized control architecture or a distributed computing architecture.
  • the control server 110 has a processor 111 and a memory 112, and may have an order pool 113 in the memory 112 , order information is stored in the order pool 113.
  • the inventory area 120 can be a dense storage area or a non-intensive storage area.
  • shelves 130 are arranged, and there are lanes 1201 for the access equipment 140 to move between the shelves 130.
  • the shelves 130 include at least one Layer shelf beams, the shelf beams divide the shelf 130 into at least one layer, the shelf beams of the shelves are provided with storage positions, and each storage position can carry a container that can accommodate stored goods, such as a material box.
  • the shelf is also provided with a shelf mark, as shown in FIG. 2 , which shows a schematic diagram of a shelf provided by an embodiment of the present application.
  • the shelf 130 is provided with a shelf mark 1301 , specifically, the shelf 130 includes at least one layer of shelf crossbeams, and the shelf mark 1301 is set on the shelf crossbeam of the shelf 130 .
  • a container 1302 capable of accommodating stored goods is placed on the shelf crossbeam.
  • the passage between the shelves is called a lane 1201 , and the access device 140 can move in the lane 1201 .
  • the access device 140 may be a container access robot, a stacker, a forklift, and the like. Access devices are used to remove containers from the rack or to place containers into storage locations on the rack.
  • control server is configured to determine a target storage location and a target access device on the target shelf, and send a movement instruction to the target access device.
  • the target access device is configured to, in response to the movement instruction, move to the target storage position according to the shelf identifier in the target aisle corresponding to the target shelf.
  • the control server 110 may determine the target storage position on the target shelf according to the target business task, and the specific business task may be a picking task, a loading task, an inventory task, and the like.
  • the target storage location may specifically be a storage location where the container needs to be taken out or a storage location where the container needs to be put in.
  • each row of shelves may be composed of multiple shelves, and the target storage location may be a storage location on one of the target shelves 130 .
  • the control server is also configured to determine the target access device 140 for performing the target business task, and send a movement instruction to the target access device 140, so that the target access device 140 moves to the target storage position of the target shelf.
  • the target access device 140 is configured to, in response to the moving instruction, when moving in the target aisle 1201 corresponding to the target shelf 130, locate according to the shelf identification on the shelf, and move to the target storage position of the target shelf.
  • the access device 140 uses the shelf marks on the shelves for positioning and navigation.
  • the shelf signs assist the access equipment to locate and navigate, reducing the work of setting up navigation signs in the roadway and reducing the workload.
  • Pasting the shelf logo on the shelf can assist access equipment navigation, but if the shelf logo is pasted independently, the workload is relatively large. Although there is a shelf beam as a reference, there may be some errors. Based on this, you can also set it on the shelf.
  • An identification belt the shelf identification is set on the identification belt, specifically, a plurality of the shelf identifications are arranged on a continuous identification belt at intervals according to preset rules, and the preset rules can be equidistant settings, It can also be a non-equidistant setting. In the case of a non-equidistant setting, it is only necessary to ensure that the distance between each shelf mark is regular. Preferably, a plurality of said shelf marks are equidistantly arranged on one continuous said mark strip.
  • the identification tape can be fixed on the shelf by pasting, or by spraying, painting, etc.
  • the shelf label can be used to locate the container, and it can also be used to provide navigation and positioning for the access device. Different types of containers are mixed and stored, which can correspond to different shelf labels. As shown in Figure 2, a shelf label It can correspond to one container position, or two or more shelf marks can correspond to one container position. In practical applications, the control server has already stored the container corresponding to each shelf ID, and can directly send a device access instruction to the access device to access the corresponding target container.
  • the shelf mark can be a positioning two-dimensional code, or other marks other than the positioning two-dimensional code, such as a bar code, a digital serial number, and the like.
  • the shelf mark can be pasted to the side of the shelf beam of the shelf in the form of pasting, or can be fixed on the side of the shelf beam of the shelf by printing, spraying, etc.
  • a piece of mark tape can also be used for continuous numbering, for example, a shelf can be pasted 6 shelf marks, two adjacent shelves can use 1-12 as shelf marks.
  • the shelf marks on the mark tape can be set at intervals according to the preset spacing. For example, if the width of the shelf is 200 cm, a 200 cm long mark tape can be set. Set a shelf mark every 40 cm on the shelf.
  • the storage system also includes a high-speed passage 150, which is adjacent to the storage area 120, and the access equipment passes through the high-speed passage 150 between the storage area 120 and the workstation. Move containers from storage areas to workstations, or from workstations back to storage areas.
  • the access equipment in the expressway 150 also needs to perform navigation and positioning according to the ground signs set on the ground, that is, when the access equipment moves in the expressway 150
  • the navigation is performed according to the ground signs in the expressway 150 .
  • the navigation is performed according to the shelf marks on the shelves 130 .
  • the ground sign 150 in the expressway and the shelf sign in the roadway can provide navigation and positioning for the access device, guide the access device 140 to the target shelf 130, and reduce the workload of pasting the ground mark in the roadway.
  • the access device 140 includes a scanning device, and the access device scans the shelf mark through the scanning device to realize navigation and positioning.
  • the shelf includes at least one layer of shelf beams, the shelf mark setting On the beams of each shelf, the access equipment can perform positioning and navigation according to the shelf marks on the beams of any shelf, and then move to the target storage position.
  • the access device when it is moving, it can also scan the shelf mark whose height from the ground is not greater than the preset height threshold for navigation and positioning, which can effectively reduce the center of gravity of the access device and increase the moving speed of the access device and security. That is, after the access device enters the target aisle, the scanning device of the access device is aligned with the shelf mark not higher than the preset height threshold for navigation and positioning.
  • the container handling robot When the access device 140 is a container handling robot, the container handling robot is provided with a pick-and-place container device, which is used to take out the container from the shelf or put the container into the shelf 130 .
  • the scanning device of the container handling robot is set on the pick-and-place container device.
  • the container handling robot can store or take out the container by scanning the shelf mark through the scanning device. Specifically, when the container handling robot enters the target lane, the scanning device is aligned with the distance The shelf mark whose ground height is not greater than the preset height threshold performs a scanning operation to obtain the current position of the access device, and lowers the pick-and-place container device below the preset height threshold, which can effectively reduce the center of gravity of the access device and improve the efficiency of the access device.
  • the pick-and-place container device is moved to the target storage position by adjusting the height of the pick-and-place container device. That is, the moving speed of the container handling robot is guaranteed, and the safety and stability of the container handling robot during the moving process are guaranteed, and the storage and picking efficiency of the storage system is improved.
  • the container can be taken out from the target storage position, or the container can be put into the target storage position to complete the control The target business task of the server 110 .
  • the access device includes left and right wheels, and encoders are installed in the motors of the left and right wheels, and the mileage information of the access device can be obtained by driving the code disc on the motor.
  • the gyroscope can obtain the angle information, and the relative position of the access device can be deduced and calculated by combining the gyroscope and mileage information.
  • the relative position of the access device will exist Accumulated errors, therefore, in the area of expressway 150, the ground markings pasted on the ground can be used as the absolute position and the relative position of the access equipment for correction to obtain the optimal position information of the access equipment.
  • the height of the pick-and-place container device of the scanning device can be moved to a height that is not greater than the preset height threshold according to the height from the ground, for example, put
  • the height of the scanning device pick-and-place container device is adjusted to the lowest shelf beam on the shelf in the vertical direction, so that the scanning device on the pick-and-place container device can be aligned with the shelf mark on the shelf beam.
  • the target access device includes a sensor; the target access device is further configured to determine the position of the target access device and the corresponding shelf identifier according to the sensor before taking and placing a container from the target storage position deviation, adjusting the position of the target access device to eliminate the position deviation.
  • the senor may be a vision sensor, a depth information sensor or a 3D sensor.
  • the access device When the access device is a container handling robot, the access device scans the shelf mark on the shelf beam through the sensor, moves to the front of the target shelf, and then adjusts the height of the pick-and-place container device to move the pick-and-place container device to the target storage position place.
  • the height of the pick-and-place container device can also be positioned through the shelf marks on the beams of each shelf in the vertical direction.
  • a shelf positioning method is provided, as shown in Figure 3, which shows a shelf positioning provided by an embodiment of the present invention
  • the flowchart of the method specifically includes the following steps.
  • Step 302 acquiring the target image collected by the image sensor for the target shelf and the calibration parameters of the image sensor, wherein the target shelf is provided with an identification tape, and the identification tape is composed of a visual identification.
  • an image sensor can also be used to achieve precise positioning of the shelf, wherein the image sensor is a sensor that can support the conversion of an optical image into a digital signal and transmits the digital signal of the image to the processor, such as a CCD camera.
  • the image sensor shoots the scene within the field of view, and transmits the captured target image to the execution subject of the embodiment of the present invention, and the execution subject acquires the target image collected by the image sensor.
  • the scene within the field of view of the image sensor is the target shelf.
  • the image sensor can only capture a specific part of the target shelf, and the target shelf is equipped with an identification tape (the identification tape is generally placed on the target shelf artificially).
  • the identification tape can be pasted on the beam, pole, bottom, top and other positions of the target shelf that can be captured by the image sensor. Rhombus etc.
  • the image sensor is installed at the front end of the rack handling equipment.
  • the specific part may be a beam at the bottom of the target rack.
  • the identification tape is composed of visual signs, which can be two-dimensional codes, special geometric patterns, special signs or nameplates, etc. That is to say, the identification tape includes at least one identification mark, and the identification tape can be composed of multiple visual signs.
  • the identification tape is in the shape of a rectangle, and the identification tape is arranged on the beam of the target shelf.
  • the beam of the shelf is located on the side of the shelf, it can be easily captured by the image sensor, and the beam of the shelf is generally long. Therefore, the use of a rectangular identification tape can improve the positioning accuracy. In order to ensure higher positioning accuracy, the identification tape can be set The dimensions are the same as the dimensions of the beam.
  • the image sensor is pre-calibrated, therefore, the calibration parameters of the image sensor can be obtained.
  • the specific calibration method can be to use the traditional calibration method, the active vision calibration method, the self-calibration method or the zero-distortion calibration method.
  • the image sensor is calibrated, and the specific calibration implementation process is not the main content of the present invention, so it will not be repeated here.
  • Calibration parameters mainly include internal reference, external reference and distortion parameters.
  • the internal parameter is a 3x3 parameter matrix, which is used to reflect the influence of the focal length attribute of the image sensor on the image
  • the external parameter refers to the setting parameters of the image sensor relative to the world coordinate system
  • the distortion parameter includes radial distortion and tangential distortion, and is A group of arrays related to the degree of distortion of the image captured by the image sensor.
  • the formats of the distortion parameters of different image sensors are different.
  • the field of view should include the identification tape, and the identification tape should be located in the center of the field of view as much as possible. In order to improve the positioning accuracy, it should include as much as possible Complete identification tape.
  • Step 304 extracting the logo image and the preset logo points visually identified in the logo image from the target image.
  • the identification image refers to the image of the area where the identification band is located in the target image collected by the image sensor.
  • the specific way to extract the logo image can be based on the neural network, input the image into the pre-trained recognition model, and the recognition model can directly output the logo image in the image; the way to extract the logo image can also be based on image processing Ways, such as binarization of images, pixel recognition, etc.
  • the preset identification points refer to the points on the visual identification that are easy to extract, have a definite positional relationship, and relatively wrap the visual identification, such as a square two-dimensional
  • the code will extract the four corner points, the circular lattice will extract the center of each small circle, the triangle polygon or other geometric features will also extract the corner points, etc.
  • step 304 may specifically be implemented in the following manner:
  • the target image can be binarized, and the binarization process can use adaptive binarization, which is used to judge whether a pixel in the target image is in a nearby interval
  • the darker or brighter area is obtained by comparing the pixel value of this pixel with the average or weighted average of the pixel values of the surrounding area, so as to obtain a binarized image.
  • the binarized image shows obvious only black and White visual effect, the coverage area of the visual logo will be significantly different from the surrounding area, therefore, the coverage area of the visual logo can be identified from the binarized image, further, the logo image composed of the coverage area can be determined, and, Preset marker points for visual identities can be extracted from the coverage area.
  • the binarization processing method is used to improve the efficiency and accuracy of extracting logo images.
  • the step of binarizing the target image to obtain the binarized image can be specifically implemented in the following manner:
  • the region of interest (ROI, Region of Interest) in the target image can be extracted first according to such acquisition rules.
  • Interest the method of extracting the ROI region in the target image, in addition to the above-mentioned extraction according to the collection rules, can also be extracted using a neural network-based approach, that is, the target image is input into a pre-trained neural network model, and the neural network model Identify the area in the target image where there may be a logo band, and this area is the ROI area.
  • the ROI area is binarized to obtain a binarized image.
  • the step of identifying the coverage area of the visual identification in the binarized image can be specifically implemented in the following manner:
  • Boundary refers to the boundary line with high black-and-white contrast in the binary image, that is, there is a high black-white contrast on both sides of the boundary. Based on this, the boundary detection of the binary image can be performed according to the change of black-white contrast to determine The boundary information of the binarized image, which may include the position, shape, length, etc. of the boundary. Based on the boundary information of the binarized image, all possible lines in the binarized image can be obtained, and these lines can be combined to identify the coverage area that may be a visual mark. By means of boundary detection, the coverage areas of visual signs are determined, and the coverage areas of each visual sign are identified more conveniently and quickly, thereby improving the accuracy and efficiency of extracting sign images.
  • the boundary information is filtered using a preset filtering method to obtain the filtered boundary information .
  • the boundary information is filtered.
  • the filtering is mainly to filter the characteristics of the above-mentioned boundary lines or line combinations, such as excluding lines whose length or direction does not meet the requirements, excluding certain angles of line combinations that do not meet the requirements, etc.
  • the specific filtering method can be based on the visual The type of identification is determined.
  • the filtered boundary information is more in line with the actual requirements, and the extracted logo image is more in line with the actual situation and has higher precision.
  • Step 306 based on the calibration parameters, the logo image and the preset logo points, calculate the pose information of the logo tape relative to the image sensor.
  • the pose information of the identification tape relative to the image sensor can be calculated based on the calibration parameters, the identification image and the preset identification points.
  • the calibration parameters stipulate some attributes of the image sensor when collecting images, such as the distortion of the collected images, focal length, etc., and the preset identification points that identify the visual identification on the image are some convenient extraction, with a certain positional relationship, And relative to the points wrapped by the visual marker, the pose information of the marker band relative to the image sensor can be calculated based on the calibration parameters, marker image and preset marker points.
  • the internal parameters are used to normalize the preset marker points of the visual marker in the marker image, and the first coordinates (u, v, 1) of the preset marker points of the visual marker on the marker image can be obtained.
  • the second coordinates (x, y, z) of the preset logo point in the world coordinate system can be obtained.
  • (u, v, 1) is actually the projection of the line from the image sensor to (x, y, z) on the image plane, and the preset identification points of the visual identification are on the same plane, therefore, according to the first
  • the coordinates and the second coordinates are used to calculate the relative position information of the preset marker point relative to the image sensor, and the relative posture information of the plane where the preset marker point is located relative to the image sensor by using the triangular relationship.
  • the specific calculation process can be as follows: the direct rotation relationship between the image plane and the plane to which the preset marker points of the visual marker belong is represented by a rotation matrix R, and the translation relationship between the image plane and the origin of the plane to which the preset marker points belong is expressed by a displacement vector T , according to the conversion relationship between coordinate systems, there is a relationship between the first coordinate (u, v, 1) and the second coordinate (x, y, z) as shown in formula (1):
  • R and T can be obtained by solving formula (1), that is, the relative rotation and relative displacement between the image plane and the plane to which the preset marker points belong.
  • formula (1) that is, the relative rotation and relative displacement between the image plane and the plane to which the preset marker points belong.
  • the identification band is composed of multiple visual identifications; the calibration parameters include internal references.
  • step 306 may specifically be implemented in the following manner:
  • the center point of each visual mark satisfies the preset arrangement condition as a constraint condition, and calculates the pose information of the mark band relative to the image sensor.
  • the identification band can be composed of multiple visual identifications, and the preset identification points of each visual identification in the identification image are normalized using internal parameters, and the preset identification points of each visual identification can be obtained in the identification image.
  • the second coordinates of the preset sign points of each visual sign in the world coordinate system can be obtained.
  • the third coordinates of the central points of each visual mark on the mark image can be calculated according to the first coordinates. For example, if the visual mark is a rectangular two-dimensional code, the preset mark points are four corner points, and the corner points are connected The intersection point of the lines is the center point of the visual sign, and the third coordinate of the center point can be calculated according to the first coordinates of the four corner points.
  • the fourth coordinate of the center point of each visual sign in the world coordinate system can also be calculated according to the second coordinate. Then, according to the third coordinate and the fourth coordinate, with the center point of each visual mark satisfying the preset arrangement condition as the constraint condition, calculate the pose information of the sign belt relative to the image sensor, wherein the preset arrangement condition is the sign belt.
  • the arrangement of visual signs in the center is generally affected by the shape of the signage tape. For example, if the signage strip is rectangular, the visual signs in it are arranged in rows. Therefore, the default arrangement condition is that the center points are on the same straight line, and then For example, if the identification tape is circular, the visual identification therein is arranged in a circle, so the default arrangement condition is that the center point is on the circumference of the same circle.
  • a specific formula for calculating pose information may be formula (1).
  • the center point of each visual identification on the same identification belt should meet the preset arrangement conditions, and use this as a constraint to calculate the pose information, which can preclude the situation of pose jumping and ensure The accuracy of pose information calculation is improved.
  • the calibration parameters include internal references.
  • step 306 may also specifically be implemented in the following manner:
  • the first coordinate and the second coordinate calculate the pose information of the identification tape relative to the image sensor
  • the difference between the sixth coordinate and the seventh coordinate is obtained to obtain the re-projection error, and the pose information is adjusted with the re-projection error being smaller than the preset threshold as a constraint condition.
  • the first coordinates of the preset marker points on the marker image can be obtained.
  • the preset Set the second coordinate of the marker point in the world coordinate system.
  • iterative optimization needs to be performed on the premise that preset angle jumps may exist.
  • the part with more pixels on the image can achieve higher calculation accuracy.
  • the rectangular identification tape can further improve the accuracy of roll angle and yaw angle.
  • the optimal pose information can be obtained. This process is also called reverse calculation, which can effectively deal with the situation of pose jumping and ensure the accuracy of pose information calculation.
  • the logo image is de-distorted to obtain the processed logo image.
  • the extracted logo image is de-distorted using the distortion parameters, and the processed logo image is a plane image, which offsets the error caused by the distortion and can improve the final positioning accuracy.
  • Step 308 determine the positioning result of the target shelf according to the pose information.
  • the positioning result of the target shelf can be determined according to the position information, and the positioning result of the target shelf is According to the pose information and the preset position of the identification tape, the pose information of the target shelf relative to the image sensor is converted.
  • the calibration parameters also include external parameters; multiple identification tapes are arranged on the target shelf;
  • step 308 the following steps may also be included:
  • the current shape of the target shelf is obtained.
  • multiple identification tapes can be arranged on the target shelf, and after obtaining the pose information of each identification tape relative to the image sensor, the pose information of each identification tape relative to the image sensor can also be combined , thus describing the shape of the target shelf.
  • the point on a certain surface of the target shelf be One of the specified points is the point pasted by a logo tape.
  • the pose of the point ajk is Rjk, Tjk.
  • the rotation angle of the point around the three rotation axes in the world coordinate system can be obtained. Since the pose information of the m ⁇ n points relative to the image sensor is obtained, the pose information of each point in the world coordinate system can be converted under the premise of having the external parameters of the image sensor relative to the world coordinate system.
  • the shelf is a frame structure, between any two points, the curve shape between the two points can be fitted according to the rotation angle along the frame direction at the two points and the pose information of the two points relative to the world coordinate system.
  • a shape evaluation function of the target shelf is added, which can provide more suitable data support for shelf handling and the like.
  • the identification tape may be in the shape of a rectangle, and the identification tape may be arranged on the crossbeam of the target shelf; the target shelf may include a plurality of crossbeams, and the identification tape is arranged on each crossbeam.
  • the target shelf is arranged with a sign tape composed of visual signs
  • the sign image and the sign are extracted from the target image collected by the image sensor.
  • the preset marker points of the visual marker in the marker image are calculated based on the calibration parameters, the marker image and the preset marker points, and the pose information of the marker tape relative to the image sensor is calculated, and the positioning result of the target shelf is determined according to the pose information.
  • the position of the logo tape relative to the image sensor is calculated. Since the identification tape is pre-arranged on the target shelf and is composed of visual signs, the positioning result of the target shelf can be obtained. Using the identification tape to locate the target shelf avoids signal interference in the actual scene, thereby improving the accuracy of the shelf positioning result.
  • FIG. 4 shows a flow chart of another shelf positioning method provided by an embodiment of the present invention, which specifically includes the following steps.
  • Step 402 acquiring the target image collected by the image sensor for the target shelf and the calibration parameters of the image sensor, wherein, the target shelf is arranged with an identification tape, and the identification tape is composed of a visual identification.
  • Step 404 extracting the logo image and the preset logo points visually identified in the logo image from the target image.
  • Step 410 using a preset verification method to verify whether the visual mark in the mark image conforms to the preset visual mark features. If the verification result is that the verification is passed, execute step 406 , otherwise return to execute step 404 .
  • Step 406 based on the calibration parameters, the logo image and the preset logo points, calculate the pose information of the logo tape relative to the image sensor.
  • Step 408 determine the positioning result of the target shelf according to the pose information.
  • steps 402 , 404 , 406 , and 408 are the same as the steps in the embodiment shown in FIG. 3 , and will not be repeated here.
  • the preset marker points of the visual markers in the extracted marker image may not be the marker points of the actual visual markers due to various reasons, but are caused by insufficient filtering, it is necessary to check whether the visual markers in the marker image conform to the preset visual marker features Perform verification to confirm whether the visual identity is the preset visual identity. For example, two-dimensional codes will be decoded, geometric features will be subject to feature inspection, and special signs will be image-matched. Only when the verification is passed, the step of calculating the pose information will be executed. The specific verification method may be to input the identification image into the corresponding verification function or model, and confirm whether the output result is verified.
  • the target shelf is arranged with a sign tape composed of visual signs
  • the sign image and the sign are extracted from the target image collected by the image sensor.
  • the preset marker points of the visual marker in the marker image are calculated based on the calibration parameters, the marker image and the preset marker points, and the pose information of the marker tape relative to the image sensor is calculated, and the positioning result of the target shelf is determined according to the pose information.
  • the position of the logo tape relative to the image sensor is calculated. Since the identification tape is pre-arranged on the target shelf and is composed of visual signs, the positioning result of the target shelf can be obtained. Using the identification tape to locate the target shelf avoids signal interference in the actual scene, thereby improving the accuracy of the shelf positioning result.
  • the step of pose information calculation will be performed to avoid It solves the problem that the preset identification points caused by insufficient filtering do not conform to the actual situation, and the positioning results are more accurate.
  • a CCD camera is installed on the front end of the shelf handling equipment, and the shelf handling equipment also includes a processor, and the processor executes the shelf positioning method as shown in Figure 5, and Figure 5 shows that an embodiment of the present invention provides A flow chart of another shelf positioning method, which specifically includes the following steps.
  • the first step is to calibrate the CCD camera, obtain the internal reference and distortion parameters of the CCD camera, and arrange the visual signs on specific parts of the shelf.
  • the CCD camera is used to collect pictures containing the complete identification band in the field of view.
  • the third step is to perform binarization processing on the ROI region of the camera containing the identification band to obtain a binarized image.
  • the binarization process here is adaptive binarization, which is used to judge whether the pixel point in the picture is a darker or brighter area in the nearby interval, that is, the average value of the pixel value of the pixel point and the pixel value of the surrounding area or weighted averages for comparison.
  • the fourth step is to perform boundary detection in the binarized image, find all the boundaries in the binary image, and perform boundary filtering to extract the logo image and the preset logo points for each visual logo in the logo image to participate in the pose calculation.
  • the boundary detection is performed on the binary image to determine all the boundaries in the binary image. Extracting the boundary can be approximately understood as extracting all possible lines in the binary image, and These lines are combined and filtered to identify areas that may be visual identities. Boundary filtering is mainly to filter the characteristics of these lines or line combinations, such as excluding lines whose length or direction does not meet the requirements, or excluding lines whose angles do not meet the requirements. The specific filtering method is determined according to the type of visual identification. According to the result of the boundary detection, the logo image can be extracted from the binarized image, and the preset logo points of each visual logo participating in the pose calculation can be extracted from the logo image, such as four corner points of a square two-dimensional code.
  • the fifth step is to verify each visual logo in the extracted logo image, and after verification, calculate the pose information of the logo belt relative to the CCD camera according to the internal parameters and distortion parameters of the CCD camera and the preset logo points.
  • the preset identification points extracted in the previous step may not be the identification points of the actual visual identification due to various reasons, but are caused by insufficient filtering, it is necessary to verify the unique special visual features of each visual identification in the identification image , to confirm whether it conforms to the preset visual identity features.
  • the position and orientation information of the marker band relative to the CCD camera is calculated according to the internal reference and distortion parameters of the CCD camera and the preset marker points.
  • the pose information is output in the form of coordinates, and the output results are the position T and attitude R of the visual marker relative to the CCD camera.
  • the position T is the position (x, y, z) of the visual marker relative to the CCD camera
  • the attitude R is the relative position of the visual marker
  • the positioning result of the target shelf can be converted, and the positioning result can be directly output, so that the shelf handling equipment can accurately transport the shelf according to the positioning result.
  • This application uses visual signs to locate target shelves.
  • the positioning result of the target shelf is obtained by calculating the pose information of the visual mark, which has higher accuracy than the positioning scheme of radio frequency technology or WLAN technology, and has higher adaptability than the deep learning scheme.
  • the shelf positioning method provided in this embodiment includes: acquiring the target image collected by the image sensor for the target shelf and the calibration parameters of the image sensor, wherein the target shelf is arranged with an identification band composed of visual signs, and is extracted from the target image collected by the image sensor.
  • the logo image and the preset logo points of the visual logo in the logo image based on the calibration parameters, logo image and preset logo points, calculate the pose information of the logo tape relative to the image sensor, and determine the positioning result of the target shelf according to the pose information .
  • the logo image in the collected target image and the preset logo points of the visual logo in the logo image based on the calibration parameters of the image acquisition device, the logo image and the preset logo points, the position of the logo tape relative to the image sensor is calculated. Since the identification tape is pre-arranged on the target shelf and is composed of visual signs, the positioning result of the target shelf can be obtained. Using the identification tape to locate the target shelf avoids signal interference in the actual scene, thereby improving the accuracy of the shelf positioning result.
  • the rotation angle of the access device is determined by the cumulative calculation of the gyroscope. Due to the influence of the gyroscope drift, the rotation angle of the access device will have cumulative errors. Only relying on the gyroscope to determine the rotation angle of the access device will cause collisions between the access device. Therefore, it is also necessary to provide a solution to ensure that the access equipment will not hit the shelves when moving in the aisle.
  • a shielding mark 1303 is provided at the bottom of the shelf 130, and the target access device includes a distance measuring device;
  • the target access device 140 is further configured to calculate the running distance from the occlusion signs 1303 on both sides of the target roadway 1201 through the distance measuring device, and adjust its position in the target roadway 1201 based on the running distance. Location.
  • Fig. 6 shows a schematic view of a shelf provided by an embodiment of the present application.
  • the shelf includes a shelf crossbeam, and an identification band is provided on the side of each shelf crossbeam, and the identification band is provided with A shelf sign 1301, a container 1302 is placed on the beam of the shelf, and a shielding sign 1303 is set between the legs of the shelf at the bottom of the shelf.
  • the shielding mark is arranged around the bottom of the shelf, specifically, the shielding mark can be set on the ground at the bottom of the shelf, and can be fixed between the legs of the shelf by a fixing device, It can be fixed on the bottom of the bottom shelf beam of the shelf by the fixing device.
  • the access device In addition to setting a shield mark at the bottom of the shelf, in another specific embodiment provided by this application, it is also possible to place the access device according to the shelf legs of the shelf to collide with the shelf.
  • the target access device is also identified It is configured to use the distance measuring device to calculate the running distance from the rack legs on both sides of the target aisle, and adjust its position in the target aisle based on the running distance.
  • the target access device 140 recognizes the occlusion mark 1303 or the shelf leg of the shelf through the distance measuring device, and calculates the running distance of the target access device 140 from the occlusion mark 1303 or the shelf leg, and by adjusting the movement distance, it is ensured that the target access device is within In the process of moving in the target lane, always keep a certain safe distance from the shelf, and move according to the preset angle direction.
  • the navigation system based on the shelf identification includes a control server, an inventory area, an access device, and a shelf, wherein shelves are set in the inventory area, and a shelf for the access device to move is formed between the shelves.
  • the shelf In the laneway, the shelf is provided with a shelf mark extending along the laneway;
  • the control server is configured to determine the target storage position and the target access device on the target shelf, and send a movement instruction to the target access device;
  • the target The access device is configured to, in response to the movement instruction, move to the target storage position according to the shelf identifier in the target aisle corresponding to the target shelf.
  • the method of setting shelf marks on the identification tape reduces the complexity of construction and improves the construction efficiency.
  • another navigation system based on shelf identification includes an inventory area 120, an access device 140, and a shelf 130, wherein the inventory area is provided with a shelf 130, The shelf is provided with a blocking sign 1303;
  • the access device 140 is configured to run in the aisle between the shelves according to the occlusion mark.
  • the access device 140 includes a distance measuring device
  • the access device 140 is further configured to calculate the running distance from the occlusion sign 1303 through the distance measuring device, and adjust the running distance to a preset distance interval during the movement in the roadway.
  • the shielding sign is arranged at the bottom of the shelf. Furthermore, the shielding sign includes a shielding baffle.
  • the navigation system based on the shelf identification provided by the embodiment of the present application includes an inventory area, an access device, and a shelf, wherein the inventory area is provided with a shelf, and the shelf is provided with a shielding identification; the access device is configured as Run in the lane between the shelves according to the occlusion sign. It can prevent the access equipment from colliding with the shelf when it moves in the storage area, and improves the safety of the access equipment operation.
  • FIG. 7 shows a flow chart of a navigation method based on shelf identification provided by an embodiment of the present application.
  • the method is applied to a storage system, and the storage system includes a control server, an inventory area, an access device, and a shelf , wherein, the storage area is provided with shelves, and an aisle for the movement of access equipment is formed between the shelves, and a shelf mark extending along the aisle is provided on the shelf.
  • Step 702 The control server determines the target storage location and the target access device on the target shelf, and sends a movement instruction to the target access device.
  • Step 704 In response to the movement instruction, the target access device moves to the target storage position according to the shelf identifier in the target aisle corresponding to the target shelf.
  • system further includes a high-speed passage, the high-speed passage is adjacent to the inventory area, and the high-speed passage is provided with a ground sign;
  • the method also includes:
  • the target access equipment moves to a roadway entrance of the target roadway according to the ground marking in the expressway.
  • the target access device uses the ground signs on the ground for navigation and positioning.
  • the access device moves to the target shelf corresponding to At the entrance of the target aisle, navigate and locate according to the shelf identification on the shelf.
  • the shelf includes at least one layer of shelf crossbeams, and the shelf mark is arranged on each shelf crossbeam;
  • the target access device moves to the target storage position according to the shelf identification, including:
  • the target access device moves to the target storage position according to the shelf identification on the beam of any shelf.
  • the target access device moves to the target storage position according to the shelf identification on the beam of any shelf, including:
  • the target access device moves to the target storage location according to the shelf mark whose distance from the ground is not greater than a preset height threshold.
  • the method also includes:
  • the target access device takes out the container from the target storage position or places the container in the target storage position according to the shelf ID.
  • the target access device includes a sensor
  • the method also includes:
  • the target access device picks and places the container from the target storage position, according to the position deviation of the target access device determined by the sensor and the corresponding shelf mark, adjust the position of the target access device to eliminate the The above position deviation.
  • the senor includes an image sensor, and the position deviation of the target access device from the corresponding shelf mark is determined according to the sensor, and the position of the target access device is adjusted, Elimination of said positional deviations, including:
  • the target shelf is provided with an identification tape, and the identification tape is composed of visual identification;
  • the logo image and the preset logo points calculate the pose information of the logo band relative to the image sensor
  • a positioning result of the target shelf is determined according to the pose information, and a position of the target access device is adjusted according to the positioning result to eliminate the position deviation.
  • the step of extracting the logo image and the preset logo points of the visual logo in the logo image from the target image includes:
  • An identification image composed of the coverage area is determined, and preset identification points of the visual identification are extracted from the coverage area.
  • the step of identifying the coverage area of the visual identification in the binarized image includes:
  • a visually identified coverage area in the binarized image is identified.
  • the identification band is composed of multiple visual identifications; the calibration parameters include internal references;
  • the step of calculating the pose information of the identification tape relative to the image sensor based on the calibration parameters, the identification image and the preset identification points includes:
  • the calibration parameters include internal references
  • the step of calculating the pose information of the identification tape relative to the image sensor based on the calibration parameters, the identification image and the preset identification points includes:
  • a difference between the sixth coordinate and the seventh coordinate is obtained to obtain a re-projection error, and the pose information is adjusted with the re-projection error being smaller than a preset threshold as a constraint condition.
  • the calibration parameters also include distortion parameters; the method further includes:
  • de-distortion processing is performed on the logo image to obtain the processed logo image.
  • the calibration parameters also include external parameters; multiple identification tapes are arranged on the target shelf;
  • the step of determining the positioning result of the target shelf according to the pose information it also includes:
  • the current form of the target shelf is obtained.
  • the identification tape is in the shape of a rectangle, and the identification tape is arranged on the crossbeam of the target shelf; the target shelf includes a plurality of crossbeams, and each crossbeam is arranged with The identification tape.
  • the shelf includes shelf legs, and the target access device includes a distance measuring device;
  • the method also includes:
  • the target access device calculates the running distance from the rack legs on both sides of the target aisle through the distance measuring device, and adjusts its position in the target aisle based on the running distance.
  • the shelf includes a shielding mark
  • the target access device includes a distance measuring device
  • the target access device calculates the running distance from the occlusion marks on both sides of the target roadway through the distance measuring device, and adjusts its position in the target roadway based on the running distance.
  • the navigation method based on the shelf identification provided by the embodiment of the present application is applied to a storage system, and the system includes a control server, an inventory area, an access device, and a shelf, wherein shelves are set in the inventory area, and a supply chain is formed between the shelves.
  • the aisle where the access device moves, the shelf is provided with a shelf mark extending along the aisle;
  • the control server determines the target storage position and the target access device on the target shelf, and sends a movement instruction to the target access device;
  • the target access device moves to the target storage position according to the shelf identifier in the target aisle corresponding to the target shelf.
  • the method of setting shelf marks on the identification tape reduces the complexity of construction and improves the construction efficiency.
  • FIG. 8 shows a schematic structural diagram of a shelf positioning device provided by an embodiment of the present invention. As shown in Figure 8, the device includes:
  • the acquisition module 820 is configured to acquire the target image collected by the image sensor for the target shelf and the calibration parameters of the image sensor, wherein the target shelf is arranged with an identification tape, and the identification tape is composed of visual identification;
  • the extraction module 840 is configured to extract the logo image and the preset logo points of the visual logo in the logo image from the target image;
  • the pose calculation module 860 is configured to calculate the pose information of the marker tape relative to the image sensor based on the calibration parameters, the marker image and the preset marker points;
  • the positioning module 880 is configured to determine the positioning result of the target shelf according to the pose information.
  • the target shelf is arranged with a sign tape composed of visual signs
  • the sign image and the sign are extracted from the target image collected by the image sensor.
  • the preset marker points of the visual marker in the marker image are calculated based on the calibration parameters, the marker image and the preset marker points, and the pose information of the marker tape relative to the image sensor is calculated, and the positioning result of the target shelf is determined according to the pose information.
  • the position of the logo tape relative to the image sensor is calculated. Since the identification tape is pre-arranged on the target shelf and is composed of visual signs, the positioning result of the target shelf can be obtained. Using the identification tape to locate the target shelf avoids signal interference in the actual scene, thereby improving the accuracy of the shelf positioning result.
  • the extraction module 840 includes: a binarization processing unit, an identification unit and an extraction unit;
  • a binarization processing unit configured to perform binarization processing on the target image to obtain a binarization image
  • a recognition unit configured to recognize coverage areas of visual signs in the binarized image
  • the extraction unit is configured to determine the identification image composed of the coverage area, and extract the preset identification points of the visual identification from the coverage area.
  • the identification unit is further configured to perform boundary detection on the binarized image, determine boundary information of the binary image; based on the boundary information, identify the coverage area of the visual mark in the binary image.
  • the identification band is composed of multiple visual identifications; the calibration parameters include internal references;
  • the pose calculation module 860 is further configured to use the internal reference to perform normalization processing on the preset marker points of each visual marker in the marker image to obtain the first coordinates of the preset marker points of each visual marker on the marker image; obtain The second coordinates of the preset logo points of each visual logo in the world coordinate system; according to the first coordinate, calculate the third coordinate of the center point of each visual logo on the logo image; according to the second coordinate, calculate the center of each visual logo The fourth coordinate of the point in the world coordinate system; according to the third coordinate and the fourth coordinate, with the center point of each visual sign satisfying the preset arrangement condition as a constraint condition, calculate the pose information of the sign belt relative to the image sensor.
  • the calibration parameters include internal references
  • the pose calculation module 860 is further configured to use the internal reference to perform normalization processing on the preset marker points to obtain the first coordinates of the preset marker points on the marker image; obtain the first coordinates of the preset marker points in the world coordinate system Two coordinates; according to the first coordinate and the second coordinate, calculate the pose information of the identification tape relative to the image sensor; obtain the fifth coordinate of any point on the identification tape in the world coordinate system and the fifth coordinate of the any point on the identification image Six coordinates; calculate the seventh coordinate of any point based on the fifth coordinate and pose information; make a difference between the sixth coordinate and the seventh coordinate to obtain the reprojection error, and adjust the pose information.
  • the calibration parameters also include distortion parameters;
  • the device also includes: a de-distortion module;
  • the de-distortion module is configured to use the distortion parameters to perform de-distortion processing on the logo image to obtain the processed logo image.
  • the calibration parameters also include external parameters; multiple identification bands are arranged on the target shelf; the shelf positioning device also includes: a shape estimation module;
  • the shape estimation module is configured to calculate the rotation angle of each identification band around the rotation axis in the world coordinate system according to the pose information of each identification band relative to the image sensor and the preset rotation matrix;
  • the pose information and external parameters of the image sensor are used to calculate the pose information of each logo in the world coordinate system; according to the rotation angle and pose information of any two logos in the world coordinate system, the The curve shape between any two identification bands; according to the curve shape between any two identification bands, the current shape of the target shelf is obtained.
  • the identification tape is in the shape of a rectangle, and the identification tape is arranged on the crossbeam of the target shelf; the target shelf includes a plurality of crossbeams, and the identification tape is arranged on each crossbeam.
  • the above is a schematic solution of a shelf positioning device in this embodiment. It should be noted that the technical solution of the shelf positioning device and the technical solution of the above-mentioned shelf positioning method belong to the same concept, and details of the technical solution of the shelf positioning device that are not described in detail can be found in the description of the technical solution of the above-mentioned shelf positioning method .
  • Fig. 9 shows a structural block diagram of a shelf handling device provided by another embodiment of the present application.
  • Components of the rack handling device 900 include, but are not limited to, an image sensor 910 , a memory 920 and a processor 930 .
  • the processor 930 is connected with the image sensor 910 and the memory 920 through the bus 940, and the database 960 is used for saving data.
  • the rack handling device 900 also includes an access device 950 that enables the rack handling device 500 to communicate via one or more networks 970 .
  • networks include Public Switched Telephone Network (PSTN, Public Switched Telephone Network), Local Area Network (LAN, Local Area Network), Wide Area Network (WAN, Wide Area Network), Personal Area Network (PAN, Personal Area Network) or networks such as the Internet Composition of communication networks.
  • PSTN Public Switched Telephone Network
  • LAN Local Area Network
  • WAN Wide Area Network
  • PAN Personal Area Network
  • Internet Composition of communication networks such as the Internet Composition of communication networks.
  • the access device 950 may include one or more of wired or wireless network interfaces of any type (for example, a Network Interface Card (NIC, Network Interface Card)), such as IEEE802.11 Wireless Local Area Networks (WLAN, Wireless Local Area Networks) Wireless interface, World Interoperability for Microwave Access (Wi-MAX, World Interoperability for Microwave Access) interface, Ethernet interface, Universal Serial Bus (USB, Universal Serial Bus) interface, cellular network interface, Bluetooth interface, near field communication (NFC , Near Field Communication) interface, and so on.
  • NIC Network Interface Card
  • NIC Network Interface Card
  • NIC Network Interface Card
  • Wi-MAX World Interoperability for Microwave Access
  • Ethernet interface Ethernet interface
  • USB Universal Serial Bus
  • USB Universal Serial Bus
  • Bluetooth interface near field communication
  • NFC Near Field Communication
  • the above-mentioned components of the rack handling device 900 and other components not shown in FIG. 9 may also be connected to each other, for example, via a bus. It should be understood that the structural block diagram of the rack handling equipment shown in FIG. 9 is only for the purpose of illustration, rather than limiting the scope of the present invention. Those skilled in the art can add or replace other components as needed.
  • the image sensor 910 is used to collect the target image for the target shelf, and transmits the target image to the processor 530;
  • the processor 530 is used to execute the following computer-executable instructions, which are implemented when the computer-executable instructions are executed by the processor:
  • the target shelf is provided with an identification tape, and the identification tape is composed of a visual identification
  • the logo image and the preset logo points calculate the pose information of the logo tape relative to the image sensor
  • the positioning result of the target shelf is determined.
  • the shelf handling equipment can be controlled to move the target shelf according to the positioning result.
  • An embodiment of the present invention also provides a computer-readable storage medium, which stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, the steps of the above-mentioned shelf positioning method are implemented.
  • the computer instructions include computer program code, which may be in source code form, object code form, executable file or some intermediate form, and the like.
  • the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a removable hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.
  • ROM Read-Only Memory
  • RAM Random Access Memory
  • electrical carrier signal telecommunication signal and software distribution medium, etc.
  • a computer-readable storage medium stores a computer program, and it is characterized in that, when the computer program is executed by a processor, the above-mentioned Steps in the navigation method for shelf identification.

Abstract

一种基于货架标识的导航系统、方法,基于货架标识的导航系统包括控制服务器(110)、库存区域(120)、存取设备(140)、货架(130),其中,库存区域(120)中设置有货架(130),货架(130)间形成了供存取设备移动的巷道(1201),货架(130)上设置有沿巷道(1201)延伸的货架标识(1301);控制服务器(110),被配置为确定目标货架上的目标存储位和目标存取设备,并向目标存取设备发送移动指令;目标存取设备,被配置为响应于移动指令,在目标货架对应的目标巷道中,根据货架标识(1301)移动至目标存储位。在目标存取设备在目标巷道移动的过程中,根据货架上的货架标识(1301)来实现定精准导航定位的效果,降低了施工成本。

Description

基于货架标识的导航系统、方法
本申请要求于2021年10月15日提交中国专利局、申请号为202111205706.8、发明名称为“货架定位方法、装置、货架搬运设备及存储介质”,2021年10月29日提交中国专利局、申请号为202111274173.9、发明名称为“基于货架标识的导航系统、方法”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本申请涉及仓储技术领域,特别涉及基于货架标识的导航系统、方法。
背景技术
随着电子商务的快速发展,电子商务在消费者生活中扮演着越来越重要的角色,每年用户订单数量呈几何倍数增长,同时电子商务越来越成熟,用户对电子商务的服务质量也提出了更高的要求,仓储管理的处理能力也在经受严峻的考验。
在仓储管理中,货箱机器人完成货箱的入库和出库工作,要实现货箱精准取放需要保证机器人在仓库全局精准定位和仓储货箱位的精确定位。通常机器人在仓库地图的定位中使用地面粘贴二维码,通过识别地面二维码相对偏差对机器人位置进行二次精准校正,货箱中心和机器人上的二维码中心对其,实现货箱精准定位。但是都在地面粘贴二维码的方式,粘贴二维码的点不好定位,地面二维码比较密集,工作量较大,一张张粘贴二维码的精准度和牢固性都比较差,在实际应用中会导致导航的偏差。
发明内容
有鉴于此,本申请实施例提供了基于货架标识的导航系统、方法,以解决现有技术中存在的技术缺陷。
根据本申请实施例的第一方面,提供了一种基于货架标识的导航系统,所述系统包括控制服务器、库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架间形成了供存取设备移动的巷道,货架上设置有沿巷道延伸的货架标识;
所述控制服务器,被配置为确定目标货架上的目标存储位和目标存取设备,并向所述目标存取设备发送移动指令;
所述目标存取设备,被配置为响应于所述移动指令,在所述目标货架对应的目标巷道中,根据货架标识移动至所述目标存储位。
可选的,所述货架上设置有标识带,所述货架标识设置在所述标识带上。
可选的,多个所述货架标识根据预设规则间隔设置在一条连续的所述标识带上。
可选的,多个所述货架标识等间距地设置在一条连续的所述标识带上。
可选的,所述货架标识包括二维码或除二维码外的其他标识。
可选的,所述系统还包括高速通道,所述高速通道与所述库存区域相邻,所述高速通道设置有地面标识;
所述目标存取设备,还被配置为在所述高速通道中根据地面标识移动至所述目标巷道的巷道口。
可选的,所述货架包括至少一层货架横梁,所述货架标识设置在每层货架横梁上。
可选的,所述目标存取设备,进一步被配置为根据任意一层货架横梁上的货架标识移动至所述目标存储位。
可选的,所述目标存取设备,进一步被配置为根据距离地面不大于预设高度阈值的货架标识移动至所述目标存储位。
可选的,所述目标存取设备,还被配置为根据货架标识从所述目标存储位中取出容器或将容器放置在所述目标存储位中。
可选的,所述目标存取设备包括传感器;
所述目标存取设备,还被配置为在从所述目标存储位中取放容器之前,根据所述传感器确定所述目标存取设备与对应货架标识的位置偏差,调整所述目标存取设备的位置,消除所述位置偏差。
可选的,所述传感器包括图像传感器,所述目标存取设备,还被配置为获取图像传感器针对目标货架采集的目标图像及所述图像传感器的标定参数,所述目标货架上布置有标识带,所述标识带由视觉标识组成;从所述目标图像中提取标识图像及所述标识图像中视觉标识的预设标识点;基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息;根据所述位姿信息,确定所述目标货架的定位结果。
根据所述定位结果,调整所述目标存取设备的位置,消除所述位置偏差。
可选的,所述目标存取设备,还被配置为:对所述目标图像进行二值化处理,得到二值化图像;识别所述二值化图像中视觉标识的覆盖区域;确定所述覆盖区域组成的标识图像,并从所述覆盖区域中提 取所述视觉标识的预设标识点。
可选的,所述目标存取设备,还被配置为:对所述二值化图像进行边界检测,确定所述二值化图像的边界信息;基于所述边界信息,识别所述二值化图像中视觉标识的覆盖区域。
可选的,所述标识带由多个视觉标识组成;所述标定参数包括内参;
所述目标存取设备,还被配置为:利用所述内参,对所述标识图像中各视觉标识的预设标识点进行归一化处理,得到所述各视觉标识的预设标识点在所述标识图像上的第一坐标;获得所述各视觉标识的预设标识点在世界坐标系下的第二坐标;根据所述第一坐标,计算所述各视觉标识的中心点在所述标识图像上的第三坐标;根据所述第二坐标,计算所述各视觉标识的中心点在世界坐标系下的第四坐标;根据所述第三坐标和所述第四坐标,以所述各视觉标识的中心点满足预设排布条件为约束条件,计算所述标识带相对于所述图像传感器的位姿信息。
可选的,所述标定参数包括内参;
所述目标存取设备,还被配置为:利用所述内参,对所述预设标识点进行归一化处理,得到所述预设标识点在所述标识图像上的第一坐标;获得所述预设标识点在世界坐标系下的第二坐标;根据所述第一坐标和所述第二坐标,计算所述标识带相对于所述图像传感器的位姿信息;获取所述标识带上的任一点在世界坐标系下的第五坐标以及所述任一点在所述标识图像上的第六坐标;根据所述第五坐标及所述位姿信息,计算所述任一点的第七坐标;将所述第六坐标与所述第七坐标作差,得到重投影误差,以所述重投影误差小于预设阈值为约束条件,调整所述位姿信息。
可选的,所述标定参数还包括畸变参数;
所述目标存取设备,还被配置为:利用所述畸变参数,对所述标识图像进行去畸变处理,得到处理后的所述标识图像。
可选的,所述标定参数还包括外参;所述目标货架上布置有多个标识带;
所述目标存取设备,还被配置为:根据每个标识带相对于所述图像传感器的位姿信息及预设旋转矩阵,计算所述每个标识带在世界坐标系下绕旋转轴的旋转角度;根据所述每个标识带相对于所述图像传感器的位姿信息及所述外参,计算所述每个标识带在世界坐标系下的位姿信息;根据任两个标识带在世界坐标系下的旋转角度及位姿信息,拟合出所述任两个标识带间的曲线形状;根据所述任两个标识带间的曲线形状,获得所述目标货架的当前形态。
可选的,所述标识带为长方形形状,所述标识带设置在所述目标货架的横梁上;所述目标货架包括多个横梁,每个横梁上均布置有所述标识带。
可选的,所述货架包括货架腿,所述目标存取设备包括测距装置;
所述目标存取设备,还被配置为通过所述测距装置计算与所述目标巷道两侧货架腿的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
可选的,所述货架底部设置有遮挡标识,所述目标存取设备包括测距装置;
所述目标存取设备,还被配置为通过所述测距装置计算与所述目标巷道两侧遮挡标识的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
可选的,所述遮挡标识包括遮挡挡板。
可选的,所述遮挡标识围绕所述货架底部设置。
可选的,所述测距装置包括测距雷达或视觉传感器。
根据本申请实施例的第二方面,提供了一种基于货架标识的导航系统,所述系统包括库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架间形成了供存取设备移动的巷道,货架上设置有遮挡标识,所述存取设备包括测距装置;
所述存取设备,被配置为通过所述测距装置计算与所述巷道两侧遮挡标识的运行距离或者与巷道两侧的货架的货架腿的距离,并基于所述运行距离调整其在所述巷道中的位置。
可选的,所述遮挡标识围绕所述货架底部设置。
可选的,所述遮挡标识包括遮挡挡板。
根据本申请实施例的第三方面,提供了一种基于货架标识的导航方法,所述方法应用于仓储系统,所述系统包括控制服务器、库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架间形成了供存取设备移动的巷道,货架上设置有沿巷道延伸的货架标识;
所述方法包括:
所述控制服务器确定目标货架上的目标存储位和目标存取设备,并向所述目标存取设备发送移动指令;
所述目标存取设备响应于所述移动指令,在所述目标货架对应的目标巷道中,根据货架标识移动至所述目标存储位。
可选的,所述系统还包括高速通道,所述高速通道与所述库存区域相邻,所述高速通道设置有地面 标识;
所述方法还包括:
所述目标存取设备在所述高速通道中根据地面标识移动至所述目标巷道的巷道口。
可选的,所述货架包括至少一层货架横梁,所述货架标识设置在每层货架横梁上;
所述目标存取设备根据货架标识移动至所述目标存储位,包括:
所述目标存取设备根据任意一层货架横梁上的货架标识移动至所述目标存储位。
可选的,所述目标存取设备根据任意一层货架横梁上的货架标识移动至所述目标存储位,包括:
所述目标存取设备根据距离地面不大于预设高度阈值的货架标识移动至所述目标存储位。
可选的,所述方法还包括:
所述目标存取设备根据货架标识从所述目标存储位中取出容器或将容器放置在所述目标存储位中。
可选的,所述目标存取设备包括传感器;
所述方法还包括:
所述目标存取设备在从所述目标存储位中取放容器之前,根据所述传感器确定所述目标存取设备与对应货架标识的位置偏差,调整所述目标存取设备的位置,消除所述位置偏差。
可选的,所述货架包括货架腿,所述目标存取设备包括测距装置;
所述方法还包括:
所述目标存取设备通过所述测距装置计算与所述目标巷道两侧货架腿的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
可选的,所述货架包括遮挡标识,所述目标存取设备包括测距装置;
所述目标存取设备通过所述测距装置计算与所述目标巷道两侧遮挡标识的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
可选的,所述传感器包括图像传感器,所述根据所述传感器确定所述目标存取设备与对应货架标识的位置偏差,调整所述目标存取设备的位置,消除所述位置偏差,包括:
获取所述图像传感器针对目标货架采集的目标图像及所述图像传感器的标定参数,所述目标货架上布置有标识带,所述标识带由视觉标识组成;
从所述目标图像中提取标识图像及所述标识图像中视觉标识的预设标识点;
基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息;
根据所述位姿信息,确定所述目标货架的定位结果,根据所述定位结果,调整所述目标存取设备的位置,消除所述位置偏差。
可选的,所述从所述目标图像中提取标识图像及所述标识图像中视觉标识的预设标识点的步骤,包括:
对所述目标图像进行二值化处理,得到二值化图像;
识别所述二值化图像中视觉标识的覆盖区域;
确定所述覆盖区域组成的标识图像,并从所述覆盖区域中提取所述视觉标识的预设标识点。
可选的,所述识别所述二值化图像中视觉标识的覆盖区域的步骤,包括:
对所述二值化图像进行边界检测,确定所述二值化图像的边界信息;
基于所述边界信息,识别所述二值化图像中视觉标识的覆盖区域。
可选的,所述标识带由多个视觉标识组成;所述标定参数包括内参;
所述基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息的步骤,包括:
利用所述内参,对所述标识图像中各视觉标识的预设标识点进行归一化处理,得到所述各视觉标识的预设标识点在所述标识图像上的第一坐标;
获得所述各视觉标识的预设标识点在世界坐标系下的第二坐标;
根据所述第一坐标,计算所述各视觉标识的中心点在所述标识图像上的第三坐标;
根据所述第二坐标,计算所述各视觉标识的中心点在世界坐标系下的第四坐标;
根据所述第三坐标和所述第四坐标,以所述各视觉标识的中心点满足预设排布条件为约束条件,计算所述标识带相对于所述图像传感器的位姿信息。
可选的,所述标定参数包括内参;
所述基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息的步骤,包括:
利用所述内参,对所述预设标识点进行归一化处理,得到所述预设标识点在所述标识图像上的第一坐标;
获得所述预设标识点在世界坐标系下的第二坐标;
根据所述第一坐标和所述第二坐标,计算所述标识带相对于所述图像传感器的位姿信息;
获取所述标识带上的任一点在世界坐标系下的第五坐标以及所述任一点在所述标识图像上的第六坐标;
根据所述第五坐标及所述位姿信息,计算所述任一点的第七坐标;
将所述第六坐标与所述第七坐标作差,得到重投影误差,以所述重投影误差小于预设阈值为约束条件,调整所述位姿信息。
可选的,所述标定参数还包括畸变参数;所述方法还包括:
利用所述畸变参数,对所述标识图像进行去畸变处理,得到处理后的所述标识图像。
可选的,所述标定参数还包括外参;所述目标货架上布置有多个标识带;
在所述根据所述位姿信息,确定所述目标货架的定位结果的步骤之后,还包括:
根据每个标识带相对于所述图像传感器的位姿信息及预设旋转矩阵,计算所述每个标识带在世界坐标系下绕旋转轴的旋转角度;
根据所述每个标识带相对于所述图像传感器的位姿信息及所述外参,计算所述每个标识带在世界坐标系下的位姿信息;
根据任两个标识带在世界坐标系下的旋转角度及位姿信息,拟合出所述任两个标识带间的曲线形状;
根据所述任两个标识带间的曲线形状,获得所述目标货架的当前形态。
可选的,所述标识带为长方形形状,所述标识带设置在所述目标货架的横梁上;所述目标货架包括多个横梁,每个横梁上均布置有所述标识带。
根据本申请实施例的第四方面,提供了一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,其特征在于,所述计算机程序被处理器执行时实现上述基于货架标识的导航方法的步骤。
本申请实施例提供的基于货架标识的导航系统,所述系统包括控制服务器、库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架间形成了供存取设备移动的巷道,货架上设置有沿巷道延伸的货架标识;所述控制服务器,被配置为确定目标货架上的目标存储位和目标存取设备,并向所述目标存取设备发送移动指令;所述目标存取设备,被配置为响应于所述移动指令,在所述目标货架对应的目标巷道中,根据货架标识移动至所述目标存储位。在目标存取设备在目标巷道移动的过程中,根据货架上的货架标识来实现定精准导航定位的效果,降低了施工成本。
附图说明
图1是本申请实施例提供的仓储系统的系统结构示意图;
图2是本申请实施例提供的货架的正视图;
图3是本申请实施例提供的一种货架定位方法的流程图;
图4是本申请一个实施例提供的另一种货架定位方法的流程图;
图5是本申请一个实施例提供的又一种货架定位方法的流程图;
图6是本申请另一实施例提供的货架的示意图;
图7是本申请实施例提供的基于货架标识的导航方法的流程图;
图8是本申请实施例提供的一种货架定位装置的结构示意图;
图9是本申请一个实施例提供的一种货架搬运设备的结构框图。
附图标记
110-控制服务器,111-处理器,112-存储器,113-订单池,114-操作台,120-库存区域,1201-巷道,130-货架,1301-货架标识,1302-容器,1303-遮挡标识,140-存取设备,150-高速通道。
具体实施方式
在下面的描述中阐述了很多具体细节以便于充分理解本申请。但是本申请能够以很多不同于在此描述的其它方式来实施,本领域技术人员可以在不违背本申请内涵的情况下做类似推广,因此本申请不受下面公开的具体实施的限制。
在本申请一个或多个实施例中使用的术语是仅仅出于描述特定实施例的目的,而非旨在限制本申请一个或多个实施例。在本申请一个或多个实施例和所附权利要求书中所使用的单数形式的“一种”、“所述”和“该”也旨在包括多数形式,除非上下文清楚地表示其他含义。还应当理解,本申请一个或多个实施例中使用的术语“和/或”是指包含一个或多个相关联的列出项目的任何或所有可能组合。
应当理解,尽管在本申请一个或多个实施例中可能采用术语第一、第二等来描述各种信息,但这些信息不应限于这些术语。这些术语仅用来将同一类型的信息彼此区分开。例如,在不脱离本申请一个或多个实施例范围的情况下,第一也可以被称为第二,类似地,第二也可以被称为第一。取决于语境,如在此所使用的词语“如果”可以被解释成为“在……时”或“当……时”或“响应于确定”。
货架搬运设备是一种能够搬运货架的自动化设备。
二值化处理:将图像上的像素点的灰度值设置为0或255的技术,也就是将整个图像呈现出明显的只有黑和白的视觉效果。
CCD相机:CCD是电荷耦合器件(Charge Coupled Device)的简称,能够将光线变为电荷并将电荷存储及转移,也可将存储的电荷取出使电压发生变化。CCD相机具有体积小、重量轻、不受磁场影响、具有抗震动和撞击之特性而被广泛应用。
图像传感器的标定:在图像测量过程以及机器视觉应用中,为确定空间物体表面某点的三维几何位置与其在图像中对应点之间的相互关系,必须建立图像传感器成像的几何模型,这些几何模型参数就是图像传感器的参数。在大多数条件下这些参数必须通过实验与计算才能得到,这个求解参数的过程就称之为标定。标定方法主要有:传统标定法、主动视觉标定方法、自标定法、零失真标定法。
视觉标识:具有特殊识别特征的图案,包括二维码、特殊几何图案、特殊标识或者名牌等。
在仓储系统中,容器存取设备完成容器的入库和出库工作,要实现容器精准取放,需要保证机器人在仓库全局精准定位和容器存储位的精准定位,目前存取设备在仓储区域中的定位通常是地面粘贴二维码,通过识别地面二维码相对偏差对存取设备位置进行二次精准校正,在存取容器时,通过容器中心和货架上的二维码中心对齐,实现货箱精准定位,但是在地面粘贴的二维码比较密集,工作量较大,货架上的二维码粘贴工作量同样较大,粘贴精度和牢固性通常比较差。
基于此,在本申请中,提供了基于货架标识的导航系统、方法,在下面的实施例中逐一进行详细说明。
参见图1,图1示出了本申请实施例提供的仓储系统的系统结构示意图,该系统包括控制服务器110、库存区域120、货架130、存取设备140,其中,所述库存区域120中设置有货架130,货架上设置有货架标识,货架130间包括供存取设备移动的巷道1201。
控制服务器110与存取设备140无线通信连接,工作人员通过操作台114操作控制服务器110工作,所述控制服务器110为在服务器上运行的、具有数据存储、信息处理能力的软件系统,可以通过无线与存取设备140连接,控制服务器110可以包括一个或多个服务器,可以为集中式控制架构或者分布式计算架构,控制服务器110具有处理器111和存储器112,在存储器112中可以具有订单池113,订单池113中存有订单信息。
所述库存区域120可以为密集存储区,也可以为非密集存储区,在库存区域120中设置有货架130,货架130之间存有供存取设备140移动的巷道1201,货架130包括至少一层货架横梁,货架横梁将货架130分割为至少一层,货架的货架横梁上设置有存储位,每个存储位可承载可容纳仓储货物的容器,例如料箱。
货架上还设置有货架标识,参见图2,图2示出了本申请一实施例提供的货架的示意图。货架130上设置有货架标识1301,具体的,货架130包括至少一层货架横梁,货架标识1301设置在货架130的货架横梁上。同时在货架横梁上放置有可容纳仓储货物的容器1302。
货架之间的通道称为巷道1201,存取设备140可以在巷道1201内移动。存取设备140具体的可以是货箱存取机器人、堆垛机、叉车等。存取设备用于从货架中取出容器或将容器放入到货架的存储位中。
在本申请提供的一具体实施方式中,控制服务器被配置为确定目标货架上的目标存储位和目标存取设备,并向所述目标存取设备发送移动指令。
目标存取设备,被配置为响应于所述移动指令,在所述目标货架对应的目标巷道中,根据货架标识移动至目标存储位。
具体的,控制服务器110可以根据目标业务任务确定目标货架上的目标存储位,具体业务任务可以是拣选任务、上货任务、盘点任务等等。目标存储位具体可以是需要取出容器的存储位或需要将容器放入的存储位。在库存区域120中,每一排货架可能是由多个货架组成,而目标存储位可能是其中一个目标货架130上的一个存储位。控制服务器还被配置为确定用于执行目标业务任务的目标存取设备140,并向目标存取设备140发送移动指令,以使目标存取设备140移动至目标货架的目标存储位处。
目标存取设备140,被配置为响应于所述移动指令,在目标货架130对应的目标巷道1201中移动时,根据货架上的货架标识进行定位,移动至目标货架的目标存储位处。
在存储区域的巷道中无需设置地面标识,存取设备140在巷道中移动时,通过货架上的货架标识进行定位导航,在实际应用中,货架标识可以位于货架的每个货架横梁侧面,利用货架上的货架标识协助存取设备进行定位导航,减少了在巷道中设置导航标识的工作,减轻了工作量。
在货架上粘贴货架标识可以协助存取设备导航,但是如果货架标识是独立粘贴,工作量也比较大,虽然有货架横梁做参考,也可能会有一定误差,基于此,还可以在货架上设置标识带,所述货架标识设置在所述标识带上,具体的,多个所述货架标识根据预设规则间隔设置在一条连续的所述标识带上,预设的规则可以是等距设置,也可以是非等距设置,在非等距设置的情况下,只需要保证各个货架标识之间的距离是有规律的即可。优选的,多个所述货架标识等间距地设置在一条连续的所述标识带上。在实 际应用中标识带可以通过粘贴的形式固定在货架上,也可以通过喷涂、绘制等方式固定在货架上。
货架标识在实际应用中,可以用于定位货箱,同时也可以用于为存取设备提供导航定位,不同型号的容器混合存放,可以对应不同的货架标识,如图2所示,一个货架标识可以对应一个容器位,也可以两个或多个货架标识对应一个容器位。在实际应用中,控制服务器中已经存储有每个货架标识对应的容器,可以直接给存取设备发送设备存取指令,存取对应的目标容器。
在实际应用中,货架标识可以是定位二维码,也可以是除定位二维码之外的其他标识,例如条形码、数字序号等等。货架标识可以是以粘贴的形式粘贴到货架的货架横梁侧面,也可以是通过印刷、喷涂等方式固定在货架的货架横梁侧面。
在本申请提供的一具体实施方式中,以条码带上用数字序号作为货架标识为例,对于连续相邻的货架之间的货架标识也可以使用一条标识带做连续编号,例如一个货架可以粘贴6个货架标识,相邻的两个货架可以用1-12作为货架标识。为了更便于存取设备根据货架标识定位,优选的,可以根据预设间距将标识带上的货架标识做间隔设置,例如货架的宽度为200厘米,可以设置200厘米长的标识带,在标识带上每隔40厘米设置一个货架标识。
在本申请提供的另一具体实施方式中,参见图1,在仓储系统中还包括有高速通道150,高速通道150与存储区域120相邻,存取设备通过高速通道150在存储区域120和工作站之间移动,将存储区域中的容器搬运至工作站,或从工作站将容器搬运回存储区域。
在高速通道中没有货架130,存取设备无法根据货架标识进行定位,因此存取设备在高速通道150中还需要根据设置在地面的地面标识进行导航定位,即存取设备在高速通道150中移动时是根据高速通道150中的地面标识进行导航,当导航至目标巷道1201的巷道口时,转入到目标巷道1201后,根据货架130上的货架标识进行导航。通过高速通道中150的地面标识和巷道中的货架标识可以为存取设备提供导航定位,指引存取设备140抵达目标货架130,减少了在巷道中粘贴地面标识的工作量。
在本申请提供的一具体实施方式中,存取设备140包括有扫描装置,存取设备通过扫描装置扫描货架标识来实现导航定位,具体的,由于货架中包括至少一层货架横梁,货架标识设置在每层货架横梁上,因此存取设备可以根据任意一层货架横梁上的货架标识来进行定位导航,进而移动至目标存储位。进一步的,当存取设备在移动的过程的中,还可以扫描距离地面高度不大于预设高度阈值的货架标识来进行导航定位,可以有效降低存取设备的重心,提高存取设备的移动速度和安全性。即当存取设备进入到目标巷道之后,将存取设备的扫描装置对准不高于预设高度阈值的货架标识,来进行导航定位。
当存取设备140为容器搬运机器人的情况下,容器搬运机器人设置有取放容器装置,取放容器装置用于从货架中取出容器,或将容器放入到货架130中。容器搬运机器人的扫描装置设置在取放容器装置上,容器搬运机器人可以通过扫描装置扫描货架标识来存入或取出容器,具体的,当容器搬运机器人进入到目标巷道后,将扫描装置对准距离地面高度不大于预设高度阈值的货架标识执行扫描操作,获取存取装置的当前位置,将取放容器装置降低到预设高度阈值之下,可以有效降低存取设备的重心,提高存取设备的移动速度和安全性。当容器搬运机器人移动至目标货架前时,再通过调整取放容器装置的高度,将取放容器装置移动至目标存储位处。即保证了容器搬运机器人的移动速度,又保证了容器搬运机器人在移动过程中的安全性和稳定性,提高了仓储系统的存储拣选效率。
当存取设备140移动至目标货架前,并将取放容器装置移动至目标存储位处的情况下,可以从目标存储位中取出容器,或者将容器放入到目标存储位中,以完成控制服务器110的目标业务任务。
在本申请提供的一具体实施方式中,存取设备包括左右轮,在左右轮电机中都安装有编码器,通过驱动电机上的码盘可以得到存取设备的里程信息,通过存取设备内的陀螺仪可以得到角度信息,结合陀螺仪和里程信息可以推导计算出存取设备的相对位置,但是由于轮子打滑、地面不平整、陀螺仪零位漂移等因素,存取设备的相对位置会存在累积误差,因此,在高速通道150区域内可以通过地面粘贴的地面标识作为绝对位置和存取设备的相对位置进行校正,获得存取设备的最优位置信息,在巷道中移动时,可以通过货架标识作为绝对位置和存取设备的相对位置进行校正,计算存取设备在巷道中的最优位置信息。
存取设备在高速通道中移动至目标巷道的巷道口后,在进入目标巷道前,可以将扫描装置的取放容器装置高度移动至根据距离地面的高度不大于预设高度阈值的高度,例如,将扫描装置取放容器装置的高度调整到货架上在沿竖直方向中高度最低的货架横梁,使得取放容器装置上的扫描装置可以对准货架横梁上的货架标识。
所述目标存取设备包括传感器;所述目标存取设备,还被配置为在从所述目标存储位中取放容器之前,根据所述传感器确定所述目标存取设备与对应货架标识的位置偏差,调整所述目标存取设备的位置,消除所述位置偏差。
其中,所述传感器可以为视觉传感器、深度信息传感器或3D传感器。
当存取设备为容器搬运机器人的情况下,存取设备通过传感器扫描货架横梁上的货架标识,移动至 目标货架前,再通过调整取放容器装置的高度使得取放容器装置移动至目标存储位处。在调整取放容器装置高度的过程中,也可以通过沿竖直方向上,各个货架横梁上的货架标识来定位取放容器装置的高度,当取放容器装置抵达目标存储位后,在从目标存储位中取放容器之前,还需要通过传感器确定取放容器装置与目标存储位对应的货架标识之间的位置偏差,通过调整取放容器装置的位置对取放容器装置与货架标识的相对位置进行校正,以消除取放容器装置与货架标识之间的位置偏差。
为了消除取放容器装置与货架标识之间的位置偏差,需要对目标货架进行精准地定位,获得目标货架的定位结果,并基于该定位结果消除取放容器装置与货架标识之间的位置偏差。
基于此,本申请提供的一具体实施方式中,为获得准确地货架定位结果,提供了一种货架定位方法,可参见图3,图3示出了本发明一个实施例提供的一种货架定位方法的流程图,具体包括以下步骤。
步骤302,获取图像传感器针对目标货架采集的目标图像及图像传感器的标定参数,其中,目标货架上布置有标识带,标识带由视觉标识组成。
在本申请实施例中,还可利用图像传感器实现货架精准定位,其中,图像传感器为能够支持光学影像转化为数字信号并向处理器传输图像的数字信号的传感器,例如CCD相机。图像传感器对视野范围内的场景进行拍摄,并将拍摄得到的目标图像传输至本发明实施例的执行主体处,执行主体获取到图像传感器采集的目标图像。本发明实施例中,图像传感器视野范围内的场景为目标货架,一般情况下,图像传感器仅能拍摄到目标货架的特定部位,该目标货架上布置有标识带(标识带一般为人为地在目标货架上预先粘贴上的),标识带可以粘贴在目标货架的横梁、立杆、底部、顶部等可以被图像传感器采集到的位置,标识带的形状不做具体限定,可以是长方形、圆形、菱形等。在本发明实施例的一种实现方式中,图像传感器安装在货架搬运设备前端,为了方便图像传感器的图像采集,特定部位可以是目标货架底部的横梁。标识带由视觉标识组成,视觉标识可以是二维码、特殊几何图案、特殊标识或者名牌等,也就是说,标识带中至少包括一个识别标识,可以是由多个视觉标识排列组成标识带。
在本申请实施例的一种实现方式中,标识带为长方形形状,标识带设置在目标货架的横梁上。
由于货架的横梁位于货架侧面,能够方便被图像传感器拍摄,而货架的横梁一般是长条形的,因此,使用长方形形状的标识带可以提高定位精度,为了保证定位精度更高,可以设置标识带的尺寸与横梁的尺寸相同。
本申请实施例中,图像传感器是预先标定过的,因此,可以获取到图像传感器的标定参数,具体的标定方式可以是利用传统标定法、主动视觉标定方法、自标定法或零失真标定法对图像传感器进行标定,具体的标定实现过程不是本发明主要讨论的内容,这里不再赘述。
标定参数主要包括内参、外参和畸变参数。其中,内参是一个3x3的参数矩阵,用来反应图像传感器的焦距属性对图像的影响;外参是指图像传感器相对于世界坐标系的设置参数;畸变参数包含径向畸变和切向畸变,是与图像传感器采集图像的畸变程度有关的一组数组,不同图像传感器的畸变参数的格式各不相同。
在图像传感器采集目标图像时,需保证图像传感器在当前安装位置或可移动范围内,视野范围内应包含标识带,且标识带尽可能位于视野范围的中心,为了提升定位的精度,尽可能地包含完整的标识带。
步骤304,从目标图像中提取标识图像及标识图像中视觉标识的预设标识点。
在获取到图像传感器采集的目标图像后,对图像内容进行识别,从中提取出标识图像,标识图像是指图像传感器采集的目标图像中标识带所处区域的图像。具体提取标识图像的方式,可以采用基于神经网络的方式,将图像输入到预先训练的识别模型中,该识别模型可以直接输出图像中的标识图像;提取标识图像的方式,还可以采用基于图像处理的方式,例如对图像进行二值化处理、像素识别等方式。
为了后续的位姿信息计算,需要提取标识图像中视觉标识的预设标识点,预设标识点是指视觉标识上方便提取、具有确定位置关系、且相对包裹视觉标识的点,如方形二维码会提取四个角点、圆形点阵会提取各小圆圆心、三角形多边形或其他几何特征也会提取角点等。
在本申请实施例的一种实现方式中,步骤304具体可以通过如下方式实现:
对目标图像进行二值化处理,得到二值化图像;
识别二值化图像中视觉标识的覆盖区域;
确定覆盖区域组成的标识图像,并从覆盖区域中提取视觉标识的预设标识点。
为了方便将标识带从目标图像中提取出来,可以对目标图像进行二值化处理,二值化处理可以采用自适应二值化,即用来判断目标图像中某一像素点是否是附近区间内较暗或较亮的区域,即将该像素点的像素值与周围区域的像素值的平均值或加权平均值进行对比得到,从而得到二值化图像,二值化图像呈现出明显的只有黑和白的视觉效果,视觉标识的覆盖区域与周围区域会有明显的区别,因此,可以从二值化图像中识别出视觉标识的覆盖区域,进一步地,可以确定覆盖区域组成的标识图像,并且,可以从覆盖区域中提取视觉标识的预设标识点。利用二值化处理方式,提升了提取标识图像的效率和准确度。
在本申请实施例的一种实现方式中,对目标图像进行二值化处理,得到二值化图像的步骤,具体可 以通过如下方式实现:
提取目标图像中的感兴趣区域;
对感兴趣区域进行二值化处理,得到二值化图像。
由于图像采集时,一般设定有一定的采集规则,比如标识带处于图像传感器视野范围的中心附近等,则可以按照这样的采集规则,首先提取出目标图像中的感兴趣区域(ROI,Region of Interest),当然提取目标图像中的ROI区域的方式,除了上述的按照采集规则提取以外,也可以利用基于神经网络的方式进行提取,即将目标图像输入预先训练的神经网络模型,由该神经网络模型识别目标图像中可能有标识带的区域,该区域即为ROI区域。在提取出ROI区域后,对ROI区域进行二值化处理,得到二值化图像。通过对目标图像中的ROI区域进行提取,然后对ROI区域进行二值化处理,ROI区域的范围比原目标图像的范围更小,其更精确地包含标识带,提升了提取标识图像的精度。
在本申请实施例的一种实现方式中,识别二值化图像中视觉标识的覆盖区域的步骤,具体可以通过如下方式实现:
对二值化图像进行边界检测,确定二值化图像的边界信息;
基于边界信息,识别二值化图像中视觉标识的覆盖区域。
边界是指二值化图像中黑白对比较高的分界线,即在边界两侧有较高的黑白对比度,基于此,可以根据黑白对比度的变化情况,对二值化图像进行边界检测,确定出二值化图像的边界信息,边界信息可以包括边界的位置、形状、长度等。基于二值化图像的边界信息,可以得到二值化图像中所有可能存在的线,对这些线进行组合,可以识别出可能是视觉标识的覆盖区域。通过边界检测的方式,确定出视觉标识的覆盖区域,更为方便、快捷地识别出各视觉标识的覆盖区域,从而,提高了提取标识图像的准确度和效率。
在本申请实施例的一种实现方式中,在基于边界信息,识别二值化图像中视觉标识的覆盖区域的步骤之前,利用预设过滤方法,对边界信息进行过滤,得到过滤后的边界信息。
由于在进行边界检测时,有些边界线并不满足实际长度情况,或者某些边界角度不符合实际角度情况,为了排除这些不符合要求的边界,在得到二值化图像的边界信息后,需要对边界信息进行过滤,过滤主要是对上述边界线或线组合的特征进行过滤,如排除长度或方向不满足要求的线、排除某些角度不符合要求的线组合等,具体的过滤方法可根据视觉标识的类型来确定。过滤后的边界信息更加符合实际要求,则提取的标识图像更加符合实际情况,精度更高。
步骤306,基于标定参数、标识图像及预设标识点,计算标识带相对于图像传感器的位姿信息。
在提取出标识图像和视觉标识的预设标识点之后,可以基于标定参数、标识图像和预设标识点,对标识带相对于图像传感器的位姿信息进行计算。具体地,标定参数约定了图像传感器在采集图像时的一些属性,例如所采集图像的畸变情况、焦距等等,并且标识图像上视觉标识的预设标识点是一些方便提取、具有确定位置关系、且相对包裹视觉标识的点,则可基于标定参数、标识图像和预设标识点,计算出标识带相对于图像传感器的位姿信息。
具体地,对标识图像中视觉标识的预设标识点利用内参进行归一化处理,可以得到视觉标识的预设标识点在标识图像上的第一坐标(u,v,1),同时,因为视觉标识是预先设计好的,所以可以获得预设标识点在世界坐标系下的第二坐标(x,y,z)。因为(u,v,1)事实上是从图像传感器到(x,y,z)连线在像平面上的投影,且视觉标识的预设标识点处于同一平面上,因此,可以根据第一坐标、第二坐标,利用三角关系计算出预设标识点相对于图像传感器的相对位置信息,以及预设标识点所在平面相对于图像传感器的相对姿态信息。
具体的计算过程可以为:令像平面和视觉标识的预设标识点所属平面直接的旋转关系用旋转矩阵R表示,像平面与预设标识点所属平面的原点间的平移关系用位移向量T表示,根据坐标系之间的转换关系,则对第一坐标(u,v,1)和第二坐标(x,y,z)之间有如公式(1)所示关系:
Figure PCTCN2022125539-appb-000001
在标识点的数量充足的前提下,求解公式(1)即可获得R、T,即像平面与预设标识点所属平面间的相对旋转和相对位移。当标识点的数量超过求解公式(1)所需的数量时,可以通过最小二乘法来进行求解。
具体地,在本申请实施例的一种实现方式中,标识带由多个视觉标识组成;标定参数包括内参。相应地,步骤306具体可以通过如下方式实现:
利用内参,对标识图像中各视觉标识的预设标识点进行归一化处理,得到各视觉标识的预设标识点在标识图像上的第一坐标;
获得各视觉标识的预设标识点在世界坐标系下的第二坐标;
根据第一坐标,计算各视觉标识的中心点在标识图像上的第三坐标;
根据第二坐标,计算各视觉标识的中心点在世界坐标系下的第四坐标;
根据第三坐标和第四坐标,以各视觉标识的中心点满足预设排布条件为约束条件,计算标识带相对于图像传感器的位姿信息。
在本实施例中,标识带可以由多个视觉标识组成,对标识图像中各视觉标识的预设标识点利用内参进行归一化处理,就可以得到各视觉标识的预设标识点在标识图像上的第一坐标,同时,因为视觉标识是预先设计好的,所以可以获得各视觉标识的预设标识点在世界坐标系下的第二坐标。进一步地,可以根据第一坐标,计算各视觉标识的中心点在标识图像上的第三坐标,例如视觉标识为长方形的二维码,则预设标识点为四个角点,则角点连线的交叉点即为该视觉标识的中心点,可以根据四个角点的第一坐标计算出中心点的第三坐标。同理,还可以根据第二坐标,计算各视觉标识的中心点在世界坐标系下的第四坐标。然后,根据第三坐标和第四坐标,以各视觉标识的中心点满足预设排布条件为约束条件,计算标识带相对于图像传感器的位姿信息,其中,预设排布条件为标识带中视觉标识的排布情况,一般受标识带的形状影响,例如,如果标识带为长方形,则其中的视觉标识就是行排布,因此预设排布条件就是中心点在同一条直线上,再例如,如果标识带为圆形,则其中的视觉标识就是圆周排布,因此预设排布条件就是中心点在同一个圆的圆周上。具体的计算位姿信息的公式可以为公式(1)。
由于是标识带,原则上来讲,同一标识带上的各视觉标识的中心点应该满足预设排布条件,以此为约束条件对位姿信息进行计算,可以预先排除位姿跳动的情况,保证了位姿信息计算的准确性。
在本申请实施例的另一种实现方式中,标定参数包括内参。相应地,步骤306具体还可以通过如下方式实现:
利用内参,对预设标识点进行归一化处理,得到预设标识点在标识图像上的第一坐标;
获得预设标识点在世界坐标系下的第二坐标;
根据第一坐标和第二坐标,计算标识带相对于图像传感器的位姿信息;
获取标识带上的任一点在世界坐标系下的第五坐标以及该任一点在标识图像上的第六坐标;
根据第五坐标及位姿信息,计算该任一点的第七坐标;
将第六坐标与第七坐标作差,得到重投影误差,以重投影误差小于预设阈值为约束条件,调整位姿信息。
对标识图像中视觉标识的预设标识点利用内参进行归一化处理,就可以得到预设标识点在标识图像上的第一坐标,同时,因为视觉标识是预先设计好的,所以可以获得预设标识点在世界坐标系下的第二坐标。对于可能出现姿态跳动的情况,需要在预设角度跳动可能存在的前提下,进行迭代优化,在优化过程中,图像上像素数更多的部分,能够取得更高的计算精度。长方形的标识带能够进一步提高滚转角和偏航角的精度。
具体地,获取标识带上的任一点在世界坐标系下的第五坐标以及该任一点在标识图像上的第六坐标,根据第五坐标及位姿信息,计算该任一点的第七坐标(第七坐标即为该任一点在当前位姿下的像点坐标),然后将第六坐标与第七坐标作差,得到重投影误差,以重投影误差小于预设阈值为约束条件,调整位姿信息。经过迭代优化,即可得到最优的位姿信息,这个过程也称为反向计算,能够有效应对位姿跳动的情况,保证了位姿信息计算的准确性。
在本申请实施例的一种实现方式中,标定参数还包括畸变参数;该方法还可以包括如下步骤:
利用畸变参数,对标识图像进行去畸变处理,得到处理后的标识图像。
将提取出的标识图像利用畸变参数进行去畸变处理,处理后的标识图像为平面图像,抵消了畸变带来的误差,能够提高最终定位的精度。
步骤308,根据位姿信息,确定目标货架的定位结果。
在得到标识带相对于图像传感器的位姿信息后,由于标识带在目标货架上的设置位置是一定的,则可以根据位姿信息,确定出目标货架的定位结果,目标货架的定位结果,就是根据位姿信息和标识带的预设设置位置,换算出来的目标货架相对于图像传感器的位姿信息。
在本申请实施例的一种实现方式中,标定参数还包括外参;目标货架上布置有多个标识带;
在步骤308,还可以包括如下步骤:
根据每个标识带相对于图像传感器的位姿信息及预设旋转矩阵,计算每个标识带在世界坐标系下绕旋转轴的旋转角度;
根据每个标识带相对于图像传感器的位姿信息及外参,计算每个标识带在世界坐标系下的位姿信息;
根据任两个标识带在世界坐标系下的旋转角度及位姿信息,拟合出该任两个标识带间的曲线形状;
根据任两个标识带间的曲线形状,获得目标货架的当前形态。
在实际应用中,目标货架上可以布置有多个标识带,则在得到每个标识带相对于图像传感器的位姿信息之后,还可以将每个标识带相对于图像传感器的位姿信息进行结合,从而描述出目标货架的形态。
具体地,令目标货架的某一面上的点为
Figure PCTCN2022125539-appb-000002
其中一个点指定就是一个标识带所粘贴的点,点ajk的位姿为Rjk,Tjk,根据预设旋转矩阵,可以得到该点在世界坐标系下绕三个旋转轴的旋转角度。由于获得了这m×n个点相对于图像传感器的位姿信息,在拥有图像传感器相对世界坐标系的外参的前提下,可以转化出各点在世界坐标系下的位姿信息。由于货架是框架结构,所以在任意两点之间,可以根据两点处沿框架方向的旋转角度和两点相对于世界坐标系的位姿信息,拟合出两点间的曲线形状,以此类推,只要将面对图像传感器一面的各点间曲线形状拟合出来,即为目标货架当前面的当前形态。
本实施例增加了目标货架的形态评估功能,能够为货架搬运等提供更为适合的数据支持。
在本发明的一个或多个实施例中,标识带可以为长方形形状,标识带可以设置在目标货架的横梁上;目标货架可以包括多个横梁,每个横梁上均布置有标识带。
应用本发明实施例,通过获取图像传感器针对目标货架采集的目标图像及图像传感器的标定参数,其中目标货架上布置有由视觉标识组成的标识带,从图像传感器采集的目标图像中提取标识图像及标识图像中视觉标识的预设标识点,基于标定参数、标识图像及预设标识点,计算标识带相对于图像传感器的位姿信息,根据该位姿信息,确定目标货架的定位结果。通过对采集的目标图像中的标识图像以及标识图像中视觉标识的预设标识点进行提取,基于图像采集设备的标定参数、标识图像及预设标识点,计算出标识带相对于图像传感器的位姿信息,由于标识带是预先布置在目标货架上的、且由视觉标识组成,从而可以获得目标货架的定位结果。使用标识带来定位目标货架,避免了实际场景中的信号干扰,从而提高了货架定位结果的精度。
基于图3所示实施例,图4示出了本发明一个实施例提供的另一种货架定位方法的流程图,具体包括以下步骤。
步骤402,获取图像传感器针对目标货架采集的目标图像及图像传感器的标定参数,其中,目标货架上布置有标识带,标识带由视觉标识组成。
步骤404,从目标图像中提取标识图像及标识图像中视觉标识的预设标识点。
步骤410,利用预设验证方法,对标识图像中的视觉标识是否符合预设视觉标识特征进行验证。如果验证结果为验证通过,则执行步骤406,否则返回执行步骤404。
步骤406,基于标定参数、标识图像及预设标识点,计算标识带相对于图像传感器的位姿信息。
步骤408,根据位姿信息,确定目标货架的定位结果。
其中,步骤402、404、406、408与图3所示实施例中的各步骤相同,这里不再赘述。
由于提取的标识图像中视觉标识的预设标识点可能由于各种原因不是实际视觉标识的标识点,而是过滤不充分造成的,因此要对标识图像中的视觉标识是否符合预设视觉标识特征进行验证,确认视觉标识是否为预先设置的视觉标识。比如二维码会进行解码、几何特征会进行特征检验、特殊标识会进行图像匹配等,也就是,需要利用预设验证方法对标识图像中的视觉标识是否符合预设视觉标识特征进行验证,只有在验证通过的情况下,才会执行位姿信息计算的步骤。具体的验证方法可以是将标识图像输入对应验证函数或模型当中,确认输出结果是否为验证通过。
应用本发明实施例,通过获取图像传感器针对目标货架采集的目标图像及图像传感器的标定参数,其中目标货架上布置有由视觉标识组成的标识带,从图像传感器采集的目标图像中提取标识图像及标识图像中视觉标识的预设标识点,基于标定参数、标识图像及预设标识点,计算标识带相对于图像传感器的位姿信息,根据该位姿信息,确定目标货架的定位结果。通过对采集的目标图像中的标识图像以及标识图像中视觉标识的预设标识点进行提取,基于图像采集设备的标定参数、标识图像及预设标识点,计算出标识带相对于图像传感器的位姿信息,由于标识带是预先布置在目标货架上的、且由视觉标识组成,从而可以获得目标货架的定位结果。使用标识带来定位目标货架,避免了实际场景中的信号干扰,从而提高了货架定位结果的精度。并且在计算位姿信息之前,先利用预设验证方法对标识图像中的视觉标识是否符合预设视觉标识特征进行验证,只有在验证通过的情况下,才会执行位姿信息计算的步骤,避免了过滤不充分导致的预设标识点不符合实际情况的问题,定位结果更为精准。
为了便于理解,下面结合具体应用场景对本发明实施例提供的货架定位方法进行介绍。在本发明实施例中,货架搬运设备的前端安装有CCD相机,货架搬运设备还包括处理器,由处理器执行如图5所 示的货架定位方法,图5示出了本发明一个实施例提供的又一种货架定位方法的流程图,具体包括以下步骤。
第一步,对CCD相机进行标定,得到CCD相机的内参和畸变参数,将视觉标识在货架特定部位进行布置。
第二步,利用CCD相机采集视场中含有完整标识带的图片。
第三步,对含有标识带的相机ROI区域进行二值化处理,得到二值化图像。
这里的二值化处理为自适应二值化,即用来判断图片中像素点是否是附近区间内较暗或较亮的区域,即将该像素点的像素值与周围区域的像素值的平均值或加权平均值进行对比得到。
第四步,在二值化图像中,进行边界检测,找到二值化图像中所有边界,并进行边界过滤,提取标识图像和标识图像中各视觉标识参与位姿计算的预设标识点。
根据边界两侧黑白对比较明显的规律,对二值化图像进行边界检测,确定出二值化图像中的所有边界,提取边界可以近似理解为提取二值化图像中所有可能存在的线,并对这些线进行组合和过滤,以找出可能是视觉标识的区域。边界过滤主要是对这些线或线组合的特征进行过滤,如排除长度或方向不满足要求的线,或排除某些角度不符合要求的线组合,具体的过滤方法根据视觉标识的类型来确定。根据边界检测的结果,可以从二值化图像中提取出标识图像,并且从标识图像中提取出各视觉标识参与位姿计算的预设标识点,如方形二维码提取四个角点。
第五步,对提取到的标识图像中的各视觉标识进行验证,验证确认后,根据CCD相机的内参和畸变参数、预设标识点计算标识带相对于CCD相机的位姿信息。
由于前一步中提取的预设标识点可能由于各种原因不是实际视觉标识的标识点,而是过滤不充分造成的,因此要对标识图像中各视觉标识的具有唯一性的特殊视觉特征进行验证,确认其是否符合预设视觉标识特征。验证通过后,再根据CCD相机的内参和畸变参数、预设标识点计算标识带相对于CCD相机的位姿信息。位姿信息以坐标的形式输出,输出结果为视觉标识相对CCD相机的位置T和姿态R,位置T即视觉标识相对于CCD相机的位置(x,y,z),姿态R即视觉标识相对于CCD相机的旋转四元数(x,y,z,w)。
在得到视觉标识到CCD相机的位姿信息后,可以换算出目标货架的定位结果,直接将定位结果输出,以使货架搬运设备根据定位结果,准确地搬运货架。
本申请使用视觉标识来进行目标货架的定位。通过对视觉标识进行位姿信息的计算获得目标货架的定位结果,相比射频技术或WLAN技术的定位方案具有更高的精度,同时相对深度学习方案具有更高的适应性。
本实施例提供的货架定位方法包括:获取图像传感器针对目标货架采集的目标图像及图像传感器的标定参数,其中目标货架上布置有由视觉标识组成的标识带,从图像传感器采集的目标图像中提取标识图像及标识图像中视觉标识的预设标识点,基于标定参数、标识图像及预设标识点,计算标识带相对于图像传感器的位姿信息,根据该位姿信息,确定目标货架的定位结果。通过对采集的目标图像中的标识图像以及标识图像中视觉标识的预设标识点进行提取,基于图像采集设备的标定参数、标识图像及预设标识点,计算出标识带相对于图像传感器的位姿信息,由于标识带是预先布置在目标货架上的、且由视觉标识组成,从而可以获得目标货架的定位结果。使用标识带来定位目标货架,避免了实际场景中的信号干扰,从而提高了货架定位结果的精度。
进一步地,在实际应用中,存取设备在库存区域的巷道内移动时,根据货架标识只能计算出存取设备在巷道内的相对位置,无法计算出存取设备与货架间的距离,存取设备转动的角度是由陀螺仪累计计算确定,由于陀螺仪漂移等影响,存取设备的转动的角度会有累计误差,仅仅依靠陀螺仪来确定存取设备的转动角度会存在存取设备撞到货架的风险,因此还需要提供一种解决方案,保证存取设备在巷道中移动时不会撞击到货架。
基于此,在本申请提供的另一具体实施方式中,货架130底部设置有遮挡标识1303,所述目标存取设备包括测距装置;
所述目标存取设备140,还被配置为通过所述测距装置计算与所述目标巷道1201两侧遮挡标识1303的运行距离,并基于所述运行距离调整其在所述目标巷道1201中的位置。
其中,遮挡标识具体可以包括遮挡挡板,测距装置具体可以包括测距雷达或视觉传感器。参见图6,图6示出了本申请一实施例提供的货架的示意图,如图6所示,货架上包括货架横梁,在每个货架横梁的侧面设置有标识带,在标识带上设置有货架标识1301,在货架横梁上放置有容器1302,在货架的底部,货架腿之间设置有遮挡标识1303。
在本申请提供的一具体实施方式中,所述遮挡标识围绕所述货架底部设置,具体的,所述遮挡标识可以设置在货架底部的地面上,可以通过固定装置固定在货架的货架腿间,可以通过固定装置固定在货架底层货架横梁的底部。
除了在货架底部设置有遮挡标识之外,在本申请提供的另一具体实施方式中,还可以根据货架的货架腿来放置存取设备与货架发生碰撞,具体的,目标存取设备,还被配置为通过所述测距装置计算与所述目标巷道两侧货架腿的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
目标存取设备140通过测距装置识别货架的遮挡标识1303或货架腿,并计算目标存取设备140距离遮挡标识1303或货架腿的运行距离,并通过调整该运动距离,保证目标存取设备在目标巷道中移动的过程中始终与货架保持一定安全距离,并按照预设的角度方向移动。
本申请实施例提供的基于货架标识的导航系统,所述系统包括控制服务器、库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架间形成了供存取设备移动的巷道,货架上设置有沿巷道延伸的货架标识;所述控制服务器,被配置为确定目标货架上的目标存储位和目标存取设备,并向所述目标存取设备发送移动指令;所述目标存取设备,被配置为响应于所述移动指令,在所述目标货架对应的目标巷道中,根据货架标识移动至所述目标存储位。在目标存取设备在目标巷道移动的过程中,根据货架上的货架标识来实现定精准导航定位的效果,降低了施工成本。
更进一步的,通过标识带,在标识带上设置货架标识的方式,降低了施工的复杂度,提升了施工效率。
最后,在货架底部设置遮挡标识,可以避免存取设备在库存区域内移动时撞到货架,提高了存取设备运行的安全性。
在本申请提供的一实施例中,提供了另一种基于货架标识的导航系统,所述系统包括库存区域120、存取设备140、货架130,其中,所述库存区域中设置有货架130,货架上设置有遮挡标识1303;
所述存取设备140,被配置为根据所述遮挡标识运行于货架间的巷道中。
更进一步的,所述存取设备140包括测距装置;
所述存取设备140,进一步被配置为通过所述测距装置计算与所述遮挡标识1303的运行距离,在巷道中移动的过程中调整所述运行距离至预设距离区间。
在本申请提供的另一实施例中,所述遮挡标识设置于货架底部。更进一步的,所述遮挡标识包括遮挡挡板。
本申请实施例提供的基于货架标识的导航系统,包括库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架上设置有遮挡标识;所述存取设备,被配置为根据所述遮挡标识运行于货架间的巷道中。可以避免存取设备在库存区域内移动时撞到货架,提高了存取设备运行的安全性。
参见图7,图7示出了本申请一实施例提供的基于货架标识的导航方法的流程图,所述方法应用于仓储系统,所述仓储系统包括控制服务器、库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架间形成了供存取设备移动的巷道,货架上设置有沿巷道延伸的货架标识。
步骤702:控制服务器确定目标货架上的目标存储位和目标存取设备,并向所述目标存取设备发送移动指令。
步骤704:目标存取设备响应于所述移动指令,在所述目标货架对应的目标巷道中,根据货架标识移动至所述目标存储位。
在本申请提供的另一具体实施方式中,所述系统还包括高速通道,所述高速通道与所述库存区域相邻,所述高速通道设置有地面标识;
所述方法还包括:
所述目标存取设备在所述高速通道中根据地面标识移动至所述目标巷道的巷道口。
具体的,目标存取设备在接收到控制服务器下发的移动指令后,在库存区域之外的高速通道中移动时,是通过地面上的地面标识进行导航定位,当存取设备移动至目标货架对应的目标巷道的巷道口时,根据货架上的货架标识进行导航定位。
在本申请提供的另一具体实施方式中,所述货架包括至少一层货架横梁,所述货架标识设置在每层货架横梁上;
所述目标存取设备根据货架标识移动至所述目标存储位,包括:
所述目标存取设备根据任意一层货架横梁上的货架标识移动至所述目标存储位。
所述目标存取设备根据任意一层货架横梁上的货架标识移动至所述目标存储位,包括:
所述目标存取设备根据距离地面不大于预设高度阈值的货架标识移动至所述目标存储位。
在本申请提供的另一具体实施方式中,所述方法还包括:
所述目标存取设备根据货架标识从所述目标存储位中取出容器或将容器放置在所述目标存储位中。
在本申请提供的另一具体实施方式中,所述目标存取设备包括传感器;
所述方法还包括:
所述目标存取设备在从所述目标存储位中取放容器之前,根据所述传感器确定所述目标存取设备与对应货架标识的位置偏差,调整所述目标存取设备的位置,消除所述位置偏差。
在本申请提供的另一具体实施方式中,所述传感器包括图像传感器,所述根据所述传感器确定所述目标存取设备与对应货架标识的位置偏差,调整所述目标存取设备的位置,消除所述位置偏差,包括:
获取所述图像传感器针对目标货架采集的目标图像及所述图像传感器的标定参数,所述目标货架上布置有标识带,所述标识带由视觉标识组成;
从所述目标图像中提取标识图像及所述标识图像中视觉标识的预设标识点;
基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息;
根据所述位姿信息,确定所述目标货架的定位结果,根据所述定位结果,调整所述目标存取设备的位置,消除所述位置偏差。
在本申请提供的另一具体实施方式中,所述从所述目标图像中提取标识图像及所述标识图像中视觉标识的预设标识点的步骤,包括:
对所述目标图像进行二值化处理,得到二值化图像;
识别所述二值化图像中视觉标识的覆盖区域;
确定所述覆盖区域组成的标识图像,并从所述覆盖区域中提取所述视觉标识的预设标识点。
在本申请提供的另一具体实施方式中,所述识别所述二值化图像中视觉标识的覆盖区域的步骤,包括:
对所述二值化图像进行边界检测,确定所述二值化图像的边界信息;
基于所述边界信息,识别所述二值化图像中视觉标识的覆盖区域。
在本申请提供的另一具体实施方式中,所述标识带由多个视觉标识组成;所述标定参数包括内参;
所述基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息的步骤,包括:
利用所述内参,对所述标识图像中各视觉标识的预设标识点进行归一化处理,得到所述各视觉标识的预设标识点在所述标识图像上的第一坐标;
获得所述各视觉标识的预设标识点在世界坐标系下的第二坐标;
根据所述第一坐标,计算所述各视觉标识的中心点在所述标识图像上的第三坐标;
根据所述第二坐标,计算所述各视觉标识的中心点在世界坐标系下的第四坐标;
根据所述第三坐标和所述第四坐标,以所述各视觉标识的中心点满足预设排布条件为约束条件,计算所述标识带相对于所述图像传感器的位姿信息。
在本申请提供的另一具体实施方式中,所述标定参数包括内参;
所述基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息的步骤,包括:
利用所述内参,对所述预设标识点进行归一化处理,得到所述预设标识点在所述标识图像上的第一坐标;
获得所述预设标识点在世界坐标系下的第二坐标;
根据所述第一坐标和所述第二坐标,计算所述标识带相对于所述图像传感器的位姿信息;
获取所述标识带上的任一点在世界坐标系下的第五坐标以及所述任一点在所述标识图像上的第六坐标;
根据所述第五坐标及所述位姿信息,计算所述任一点的第七坐标;
将所述第六坐标与所述第七坐标作差,得到重投影误差,以所述重投影误差小于预设阈值为约束条件,调整所述位姿信息。
在本申请提供的另一具体实施方式中,所述标定参数还包括畸变参数;所述方法还包括:
利用所述畸变参数,对所述标识图像进行去畸变处理,得到处理后的所述标识图像。
在本申请提供的另一具体实施方式中,所述标定参数还包括外参;所述目标货架上布置有多个标识带;
在所述根据所述位姿信息,确定所述目标货架的定位结果的步骤之后,还包括:
根据每个标识带相对于所述图像传感器的位姿信息及预设旋转矩阵,计算所述每个标识带在世界坐标系下绕旋转轴的旋转角度;
根据所述每个标识带相对于所述图像传感器的位姿信息及所述外参,计算所述每个标识带在世界坐标系下的位姿信息;
根据任两个标识带在世界坐标系下的旋转角度及位姿信息,拟合出所述任两个标识带间的曲线形状;
根据所述任两个标识带间的曲线形状,获得所述目标货架的当前形态。
在本申请提供的另一具体实施方式中,所述标识带为长方形形状,所述标识带设置在所述目标货架的横梁上;所述目标货架包括多个横梁,每个横梁上均布置有所述标识带。
在本申请提供的另一具体实施方式中,所述货架包括货架腿,所述目标存取设备包括测距装置;
所述方法还包括:
所述目标存取设备通过所述测距装置计算与所述目标巷道两侧货架腿的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
在本申请提供的另一具体实施方式中,所述货架包括遮挡标识,所述目标存取设备包括测距装置;
所述目标存取设备通过所述测距装置计算与所述目标巷道两侧遮挡标识的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
本申请实施例提供的基于货架标识的导航方法,应用于仓储系统,所述系统包括控制服务器、库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架间形成了供存取设备移动的巷道,货架上设置有沿巷道延伸的货架标识;所述控制服务器确定目标货架上的目标存储位和目标存取设备,并向所述目标存取设备发送移动指令;所述目标存取设备响应于所述移动指令,在所述目标货架对应的目标巷道中,根据货架标识移动至所述目标存储位。在目标存取设备在目标巷道移动的过程中,根据货架上的货架标识来实现定精准导航定位的效果,降低了施工成本。
更进一步的,通过标识带,在标识带上设置货架标识的方式,降低了施工的复杂度,提升了施工效率。
最后,在货架底部设置遮挡标识,可以避免存取设备在库存区域内移动时撞到货架,提高了存取设备运行的安全性。
与上述实施例中提到的货架定位方法实施例相对应,本申请还提供了货架定位装置实施例,图8示出了本发明一个实施例提供的一种货架定位装置的结构示意图。如图8所示,该装置包括:
获取模块820,被配置为获取图像传感器针对目标货架采集的目标图像及图像传感器的标定参数,其中,目标货架上布置有标识带,标识带由视觉标识组成;
提取模块840,被配置为从目标图像中提取标识图像及标识图像中视觉标识的预设标识点;
位姿计算模块860,被配置为基于标定参数、标识图像及预设标识点,计算标识带相对于图像传感器的位姿信息;
定位模块880,被配置为根据位姿信息,确定目标货架的定位结果。
应用本发明实施例,通过获取图像传感器针对目标货架采集的目标图像及图像传感器的标定参数,其中目标货架上布置有由视觉标识组成的标识带,从图像传感器采集的目标图像中提取标识图像及标识图像中视觉标识的预设标识点,基于标定参数、标识图像及预设标识点,计算标识带相对于图像传感器的位姿信息,根据该位姿信息,确定目标货架的定位结果。通过对采集的目标图像中的标识图像以及标识图像中视觉标识的预设标识点进行提取,基于图像采集设备的标定参数、标识图像及预设标识点,计算出标识带相对于图像传感器的位姿信息,由于标识带是预先布置在目标货架上的、且由视觉标识组成,从而可以获得目标货架的定位结果。使用标识带来定位目标货架,避免了实际场景中的信号干扰,从而提高了货架定位结果的精度。
可选地,提取模块840包括:二值化处理单元、识别单元和提取单元;
二值化处理单元,被配置为对目标图像进行二值化处理,得到二值化图像;
识别单元,被配置为识别二值化图像中视觉标识的覆盖区域;
提取单元,被配置为确定覆盖区域组成的标识图像,并从覆盖区域中提取视觉标识的预设标识点。
可选地,识别单元,进一步被配置为对二值化图像进行边界检测,确定二值化图像的边界信息;基于边界信息,识别二值化图像中视觉标识的覆盖区域。
可选地,标识带由多个视觉标识组成;标定参数包括内参;
位姿计算模块860,进一步被配置为利用内参,对标识图像中各视觉标识的预设标识点进行归一化处理,得到各视觉标识的预设标识点在标识图像上的第一坐标;获得各视觉标识的预设标识点在世界坐标系下的第二坐标;根据第一坐标,计算各视觉标识的中心点在标识图像上的第三坐标;根据第二坐标,计算各视觉标识的中心点在世界坐标系下的第四坐标;根据第三坐标和第四坐标,以各视觉标识的中心点满足预设排布条件为约束条件,计算标识带相对于图像传感器的位姿信息。
可选地,标定参数包括内参;
位姿计算模块860,进一步被配置为利用内参,对预设标识点进行归一化处理,得到预设标识点在标识图像上的第一坐标;获得预设标识点在世界坐标系下的第二坐标;根据第一坐标和第二坐标,计算标识带相对于图像传感器的位姿信息;获取标识带上的任一点在世界坐标系下的第五坐标以及该任一点在标识图像上的第六坐标;根据第五坐标及位姿信息,计算该任一点的第七坐标;将第六坐标与第七坐标作差,得到重投影误差,以重投影误差小于预设阈值为约束条件,调整位姿信息。
可选地,标定参数还包括畸变参数;该装置还包括:去畸变模块;
去畸变模块,被配置为利用畸变参数,对标识图像进行去畸变处理,得到处理后的标识图像。
可选地,标定参数还包括外参;目标货架上布置有多个标识带;货架定位装置还包括:形态估计模块;
形态估计模块,被配置为根据每个标识带相对于图像传感器的位姿信息及预设旋转矩阵,计算每个标识带在世界坐标系下绕旋转轴的旋转角度;根据每个标识带相对于图像传感器的位姿信息及外参,计算每个标识带在世界坐标系下的位姿信息;根据任两个标识带在世界坐标系下的旋转角度及位姿信息,拟合出该任两个标识带间的曲线形状;根据任两个标识带间的曲线形状,获得目标货架的当前形态。
可选地,标识带为长方形形状,标识带设置在目标货架的横梁上;目标货架包括多个横梁,每个横梁上均布置有标识带。
上述为本实施例的一种货架定位装置的示意性方案。需要说明的是,该货架定位装置的技术方案与上述的货架定位方法的技术方案属于同一构思,货架定位装置的技术方案未详细描述的细节内容,均可以参见上述货架定位方法的技术方案的描述。
图9示出了本申请另一个实施例提供的一种货架搬运设备的结构框图。该货架搬运设备900的部件包括但不限于图像传感器910、存储器920和处理器930。处理器930与图像传感器910、存储器920通过总线940相连接,数据库960用于保存数据。
货架搬运设备900还包括接入设备950,接入设备950使得货架搬运设备500能够经由一个或多个网络970通信。这些网络的示例包括公用交换电话网(PSTN,Public Switched Telephone Network)、局域网(LAN,Local Area Network)、广域网(WAN,Wide Area Network)、个域网(PAN,Personal Area Network)或诸如因特网的通信网络的组合。接入设备950可以包括有线或无线的任何类型的网络接口(例如,网络接口卡(NIC,Network Interface Card))中的一个或多个,诸如IEEE802.11无线局域网(WLAN,Wireless Local Area Networks)无线接口、全球微波互联接入(Wi-MAX,World Interoperability for Microwave Access)接口、以太网接口、通用串行总线(USB,Universal Serial Bus)接口、蜂窝网络接口、蓝牙接口、近场通信(NFC,Near Field Communication)接口,等等。
在本发明的一个实施例中,货架搬运设备900的上述部件以及图9中未示出的其他部件也可以彼此相连接,例如通过总线。应当理解,图9所示的货架搬运设备结构框图仅仅是出于示例的目的,而不是对本发明范围的限制。本领域技术人员可以根据需要,增添或替换其他部件。
其中,图像传感器910用于针对目标货架采集目标图像,将目标图像传输至处理器530;处理器530用于执行如下计算机可执行指令,该计算机可执行指令被处理器执行时实现:
获取图像传感器910采集的目标图像及图像传感器的标定参数,其中,目标货架上布置有标识带,标识带由视觉标识组成;
从目标图像中提取标识图像及标识图像中视觉标识的预设标识点;
基于标定参数、标识图像及预设标识点,计算标识带相对于图像传感器的位姿信息;
根据位姿信息,确定目标货架的定位结果。
在得到定位结果后,可根据定位结果,控制货架搬运设备搬运目标货架。
上述为本实施例的一种货架搬运设备的示意性方案。需要说明的是,该货架搬运设备的技术方案与上述的货架定位方法的技术方案属于同一构思,货架搬运设备的技术方案未详细描述的细节内容,均可以参见上述货架定位方法的技术方案的描述。
本发明一实施例还提供一种计算机可读存储介质,其存储有计算机可执行指令,该计算机可执行指令被处理器执行时实现上述货架定位方法的步骤。
上述为本实施例的一种计算机可读存储介质的示意性方案。需要说明的是,该存储介质的技术方案与上述的货架定位方法的技术方案属于同一构思,存储介质的技术方案未详细描述的细节内容,均可以参见上述货架定位方法的技术方案的描述。
上述对本发明特定实施例进行了描述。其它实施例在所附权利要求书的范围内。在一些情况下,在权利要求书中记载的动作或步骤可以按照不同于实施例中的顺序来执行并且仍然可以实现期望的结果。另外,在附图中描绘的过程不一定要求示出的特定顺序或者连续顺序才能实现期望的结果。在某些实施方式中,多任务处理和并行处理也是可以的或者可能是有利的。
所述计算机指令包括计算机程序代码,所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读介质可以包括:能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM,Read-Only Memory)、随机存取存储器(RAM,Random Access Memory)、电载波信号、电信信号以及软件分发介质等。需要说明的是,所述计算机可读介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减,例如在某些司法管辖区,根据立法和专利实践,计算机可读介质不包括电载波信号和电信信号。
需要说明的是,对于前述的各方法实施例,为了简便描述,故将其都表述为一系列的动作组合,但 是本领域技术人员应该知悉,本发明实施例并不受所描述的动作顺序的限制,因为依据本发明实施例,某些步骤可以采用其它顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作和模块并不一定都是本发明实施例所必须的。
在上述实施例中,对各个实施例的描述都各有侧重,某个实施例中没有详述的部分,可以参见其它实施例的相关描述。
以上公开的本申请优选实施例只是用于帮助阐述本发明。可选实施例并没有详尽叙述所有的细节,也不限制该申请仅为所述的具体实施方式。显然,根据本申请实施例的内容,可作很多的修改和变化。本发明选取并具体描述这些实施例,是为了更好地解释本申请实施例的原理和实际应用,从而使所属技术领域技术人员能很好地理解和利用本发明。本申请仅受权利要求书及其全部范围和等效物的限制。
在本申请提供的另一具体实施方式中,提供了一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,其特征在于,所述计算机程序被处理器执行时实现上述基于货架标识的导航方法的步骤。
需要说明的是,对于前述的各方法实施例,为了简便描述,故将其都表述为一系列的动作组合,但是本领域技术人员应该知悉,本申请并不受所描述的动作顺序的限制,因为依据本申请,某些步骤可以采用其它顺序或者同时进行。其次,本领域技术人员也应该知悉,说明书中所描述的实施例均属于优选实施例,所涉及的动作和模块并不一定都是本申请所必须的。
在上述实施例中,对各个实施例的描述都各有侧重,某个实施例中没有详述的部分,可以参见其它实施例的相关描述。
以上公开的本申请优选实施例只是用于帮助阐述本申请。可选实施例并没有详尽叙述所有的细节,也不限制该发明仅为所述的具体实施方式。显然,根据本申请的内容,可作很多的修改和变化。本申请选取并具体描述这些实施例,是为了更好地解释本申请的原理和实际应用,从而使所属技术领域技术人员能很好地理解和利用本申请。本申请仅受权利要求书及其全部范围和等效物的限制。

Claims (44)

  1. 一种基于货架标识的导航系统,其特征在于,所述系统包括控制服务器、库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架间形成了供存取设备移动的巷道,货架上设置有沿巷道延伸的货架标识;
    所述控制服务器,被配置为确定目标货架上的目标存储位和目标存取设备,并向所述目标存取设备发送移动指令;
    所述目标存取设备,被配置为响应于所述移动指令,在所述目标货架对应的目标巷道中,根据货架标识移动至所述目标存储位。
  2. 如权利要求1所述的基于货架标识的导航系统,其特征在于,所述货架上设置有标识带,所述货架标识设置在所述标识带上。
  3. 如权利要求2所述的基于货架标识的导航系统,其特征在于,多个所述货架标识根据预设规则间隔设置在一条连续的所述标识带上。
  4. 如权利要求3所述的基于货架标识的导航系统,其特征在于,多个所述货架标识等间距地设置在一条连续的所述标识带上。
  5. 如权利要求1所述的基于货架标识的导航系统,其特征在于,所述货架标识包括二维码或除二维码外的其他标识。
  6. 如权利要求1所述的基于货架标识的导航系统,其特征在于,所述系统还包括高速通道,所述高速通道与所述库存区域相邻,所述高速通道设置有地面标识;
    所述目标存取设备,还被配置为在所述高速通道中根据地面标识移动至所述目标巷道的巷道口。
  7. 如权利要求1所述的基于货架标识的导航系统,其特征在于,所述货架包括至少一层货架横梁,所述货架标识设置在每层货架横梁上。
  8. 如权利要求7所述的基于货架标识的导航系统,其特征在于,所述目标存取设备,进一步被配置为根据任意一层货架横梁上的货架标识移动至所述目标存储位。
  9. 如权利要求8所述的基于货架标识的导航系统,其特征在于,所述目标存取设备,进一步被配置为根据距离地面不大于预设高度阈值的货架标识移动至所述目标存储位。
  10. 如权利要求1所述的基于货架标识的导航系统,其特征在于,所述目标存取设备,还被配置为根据货架标识从所述目标存储位中取出容器或将容器放置在所述目标存储位中。
  11. 如权利要求10所述的基于货架标识的导航系统,其特征在于,所述目标存取设备包括传感器;
    所述目标存取设备,还被配置为在从所述目标存储位中取放容器之前,根据所述传感器确定所述目标存取设备与对应货架标识的位置偏差,调整所述目标存取设备的位置,消除所述位置偏差。
  12. 如权利要求11所述的基于货架标识的导航系统,其特征在于,所述传感器包括图像传感器;
    所述目标存取设备,还被配置为获取所述图像传感器针对目标货架采集的目标图像及所述图像传感器的标定参数,所述目标货架上布置有标识带,所述标识带由视觉标识组成;从所述目标图像中提取标识图像及所述标识图像中视觉标识的预设标识点;基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息;根据所述位姿信息,确定所述目标货架的定位结果。
    根据所述定位结果,调整所述目标存取设备的位置,消除所述位置偏差。
  13. 如权利要求12所述的基于货架标识的导航系统,其特征在于,所述目标存取设备,还被配置为:对所述目标图像进行二值化处理,得到二值化图像;识别所述二值化图像中视觉标识的覆盖区域;确定所述覆盖区域组成的标识图像,并从所述覆盖区域中提取所述视觉标识的预设标识点。
  14. 如权利要求13所述的基于货架标识的导航系统,其特征在于,所述目标存取设备,还被配置为:对所述二值化图像进行边界检测,确定所述二值化图像的边界信息;基于所述边界信息,识别所述二值化图像中视觉标识的覆盖区域。
  15. 如权利要求12所述的基于货架标识的导航系统,其特征在于,所述标识带由多个视觉标识组成;所述标定参数包括内参;
    所述目标存取设备,还被配置为:利用所述内参,对所述标识图像中各视觉标识的预设标识点进行归一化处理,得到所述各视觉标识的预设标识点在所述标识图像上的第一坐标;获得所述各视觉标识的预设标识点在世界坐标系下的第二坐标;根据所述第一坐标,计算所述各视觉标识的中心点在所述标识图像上的第三坐标;根据所述第二坐标,计算所述各视觉标识的中心点在世界坐标系下的第四坐标;根据所述第三坐标和所述第四坐标,以所述各视觉标识的中心点满足预设排布条件为约束条件,计算所述标识带相对于所述图像传感器的位姿信息。
  16. 如权利要求12所述的基于货架标识的导航系统,其特征在于,所述标定参数包括内参;
    所述目标存取设备,还被配置为:利用所述内参,对所述预设标识点进行归一化处理,得到所述预设标识点在所述标识图像上的第一坐标;获得所述预设标识点在世界坐标系下的第二坐标;根据所述第 一坐标和所述第二坐标,计算所述标识带相对于所述图像传感器的位姿信息;获取所述标识带上的任一点在世界坐标系下的第五坐标以及所述任一点在所述标识图像上的第六坐标;根据所述第五坐标及所述位姿信息,计算所述任一点的第七坐标;将所述第六坐标与所述第七坐标作差,得到重投影误差,以所述重投影误差小于预设阈值为约束条件,调整所述位姿信息。
  17. 如权利要求15或16所述的基于货架标识的导航系统,其特征在于,所述标定参数还包括畸变参数;
    所述目标存取设备,还被配置为:利用所述畸变参数,对所述标识图像进行去畸变处理,得到处理后的所述标识图像。
  18. 如权利要求12-16任一项所述的基于货架标识的导航系统,其特征在于,所述标定参数还包括外参;所述目标货架上布置有多个标识带;
    所述目标存取设备,还被配置为:根据每个标识带相对于所述图像传感器的位姿信息及预设旋转矩阵,计算所述每个标识带在世界坐标系下绕旋转轴的旋转角度;根据所述每个标识带相对于所述图像传感器的位姿信息及所述外参,计算所述每个标识带在世界坐标系下的位姿信息;根据任两个标识带在世界坐标系下的旋转角度及位姿信息,拟合出所述任两个标识带间的曲线形状;根据所述任两个标识带间的曲线形状,获得所述目标货架的当前形态。
  19. 如权利要求12-16中任一项所述的基于货架标识的导航系统,其特征在于,所述标识带为长方形形状,所述标识带设置在所述目标货架的横梁上;所述目标货架包括多个横梁,每个横梁上均布置有所述标识带。
  20. 如权利要求1所述的基于货架标识的导航系统,其特征在于,所述货架包括货架腿,所述目标存取设备包括测距装置;
    所述目标存取设备,还被配置为通过所述测距装置计算与所述目标巷道两侧货架腿的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
  21. 如权利要求1所述的基于货架标识的导航系统,其特征在于,所述货架底部设置有遮挡标识,所述目标存取设备包括测距装置;
    所述目标存取设备,还被配置为通过所述测距装置计算与所述目标巷道两侧遮挡标识的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
  22. 如权利要求21所述的基于货架标识的导航系统,其特征在于,所述遮挡标识包括遮挡挡板。
  23. 如权利要求21所述的基于货架标识的导航系统,其特征在于,所述遮挡标识围绕所述货架底部设置。
  24. 如权利要求20或21所述的基于货架标识的导航系统,其特征在于,所述测距装置包括测距雷达或视觉传感器。
  25. 一种基于货架标识的导航系统,其特征在于,所述系统包括库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架间形成了供存取设备移动的巷道,货架上设置有遮挡标识,所述存取设备包括测距装置;
    所述存取设备,被配置为通过所述测距装置计算与所述巷道两侧遮挡标识的运行距离或者与巷道两侧的货架的货架腿的距离,并基于所述运行距离调整其在所述巷道中的位置。
  26. 如权利要求25所述的基于货架标识的导航系统,其特征在于,所述遮挡标识围绕所述货架底部设置。
  27. 如权利要求25所述的基于货架标识的导航系统,其特征在于,所述遮挡标识包括遮挡挡板。
  28. 一种基于货架标识的导航方法,其特征在于,应用于仓储系统,所述系统包括控制服务器、库存区域、存取设备、货架,其中,所述库存区域中设置有货架,货架间形成了供存取设备移动的巷道,货架上设置有沿巷道延伸的货架标识;
    所述控制服务器确定目标货架上的目标存储位和目标存取设备,并向所述目标存取设备发送移动指令;
    所述目标存取设备响应于所述移动指令,在所述目标货架对应的目标巷道中,根据货架标识移动至所述目标存储位。
  29. 如权利要求28所述的基于货架标识的导航方法,其特征在于,所述系统还包括高速通道,所述高速通道与所述库存区域相邻,所述高速通道设置有地面标识;
    所述方法还包括:
    所述目标存取设备在所述高速通道中根据地面标识移动至所述目标巷道的巷道口。
  30. 如权利要求28所述的基于货架标识的导航方法,其特征在于,所述货架包括至少一层货架横梁,所述货架标识设置在每层货架横梁上;
    所述目标存取设备根据货架标识移动至所述目标存储位,包括:
    所述目标存取设备根据任意一层货架横梁上的货架标识移动至所述目标存储位。
  31. 如权利要求30所述的基于货架标识的导航方法,其特征在于,所述目标存取设备根据任意一层货架横梁上的货架标识移动至所述目标存储位,包括:
    所述目标存取设备根据距离地面不大于预设高度阈值的货架标识移动至所述目标存储位。
  32. 如权利要求28所述的基于货架标识的导航方法,其特征在于,所述方法还包括:
    所述目标存取设备根据货架标识从所述目标存储位中取出容器或将容器放置在所述目标存储位中。
  33. 如权利要求32所述的基于货架标识的导航方法,其特征在于,所述目标存取设备包括传感器;
    所述方法还包括:
    所述目标存取设备在从所述目标存储位中取放容器之前,根据所述传感器确定所述目标存取设备与对应货架标识的位置偏差,调整所述目标存取设备的位置,消除所述位置偏差。
  34. 如权利要求28所述的基于货架标识的导航方法,其特征在于,所述货架包括货架腿,所述目标存取设备包括测距装置;
    所述方法还包括:
    所述目标存取设备通过所述测距装置计算与所述目标巷道两侧货架腿的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
  35. 如权利要求28所述的基于货架标识的导航方法,其特征在于,所述货架包括遮挡标识,所述目标存取设备包括测距装置;
    所述目标存取设备通过所述测距装置计算与所述目标巷道两侧遮挡标识的运行距离,并基于所述运行距离调整其在所述目标巷道中的位置。
  36. 如权利要求33所述的基于货架标识的导航方法,其特征在于,所述传感器包括图像传感器;
    所述根据所述传感器确定所述目标存取设备与对应货架标识的位置偏差,调整所述目标存取设备的位置,消除所述位置偏差,包括:
    获取所述图像传感器针对目标货架采集的目标图像及所述图像传感器的标定参数,所述目标货架上布置有标识带,所述标识带由视觉标识组成;
    从所述目标图像中提取标识图像及所述标识图像中视觉标识的预设标识点;
    基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息;
    根据所述位姿信息,确定所述目标货架的定位结果,根据所述定位结果,调整所述目标存取设备的位置,消除所述位置偏差。
  37. 如权利要求36所述的基于货架标识的导航方法,其特征在于,所述从所述目标图像中提取标识图像及所述标识图像中视觉标识的预设标识点的步骤,包括:
    对所述目标图像进行二值化处理,得到二值化图像;
    识别所述二值化图像中视觉标识的覆盖区域;
    确定所述覆盖区域组成的标识图像,并从所述覆盖区域中提取所述视觉标识的预设标识点。
  38. 如权利要求37所述的基于货架标识的导航方法,其特征在于,所述识别所述二值化图像中视觉标识的覆盖区域的步骤,包括:
    对所述二值化图像进行边界检测,确定所述二值化图像的边界信息;
    基于所述边界信息,识别所述二值化图像中视觉标识的覆盖区域。
  39. 如权利要求36所述的基于货架标识的导航方法,其特征在于,所述标识带由多个视觉标识组成;所述标定参数包括内参;
    所述基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息的步骤,包括:
    利用所述内参,对所述标识图像中各视觉标识的预设标识点进行归一化处理,得到所述各视觉标识的预设标识点在所述标识图像上的第一坐标;
    获得所述各视觉标识的预设标识点在世界坐标系下的第二坐标;
    根据所述第一坐标,计算所述各视觉标识的中心点在所述标识图像上的第三坐标;
    根据所述第二坐标,计算所述各视觉标识的中心点在世界坐标系下的第四坐标;
    根据所述第三坐标和所述第四坐标,以所述各视觉标识的中心点满足预设排布条件为约束条件,计算所述标识带相对于所述图像传感器的位姿信息。
  40. 如权利要求36所述的基于货架标识的导航方法,其特征在于,所述标定参数包括内参;
    所述基于所述标定参数、所述标识图像及所述预设标识点,计算所述标识带相对于所述图像传感器的位姿信息的步骤,包括:
    利用所述内参,对所述预设标识点进行归一化处理,得到所述预设标识点在所述标识图像上的第一 坐标;
    获得所述预设标识点在世界坐标系下的第二坐标;
    根据所述第一坐标和所述第二坐标,计算所述标识带相对于所述图像传感器的位姿信息;
    获取所述标识带上的任一点在世界坐标系下的第五坐标以及所述任一点在所述标识图像上的第六坐标;
    根据所述第五坐标及所述位姿信息,计算所述任一点的第七坐标;
    将所述第六坐标与所述第七坐标作差,得到重投影误差,以所述重投影误差小于预设阈值为约束条件,调整所述位姿信息。
  41. 如权利要求39或40所述的基于货架标识的导航方法,其特征在于,所述标定参数还包括畸变参数;所述方法还包括:
    利用所述畸变参数,对所述标识图像进行去畸变处理,得到处理后的所述标识图像。
  42. 如权利要求36-40任一项所述的基于货架标识的导航方法,其特征在于,所述标定参数还包括外参;所述目标货架上布置有多个标识带;
    在所述根据所述位姿信息,确定所述目标货架的定位结果的步骤之后,还包括:
    根据每个标识带相对于所述图像传感器的位姿信息及预设旋转矩阵,计算所述每个标识带在世界坐标系下绕旋转轴的旋转角度;
    根据所述每个标识带相对于所述图像传感器的位姿信息及所述外参,计算所述每个标识带在世界坐标系下的位姿信息;
    根据任两个标识带在世界坐标系下的旋转角度及位姿信息,拟合出所述任两个标识带间的曲线形状;
    根据所述任两个标识带间的曲线形状,获得所述目标货架的当前形态。
  43. 如权利要求36-40中任一项所述的基于货架标识的导航方法,其特征在于,所述标识带为长方形形状,所述标识带设置在所述目标货架的横梁上;所述目标货架包括多个横梁,每个横梁上均布置有所述标识带。
  44. 一种计算机可读存储介质,所述计算机可读存储介质存储有计算机程序,其特征在于,所述计算机程序被处理器执行时实现如权利要求28至43中任一项所述基于货架标识的导航方法的步骤。
PCT/CN2022/125539 2021-10-15 2022-10-15 基于货架标识的导航系统、方法 WO2023061501A1 (zh)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN202111205706.8A CN113936010A (zh) 2021-10-15 2021-10-15 货架定位方法、装置、货架搬运设备及存储介质
CN202111205706.8 2021-10-15
CN202111274173.9 2021-10-29
CN202111274173.9A CN114132679A (zh) 2021-10-29 2021-10-29 基于货架标识的导航系统、方法

Publications (1)

Publication Number Publication Date
WO2023061501A1 true WO2023061501A1 (zh) 2023-04-20

Family

ID=85987301

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/125539 WO2023061501A1 (zh) 2021-10-15 2022-10-15 基于货架标识的导航系统、方法

Country Status (1)

Country Link
WO (1) WO2023061501A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003246417A (ja) * 2002-02-22 2003-09-02 Sanshin Kinzoku Kogyo Kk 保管棚移動管理システム
CN109264283A (zh) * 2018-11-16 2019-01-25 北京艾瑞思机器人技术有限公司 分拣系统及货箱分拣方法
CN112214012A (zh) * 2019-07-11 2021-01-12 深圳市海柔创新科技有限公司 一种导航方法、移动载体及导航系统
CN112573058A (zh) * 2019-09-30 2021-03-30 深圳市海柔创新科技有限公司 一种取货方法、搬运机器人、处理终端及智能仓储系统
CN113936010A (zh) * 2021-10-15 2022-01-14 北京极智嘉科技股份有限公司 货架定位方法、装置、货架搬运设备及存储介质
CN114132679A (zh) * 2021-10-29 2022-03-04 北京极智嘉科技股份有限公司 基于货架标识的导航系统、方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003246417A (ja) * 2002-02-22 2003-09-02 Sanshin Kinzoku Kogyo Kk 保管棚移動管理システム
CN109264283A (zh) * 2018-11-16 2019-01-25 北京艾瑞思机器人技术有限公司 分拣系统及货箱分拣方法
CN112214012A (zh) * 2019-07-11 2021-01-12 深圳市海柔创新科技有限公司 一种导航方法、移动载体及导航系统
CN112573058A (zh) * 2019-09-30 2021-03-30 深圳市海柔创新科技有限公司 一种取货方法、搬运机器人、处理终端及智能仓储系统
CN113936010A (zh) * 2021-10-15 2022-01-14 北京极智嘉科技股份有限公司 货架定位方法、装置、货架搬运设备及存储介质
CN114132679A (zh) * 2021-10-29 2022-03-04 北京极智嘉科技股份有限公司 基于货架标识的导航系统、方法

Similar Documents

Publication Publication Date Title
US10776661B2 (en) Methods, systems and apparatus for segmenting and dimensioning objects
US10290115B2 (en) Device and method for determining the volume of an object moved by an industrial truck
CN110595476B (zh) 一种基于gps和图像视觉融合的无人机降落导航方法及装置
CN111598952B (zh) 一种多尺度合作靶标设计与在线检测识别方法及系统
WO2019195603A1 (en) Method, system, and apparatus for correcting translucency artifacts in data representing a support structure
CN111964680B (zh) 一种巡检机器人的实时定位方法
CN110096920A (zh) 一种面向视觉伺服的高精度高速定位标签和定位方法
CN110148099B (zh) 投影关系的修正方法及装置、电子设备、计算机可读介质
CN104268499B (zh) 一种带条码商品的条码标记检测方法
CN103824275A (zh) 在图像中搜寻鞍点状结构并测定其信息的系统和方法
CN103913149A (zh) 一种基于stm32单片机的双目测距系统及其测距方法
CN114396875B (zh) 一种基于深度相机垂直拍摄的长方形包裹体积测量方法
CN113936010A (zh) 货架定位方法、装置、货架搬运设备及存储介质
CN109005349B (zh) 基于智能平台面阵相机采集的机场道面表面图像改进拼接方法
US20240005548A1 (en) Information processing apparatus, three-dimensional position estimation method, and a non-transitory computer-readable medium
WO2023061501A1 (zh) 基于货架标识的导航系统、方法
CN112182967B (zh) 一种基于热影仪的光伏组件自动建模方法
CN115586796A (zh) 一种基于视觉的无人机降落位置的处理方法、装置及设备
CN111964681B (zh) 一种巡检机器人的实时定位系统
CN117897348A (zh) 货架定位方法、货架对接方法以及装置、设备、介质
Jende et al. Low-level tie feature extraction of mobile mapping data (mls/images) and aerial imagery
CN115077563A (zh) 车辆定位精度评价方法、装置及电子设备
CN112330748A (zh) 一种基于双目深度相机的托盘识别和定位方法
CN114202548A (zh) 一种叉车托盘的定位方法、装置、存储介质及电子设备
WO2023151603A1 (zh) 货箱存放方法和机器人

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22880437

Country of ref document: EP

Kind code of ref document: A1