WO2022088716A1 - 仓库管理方法、装置、系统及电子设备 - Google Patents

仓库管理方法、装置、系统及电子设备 Download PDF

Info

Publication number
WO2022088716A1
WO2022088716A1 PCT/CN2021/102441 CN2021102441W WO2022088716A1 WO 2022088716 A1 WO2022088716 A1 WO 2022088716A1 CN 2021102441 W CN2021102441 W CN 2021102441W WO 2022088716 A1 WO2022088716 A1 WO 2022088716A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
foreign
image
foreign object
warehouse management
Prior art date
Application number
PCT/CN2021/102441
Other languages
English (en)
French (fr)
Inventor
陈德平
黄灿
王银学
童孝康
Original Assignee
北京旷视机器人技术有限公司
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
Application filed by 北京旷视机器人技术有限公司 filed Critical 北京旷视机器人技术有限公司
Publication of WO2022088716A1 publication Critical patent/WO2022088716A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/41875—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by quality surveillance of production
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/183—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source
    • H04N7/185—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source from a mobile camera, e.g. for remote control
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/32—Operator till task planning
    • G05B2219/32252—Scheduling production, machining, job shop

Definitions

  • the present application relates to the field of automation, and in particular, to a warehouse management method, device, system and electronic device.
  • the stacker needs to transport the goods to the shelves through the running track, so as to complete the storage of the goods, or transport the goods on the shelves to other places through the running track, so as to complete the warehouse. Therefore, the detection of foreign objects on the running track is particularly important to ensure the safe and stable operation of the stacker.
  • infrared detection method is mainly used to realize the function of track foreign object detection, but the infrared detection accuracy is not enough, so that the track foreign object cannot be detected in time, which may cause equipment damage to the stacker, resulting in economic losses and local warehouse management system. invalid.
  • the purpose of this application is to provide a warehouse management method, device, system and electronic device, which can improve at least one of the above problems.
  • the embodiment of the present application provides a warehouse management method, the method is applied to a server; the server is connected in communication with at least one camera device; the method includes: during the operation of the target robot, receiving the current corresponding driving area of the target robot collected by the target camera device image, the shooting angle of the target camera device points to the current corresponding driving area of the target robot; the target camera device is one of at least one camera device; detects whether the image contains foreign objects; wherein, the foreign objects are non-fixed objects in the driving area; if so, Control the target robot to stop running.
  • the above-mentioned at least one camera device is respectively installed in a designated position of the warehouse; the server pre-stores the corresponding relationship between at least one camera device and the driving area; receiving the image of the current corresponding driving area of the target robot collected by the target camera device, including: according to The corresponding relationship and the current corresponding driving area of the target robot are determined, and the target camera device is determined; the image collected by the target camera device is obtained.
  • the above method also includes: acquiring a video frame sequence within a specified time period before the image collected by the target camera device; if the foreign body type of the foreign body is a non-living type, determining the initial storage position of the foreign body based on the video frame sequence, and reporting to the first designated
  • the terminal sends first notification information; the first notification information carries prompt information corresponding to the initial storage position; if the foreign object type of the foreign object is a living body type, the tracking track information of the foreign object is determined based on the video frame sequence, and the second designated terminal is sent to the second designated terminal. notification information; the second notification information carries the tracking track information of the foreign object.
  • the above server is also connected with an alarm device, and the alarm device includes: a voice alarm device and/or a signal light alarm device; the method further includes: if the foreign body type of the foreign body is a living body type, triggering the alarm device to give an alarm.
  • the server is preconfigured with a foreign object detection model; detecting whether the image contains foreign objects includes: inputting the image into the foreign object detection model; if the image output by the foreign object detection model is marked with a foreign object area, determine that the image contains foreign objects.
  • the above method also includes: if the image contains foreign objects, extracting the foreign object area marked on the image to obtain a sub-image of the foreign object area; inputting the sub-image of the foreign object area into the image recognition model to obtain the foreign object type of the foreign object; Live type.
  • step of controlling the target robot to stop running includes one of the following: sending a shutdown command to the target robot to stop the target robot from running; sending a control command to suspend the target robot for a specified period of time to stop the target robot from running.
  • the above method further includes: if the image contains a foreign object, locating the target position corresponding to the foreign object; and marking the target position on the interface displaying the driving area in a preset labeling manner.
  • marking the target position on the interface displaying the driving area in a preset marking method includes: determining the foreign object type of the foreign object; determining the marking method according to the foreign object type; marking the target position on the interface displaying the driving area in the determined marking method.
  • the above-mentioned target robot is a target stacker
  • the driving area is an area where a track on which the target stacker can travel is located.
  • the embodiment of the present application also provides a warehouse management device, the device is applied to a server; the server is communicatively connected to at least one camera device; the device includes: an image receiving module, configured to receive the target collected by the target camera device during the operation of the target robot. The image of the driving area currently corresponding to the robot, the shooting angle of the target camera device points to the driving area corresponding to the target robot; the target camera device is one of at least one camera device; the foreign object detection module is configured to detect whether the image contains foreign objects; wherein, the foreign object is Non-fixed objects in the driving area; the robot control module is configured to control the target robot to stop running when the detection result of the foreign object detection module is yes.
  • the device further includes: a foreign body alarm module; the foreign body alarm module is configured to: obtain a video frame sequence within a specified time period before the image collected by the target camera device; if the foreign body type of the foreign body is a non-living body type, based on the video frame sequence The sequence determines the initial storage position of the foreign object, and sends the first notification information to the first designated terminal; the first notification information carries the prompt information corresponding to the initial storage position; Track the track information, and send the second notification information to the second designated terminal; the second notification information carries the tracking track information of the foreign object.
  • a foreign body alarm module is configured to: obtain a video frame sequence within a specified time period before the image collected by the target camera device; if the foreign body type of the foreign body is a non-living body type, based on the video frame sequence The sequence determines the initial storage position of the foreign object, and sends the first notification information to the first designated terminal; the first notification information carries the prompt information corresponding to the initial storage position; Track the track information, and send
  • the device further includes: a type judgment module; the type judgment module is configured to: if the image contains a foreign body, extract the foreign body area marked by the image to obtain a foreign body area sub-image; input the foreign body area sub-image into the image recognition model, The foreign body type of the obtained foreign body; the foreign body type includes living type and non-living type.
  • the device further includes: a foreign object marking module; the foreign object marking module is configured to: if the image contains a foreign object, locate the target position corresponding to the foreign object; mark the target position on the interface displaying the driving area in a preset marking manner.
  • the embodiment of the present application further provides a warehouse management system, the system includes: a server, a camera device and a target robot; the server is respectively connected in communication with the camera device and the target robot; the server is configured to execute the steps of the warehouse management method according to the first aspect .
  • the above-mentioned at least one camera device is installed at a designated position of the target robot to capture images of the driving area corresponding to the target robot; area for image acquisition.
  • the above-mentioned warehouse is a three-dimensional warehouse
  • the target robot is a stacker
  • at least one camera device is installed above the wheel sets on both sides of the stacker; and/or, at least one camera device is installed on both sides of the roadway of the three-dimensional warehouse.
  • Embodiments of the present application further provide an electronic device, including a processor and a memory, where the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the above warehouse management method.
  • Embodiments of the present application further provide a computer-readable storage medium, where the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are invoked and executed by the processor, the computer-executable instructions prompt the processor to implement the above warehouse management method.
  • FIG. 1 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • FIG. 2 is a flowchart of a warehouse management method provided by an embodiment of the present application.
  • FIG. 3 is a flowchart of a foreign body detection method provided by an embodiment of the present application.
  • FIG. 4 is a flowchart of another warehouse management method provided by an embodiment of the present application.
  • FIG. 5 is a workflow diagram of a warehouse management method provided by an embodiment of the present application.
  • FIG. 6 is a structural block diagram of a warehouse management apparatus provided by an embodiment of the present application.
  • FIG. 7 is a structural block diagram of another warehouse management apparatus provided by an embodiment of the present application.
  • FIG. 8 is a structural block diagram of a warehouse management system provided by an embodiment of the present application.
  • the stacker needs to transport the goods to the shelves through the running track, so as to complete the storage of the goods, or transport the goods on the shelves to other places through the running track, so as to complete the warehouse. Therefore, the detection of foreign objects on the running track is particularly important to ensure the safe and stable operation of the stacker.
  • infrared detection method is mainly used to realize the function of track foreign object detection.
  • the infrared detection accuracy is not enough, so that the track foreign object cannot be detected in time, which may cause equipment damage to the stacker, resulting in economic losses and partial failure of the warehouse management system. .
  • the embodiments of the present application provide a warehouse management method, device, system, and electronic device. For ease of understanding, the embodiments of the present application are described in detail below.
  • FIG. 1 An example electronic device 100 for implementing the warehouse management method, apparatus, and system of the embodiments of the present application is described with reference to FIG. 1 .
  • FIG. 1 is a schematic structural diagram of an electronic device
  • the electronic device 100 includes one or more processors 102, one or more storage devices 104, an input device 106, an output device 108, and an image acquisition device 110. These components are The bus system 112 and/or other form of connection mechanism (not shown) are interconnected. It should be noted that the components and structures of the electronic device 100 shown in FIG. 1 are only exemplary and not restrictive, and the electronic device may have some of the components shown in FIG. 1 or not shown in FIG. 1 as required. other components and structures.
  • the processor 102 may be a central processing unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 100 to perform desired functions.
  • CPU central processing unit
  • the processor 102 may be a central processing unit (CPU) or other form of processing unit having data processing capabilities and/or instruction execution capabilities, and may control other components in the electronic device 100 to perform desired functions.
  • the storage device 104 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory.
  • the volatile memory may include, for example, random access memory (RAM) and/or cache memory, or the like.
  • the non-volatile memory may include, for example, read only memory (ROM), hard disk, flash memory, and the like.
  • One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 102 may execute the program instructions to implement the client functions (implemented by the processor) in the embodiments of the present application described below. and/or other desired functionality.
  • Various application programs and various data such as various data used and/or generated by the application program, etc. may also be stored in the computer-readable storage medium.
  • the input device 106 may be a device used by a user to input instructions, and may include one or more of a keyboard, mouse, microphone, touch screen, and the like.
  • the output device 108 may output various information (eg, images or sounds) to the outside (eg, a user), and may include one or more of a display, a speaker, and the like.
  • the image acquisition device 110 may be a panoramic camera, or a common camera, a plurality of cameras installed in a designated position of a roadway in a warehouse, or a camera installed in a designated position of a target robot in the warehouse.
  • the image acquisition device in the embodiment of the present application can capture images expected by users in the warehouse (for example, photos and videos of each roadway, each shelf, each target robot driving area, etc.), and store the captured images in the storage device. 104 for use by other components.
  • the example electronic device for implementing a warehouse management method, apparatus and system may be implemented as a terminal such as a server, a smart phone, a tablet computer, a computer and the like.
  • the 2 is a warehouse management method provided by an embodiment of the present application.
  • the method is applied to a server, usually a server in a warehouse management system, and the server is connected in communication with at least one camera device; the camera device may be a plurality of cameras disposed in the warehouse
  • the number of cameras at the designated location in the warehouse is related to the installation location and the internal structure of the warehouse.
  • Cameras can be installed at multiple locations in the warehouse to capture images of various areas in the warehouse, or they can be cameras installed on robots. .
  • the above-mentioned camera device can capture images of the driving area of the robot in the warehouse.
  • the above warehouse management method mainly includes the following steps S202 to S206:
  • Step S202 during the running process of the target robot, receive an image of the current corresponding driving area of the target robot collected by the target camera device.
  • the above-mentioned target camera is one of the above-mentioned at least one camera, and its shooting angle points to the current corresponding driving area of the target robot, so that the image of the corresponding driving area of the target robot can be collected.
  • the target camera device can be installed on the target robot, and its shooting angle of view points to the driving area corresponding to the target robot, and moves with the movement of the target robot.
  • images of the driving area corresponding to the target robot can be continuously captured.
  • the above-mentioned target camera device can also be installed in a designated position in the warehouse, for example, on both sides of each roadway or a certain position on the shelf, and its shooting angle points to the roadway, so that the robot passing through the roadway can be photographed. Image. It should be understood that when the target robot is running in different driving areas, images collected by the target robot may be received from different camera devices.
  • the above-mentioned target robots can be different types of handling robots, such as stackers, flap robots, roller robots, jacking robots, traction robots, and forklifts.
  • the target robot is controlled to start running.
  • the image of the driving area of the target robot is collected in real time by the target camera device.
  • the driving area may be the area where the target stacker can travel on the track.
  • Step S204 detecting whether the image contains foreign objects; wherein, the foreign objects are non-fixed objects in the driving area.
  • the image is detected and analyzed quickly to determine whether there is a foreign object in the image.
  • detection methods There are many specific detection methods. For example, the image is compared with a pre-stored normal image without foreign objects. For example, in practical applications, a normal image of each driving area without foreign objects can be stored in advance. The similarity between the image and the pre-stored normal image belonging to the same driving area is compared to determine whether the collected image contains foreign objects; or the image can be detected by a preset trained model.
  • the trained model can It is a model trained based on a deep learning algorithm.
  • the trained model can accurately determine whether there is a foreign body in the image, output the foreign body discrimination result, and mark the location of the foreign body in the image.
  • the region images marked with real foreign object detection results can be used as training samples, and the above models can be trained through the training samples, based on the preset loss function and back propagation.
  • the algorithm adjusts the model parameters until the model can output the expected foreign object detection results to determine that the training is over.
  • the driving area of the target robot is the roadway or track.
  • the objects on the roadway or track will hinder the robot's driving, so they can be regarded as foreign objects, which are non-fixed objects in the driving area, such as dropped goods. , parts, living things, etc.
  • Step S206 if the image contains foreign objects, control the target robot to stop running.
  • the target robot can be controlled to suspend operation in a timely manner.
  • a shutdown command can be sent directly to the target robot to stop the target robot from running, or a specified time period can be sent to the target robot.
  • the specified time period can be flexibly set according to requirements, and is usually determined based on the normal foreign body processing time of the staff, such as being longer than the conventional foreign body processing time.
  • the warehouse management method provided by the embodiment of the present application can receive the image of the current corresponding driving area of the target robot collected by the target camera during the operation of the target robot; and perform foreign object detection on the image of the corresponding driving area of the target robot. , when it is determined that the image contains foreign objects, control the target robot to suspend operation in time.
  • the above-mentioned method provided by the embodiment of the present application can timely discover foreign objects on the running path of the robot, thereby effectively avoiding equipment damage or other economic losses.
  • the above at least one camera device may be installed at a designated position in the warehouse, for example, in a roadway in the warehouse
  • the two sides of the warehouse or a designated position of the shelf can collect images of all the lanes, shelves and other areas in the warehouse.
  • the server also prestores at least one correspondence between the cameras and the driving area, for example, camera 1 corresponds to driving area A, camera 2 corresponds to driving area B, and so on.
  • the above step of receiving the image of the current corresponding driving area of the target robot collected by the target camera device can be implemented in the following ways:
  • the target camera is determined; the image collected by the target camera is acquired.
  • the target camera can be determined as the camera 2 based on the above-mentioned correspondence between the camera and the driving area, and then the image collected by the camera 2 is obtained, that is, The image of the current corresponding driving area of the target robot can be obtained.
  • a foreign object detection model is preconfigured in the server in this embodiment of the present application.
  • the foreign object detection model is obtained by training a neural network through a large number of training samples, and can be used for image foreign object detection. and foreign body markings.
  • the above-mentioned step of detecting whether an image contains foreign objects can be implemented with reference to the flowchart of the foreign object detection method shown in FIG. 3 , and specifically referring to the following steps S302 to S304:
  • Step S302 input the image into the foreign object detection model.
  • Step S304 if the image output by the foreign object detection model is marked with a foreign object area, it is determined that the image contains a foreign object.
  • the above-mentioned foreign body detection model uses a deep learning algorithm, a model that can detect and segment foreign bodies is obtained by collecting data in advance and training. Therefore, using this model, the exact pixel area of the image where the foreign object is located can be obtained, and the identification of the foreign object area will be marked in the output image, such as a rectangular frame, a square frame, etc., or the pixels of the foreign object in the image can be directly output. Point coordinates, through the coordinates to mark the location area where the foreign object is located.
  • the foreign object detection model After the above image is input into the foreign object detection model, the foreign object detection model will output an image. If the foreign object detection model outputs an image marked with a foreign object area, it can be determined that the image contains foreign objects.
  • the detection of foreign objects in the above manner can improve the detection accuracy while ensuring the image detection efficiency, so as to better ensure the stable and safe operation of the target robot.
  • a shutdown instruction may also be sent to the target camera device to stop the target camera device from working. In this way, the target camera device can be prevented from continuing to collect images with foreign objects, and waste of power consumption and meaningless image acquisition can be avoided.
  • the embodiment of the present application can also issue an alarm after determining that there is a foreign object. For example, through steps S402 to S410 in the flowchart of the method shown in FIG. 4 , the alarming process for the above-mentioned foreign objects can be realized:
  • Step S402 if the image contains a foreign body, extract the foreign body area marked in the image to obtain a sub-image of the foreign body area;
  • Step S404 input the sub-image of the foreign body region into the image recognition model to obtain the foreign body type of the foreign body; the foreign body type includes living body type and non-living body type.
  • the above-mentioned image recognition model may be a model obtained through deep learning training and used to identify the type of object; the neural network structure included in the model may have various forms, which are not specifically limited here.
  • the foreign object type of the foreign object in the image can be obtained, and it can be determined whether the foreign object is a living type or a non-living type.
  • Step S406 acquiring a video frame sequence within a specified time period before the image captured by the target camera device.
  • the above specified duration is a preset duration, and for non-living foreign objects, the specified duration can be determined based on the free fall time of the object corresponding to the height of the shelf in the warehouse. In this way, it can be ensured that the image of the initial storage position of the dropped object can be found in the video frame sequence within the specified time period, and the initial storage position of the foreign object in the image can also be determined.
  • the specified time period may be another set value, which may be relatively longer, so as to determine the tracking trajectory information of the living body, and then determine how the living body enters the warehouse.
  • Step S408 if the foreign body type of the foreign body is a non-living type, determine the initial storage position of the foreign body based on the video frame sequence, and send first notification information to the first designated terminal; the first notification information carries the prompt information corresponding to the initial storage position;
  • non-living types generally refer to goods, but may also be lost items such as parts or mobile phones in the warehouse. If non-living foreign objects are detected in the image of the driving area of the target robot, it may be a Goods or parts slip off a shelf, or someone's phone falls, etc. Further, the initial storage position of the foreign object is found through the video frame sequence within the above-mentioned specified duration.
  • the server After the server determines the initial storage location of the foreign objects, it can send the first notification information to the first designated terminal.
  • the first designated terminal may be an intelligent terminal corresponding to the staff responsible for the management of the entire warehouse, or it may be responsible for the driving area.
  • the first notification information can be a short message message, or an email message, or a message sent by an instant messaging software, etc., and the information notification form is not limited here.
  • the first notification information may also carry an image containing the foreign object, so as to facilitate processing by the staff.
  • the foreign object processing staff can quickly return the goods to its original storage position for the first time, or make new goods replenishment when the goods are damaged. , or deal with the parts or mobile phones accordingly, so as to avoid additional losses on both sides when the goods are delivered, and in addition, it can also speed up the process of returning the target robot to normal work.
  • Step S410 if the foreign body type of the foreign body is a living body type, determine the tracking track information of the foreign body based on the video frame sequence, and send second notification information to the second designated terminal; the second notification information carries the tracking track information of the foreign body.
  • the designated terminal sends second notification information.
  • the second notification information includes the tracking track information of the foreign object, which can be the positioning information of the person, so that the foreign object processing staff can find the foreign object and process it in time according to the positioning information.
  • the second notification information may also carry an image containing the foreign object, so as to facilitate the handling by the staff.
  • first designated terminal and second designated terminal may be the same terminal, or may be different terminals, and different settings may be performed according to actual conditions.
  • the above server is also connected with an alarm device, and the alarm device includes: a voice alarm device and/or a signal light alarm device; if the server determines that the foreign object type is a living body type, the alarm device is directly triggered to give an alarm.
  • the alarm mode of the above-mentioned alarm device may include: voice alarm and/or signal lamp alarm.
  • the server if it detects that there is a foreign object in the image of the driving area of the target robot, it can directly control the target robot to stop running. In order to further shorten the time for the target robot to resume normal work, the following two methods (1) and (2) One of the control target robots to stop running:
  • the server When the server detects that there is a foreign object in the image of the driving area of the target robot, it directly sends a shutdown instruction to the target robot to stop the target robot from working. In this case, after the foreign object processing is completed, the server needs to send an opening command to the target robot again, so as to resume normal operation, or manually start the robot.
  • the advantage of this method is that the target robot can be restarted to work in time, and the time for the target robot to resume normal work is shortened.
  • the specified time for the above-mentioned suspension will be set relatively long.
  • the foreign object processing time of the staff under normal circumstances can be calculated in advance, and then set The specified pause time is longer than the foreign object processing time, so as to ensure that the target robot will automatically start running again after the foreign object is completely processed.
  • the advantage of this method is to reduce the steps for the server to re-send the start command once, or to reduce the operation of manually starting the robot once, which reduces the power consumption of the server and improves the processing efficiency of the server.
  • the terminal interface can also display the position of the foreign object.
  • the above method may further include: if the image contains a foreign object, locating the target position corresponding to the foreign object; The target location is marked on the interface of the area.
  • the above-mentioned preset labeling methods include at least one of the following: specifying color box labeling, bold box labeling, and blinking box labeling.
  • the embodiment of the present application may also perform foreign object labeling in the following manner: first determine the foreign body type of the foreign body; then determine the labeling method according to the foreign body type; the determined labeling method is displayed in the display The target location is marked on the interface of the driving area. For example, for the foreign body of the living type, it is circled with a red frame, and for the foreign body of the non-living type, it is circled with a green frame, etc.
  • S1 The server sends a power-on command to the stacker
  • S4 The camera executes the above startup command, the camera is turned on, and image acquisition is performed;
  • S5 The camera continuously scans and transmits images to the server in real time
  • the server stores the image A and performs foreign object detection on the image A in real time;
  • S7 The server judges that there is a foreign object, and sends a pause command to the stacker;
  • S8 The server judges that there is a foreign object, and sends a shutdown command to the camera;
  • S9 The server performs image recognition on the image A to determine the type of foreign object
  • the server determines that the foreign body type is a living body type, and sends an early warning command to the stacker;
  • S12 The server locates the foreign body of the living body type, and sends the positioning information to the client;
  • the server determines that the foreign object type is a non-living type, obtains relevant videos for analysis, determines the initial storage position of the goods, and sends the relevant video and the initial storage position of the goods to the client;
  • S14 The foreign body is processed by the staff who received the notification through the user terminal;
  • step S1 The subsequent server continues to perform step S1, and re-sends an opening command to the stacker, so that the stacker continues to run. It should be noted that the above is only a specific implementation process, and some steps may be changed or adjusted.
  • the warehouse management method can accurately determine whether there is foreign matter in the driving area of the target robot, effectively prevent the equipment damage of the stacker, thereby avoiding economic losses and warehouse management. Partial failure of the system; it can accurately identify the type of the foreign object, and give different early warning reminders for different types.
  • the reminder information includes the positioning information or initial storage position of the foreign object, so that the staff can deal with the foreign object in a timely and convenient manner. , improve the foreign body processing efficiency of the staff, and shorten the time for the target robot to return to normal work.
  • the embodiment of the present application does not need to set up high-density infrared detection equipment, and the cost is low.
  • the embodiments of the present application further provide a warehouse management device, which is applied to a server; the server is connected in communication with at least one camera device; as shown in FIG. 6 , the device includes an image receiving module 602 and a foreign object detection module 604 and Robot Control Module 606:
  • the image receiving module 602 is configured to receive the image of the current corresponding driving area of the target robot collected by the target camera device during the running process of the target robot, and the shooting angle of the target camera device points to the current corresponding driving area of the target robot;
  • the target camera device is: One of at least one camera;
  • the foreign object detection module 604 configured to detect whether the image contains foreign objects; wherein, the foreign object is a non-fixed object in the driving area;
  • the robot control module 606 is configured to be detected when the foreign object detection module The result is yes, Control the target robot to stop running.
  • the above-mentioned at least one camera device is respectively installed in the designated position of the warehouse; the above-mentioned server pre-stores the corresponding relationship between at least one camera device and the driving area; the above-mentioned image receiving module 602 is also configured to: according to the corresponding relationship and the current corresponding driving of the target robot area, determine the target camera device; acquire the image collected by the target camera device.
  • the warehouse management device further includes: a foreign object alarm module 608; the foreign object alarm module 608 is configured to: acquire a video frame sequence within a specified time period before the image captured by the target camera device ; if the foreign body type of the foreign body is a non-living type, determine the initial storage position of the foreign body based on the video frame sequence, and send the first notification information to the first designated terminal; the first notification information carries the prompt information corresponding to the initial storage position; if the foreign body The type of the foreign object is a living body type, the tracking track information of the foreign object is determined based on the video frame sequence, and second notification information is sent to the second designated terminal; the second notification information carries the tracking track information of the foreign object.
  • the foreign object alarm module 608 is configured to: acquire a video frame sequence within a specified time period before the image captured by the target camera device ; if the foreign body type of the foreign body is a non-living type, determine the initial storage position of the foreign body based on the video frame sequence, and send the first notification information
  • the above-mentioned server is also connected with an alarm device, and the alarm device includes: a voice alarm device and/or a signal light alarm device; the above-mentioned foreign body alarm module 608 is also configured to: if the foreign body type of the foreign body is a living body type, the alarm device is triggered to give an alarm.
  • the server is preconfigured with a foreign body detection model; the foreign body detection module 604 is further configured to: input the image into the foreign body detection model; if the image output by the foreign body detection model is marked with a foreign body area, it is determined that the image contains foreign bodies.
  • the above-mentioned device also includes a type judgment module 610, and the type judgment module 610 is configured to: if the image contains a foreign body, extract the foreign body area marked by the image to obtain a sub-image of the foreign body area; input the sub-image of the foreign body area into the image recognition model to obtain Foreign body types of foreign bodies; foreign body types include living types and non-living types.
  • the above-mentioned robot control module 606 is further configured to: send a shutdown command to the target robot to stop the target robot from running;
  • the above device further includes a foreign object marking module 612; the foreign object marking module 612 is configured to: if the image contains a foreign object, locate the target position corresponding to the foreign object; mark the target position on the interface displaying the driving area in a preset marking manner.
  • the above-mentioned foreign object marking module 612 is further configured to: determine the foreign object type of the foreign object; determine the marking method according to the foreign object type; mark the target position on the interface displaying the driving area in the determined marking method.
  • the above-mentioned target robot is a target stacker
  • the traveling area is an area where a track that the target stacker can travel is located.
  • the embodiments of the present application further provide a warehouse management system.
  • the system includes: a server 82 , a camera device 84 and a target robot 86 ;
  • the camera device 84 and the target robot 86 are connected in communication;
  • the server 82 is configured to execute the steps of the warehouse management method in the above method embodiment.
  • the above-mentioned at least one camera device 84 is installed at a designated position of the target robot 86 to capture images of the driving area corresponding to the target robot 86; Image acquisition is performed in the driving area corresponding to the robot 86 .
  • the above-mentioned warehouse is a three-dimensional warehouse
  • the above-mentioned target robot 86 is a stacker crane
  • at least one camera device 84 is installed above the wheel sets on both sides of the stacker crane; and/or, at least one camera device 84 Installed on both sides of the roadway of the three-dimensional warehouse.
  • the above-mentioned system further includes a user terminal, such as a terminal corresponding to a pre-stored mobile phone number, a terminal corresponding to a pre-stored email, and the like in the server.
  • a user terminal such as a terminal corresponding to a pre-stored mobile phone number, a terminal corresponding to a pre-stored email, and the like in the server.
  • the installation position of the camera device includes but is not limited to the installation position provided above.
  • the warehouse management system provided by the embodiment of the present application has the same implementation principle and technical effect as the foregoing warehouse management method embodiments.
  • the parts not mentioned in the embodiments of the warehouse management system reference may be made to the foregoing warehouse management method. Corresponding content in the examples.
  • Embodiments of the present application further provide a computer-readable storage medium, where computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are invoked and executed by a processor, the computer-executable instructions cause the processor to
  • a computer-readable storage medium where computer-executable instructions are stored in the computer-readable storage medium, and when the computer-executable instructions are invoked and executed by a processor, the computer-executable instructions cause the processor to
  • the computer program product of the warehouse management method, apparatus, system, and electronic device provided by the embodiments of the present application includes a computer-readable storage medium storing program codes, and the instructions included in the program codes can be used to execute the methods described in the foregoing method embodiments.
  • the instructions included in the program codes can be used to execute the methods described in the foregoing method embodiments.
  • the technical solution of the present application can be embodied in the form of a software product in essence, or the part that contributes to the prior art or the part of the technical solution, and the computer software product is stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application.
  • the aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .
  • the image collected by the target camera device with the shooting angle of view pointing to the current driving area of the target robot can be received, and then the image of the current driving area corresponding to the target robot can be received.
  • Perform detection to determine whether the image contains foreign objects; the foreign objects are non-fixed objects in the driving area; if foreign objects are detected in the image, the target robot will be controlled to stop running in time, which can effectively avoid equipment damage or other economic losses.

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • General Engineering & Computer Science (AREA)
  • Manufacturing & Machinery (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Warehouses Or Storage Devices (AREA)
  • Manipulator (AREA)

Abstract

本申请提供了一种仓库管理方法、装置、系统及电子设备，涉及自动化领域，其中，仓库管理方法应用于服务器；服务器与至少一个摄像装置通信连接；方法包括：在目标机器人运行过程中，接收目标摄像装置采集到的目标机器人当前对应的行驶区域的图像，目标摄像装置的拍摄视角指向目标机器人当前对应的行驶区域；目标摄像装置为至少一个摄像装置之一；检测图像是否包含异物；其中，异物为行驶区域中的非固定物；如果是，控制目标机器人停止运行。本申请能够通过对目标机器人行驶区域的图像进行检测，判断图像中是否有异物，进而在有异物时及时控制目标机器人停止运行，以避免设备损坏或其它经济损失等。

Description

仓库管理方法、装置、系统及电子设备
相关申请的交叉引用
本申请要求于2020年10月30日提交中国专利局的申请号为202011194452.X、名称为“仓库管理方法、装置、系统及电子设备”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
本申请涉及自动化领域，尤其是涉及一种仓库管理方法、装置、系统及电子设备。
背景技术
相关技术中所采用的立体仓库，堆垛机需要通过运行轨道将货物运送到货架，从而完成货物入库，或者通过运行轨道将货架上的货物运送至其它地方，从而完成出库。因此，运行轨道的异物检测对保证堆垛机的安全稳定运行尤为重要。相关技术中主要采用红外检测方式实现轨道异物检测功能，但是红外检测精度不够，导致轨道异物不能被及时发现，很有可能会引起堆垛机的设备损坏，造成经济损失和仓库管理系统的局域失效。
公开内容
本申请的目的在于提供一种仓库管理方法、装置、系统及电子设备，能够改善以上问题至少之一。
本申请实施例提供一种仓库管理方法，方法应用于服务器；服务器与至少一个摄像装置通信连接；该方法包括：在目标机器人运行过程中，接收目标摄像装置采集到的目标机器人当前对应的行驶区域的图像，目标摄像装置的拍摄视角指向目标机器人当前对应的行驶区域；目标摄像装置为至少一个摄像装置之一；检测图像是否包含异物；其中，异物为行驶区域中的非固定物；如果是，控制目标机器人停止运行。
进一步的，上述至少一个摄像装置分别安装于仓库的指定位置；服务器预存有至少一个摄像装置与行驶区域的对应关系；接收目标摄像装置采集到的目标机器人当前对应的行驶区域的图像，包括：根据对应关系和目标机器人当前对应的行驶区域，确定目标摄像装置；获取目标摄像装置采集的图像。
进一步的，上述方法还包括：获取目标摄像装置采集的图像之前指定时长内的视频帧序列；如果异物的异物类型为非活体类型，基于视频帧序列确定异物的初始存放位置，并向第一指定终端发送第一通知信息；第一通知信息携带有初始存放位置对应的提示信息； 如果异物的异物类型为活体类型，基于视频帧序列确定异物的跟踪轨迹信息，并向第二指定终端发送第二通知信息；第二通知信息携带有异物的跟踪轨迹信息。
进一步的，上述服务器还连接有报警装置，报警装置包括：语音报警器和/或信号灯报警器；方法还包括：如果异物的异物类型为活体类型，触发报警装置进行报警。
进一步的，上述服务器预先配置有异物检测模型；检测图像是否包含异物，包括：将图像输入异物检测模型；如果异物检测模型输出的图像标注有异物区域，确定图像包含有异物。
进一步的，上述方法还包括：如果图像包含有异物，提取图像标注的异物区域，得到异物区域子图像；将异物区域子图像输入图像识别模型，得到异物的异物类型；异物类型包括活体类型和非活体类型。
进一步的，上述控制目标机器人停止运行的步骤，包括以下之一：向目标机器人发送关闭指令，以使目标机器人停止运行；向目标机器人发送暂停指定时长的控制指令，以使目标机器人暂停运行。
进一步的，上述方法还包括：如果图像包含异物，定位异物对应的目标位置；以预设标注方式在显示行驶区域的界面上标注目标位置。
进一步的，以预设标注方式在显示行驶区域的界面上标注目标位置，包括：确定异物的异物类型；根据异物类型确定标注方式；以确定的标注方式在显示行驶区域的界面上标注目标位置。
进一步的，上述目标机器人为目标堆垛机，行驶区域为目标堆垛机可行驶的轨道所在区域。
本申请实施例还提供一种仓库管理装置，装置应用于服务器；服务器与至少一个摄像装置通信连接；装置包括：图像接收模块，配置成在目标机器人运行过程中，接收目标摄像装置采集到的目标机器人当前对应的行驶区域的图像，目标摄像装置的拍摄视角指向目标机器人对应的行驶区域；目标摄像装置为至少一个摄像装置之一；异物检测模块，配置成检测图像是否包含异物；其中，异物为行驶区域中的非固定物；机器人控制模块，配置成在异物检测模块的检测结果为是时，控制目标机器人停止运行。
进一步的，所述装置还包括：异物报警模块；所述异物报警模块配置成：获取目标摄像装置采集的图像之前指定时长内的视频帧序列；如果异物的异物类型为非活体类型，基于视频帧序列确定异物的初始存放位置，并向第一指定终端发送第一通知信息；第一通知信息携带有初始存放位置对应的提示信息；如果异物的异物类型为活体类型，基于视频帧序列确定异物的跟踪轨迹信息，并向第二指定终端发送第二通知信息；第二通知信息携带有异物的跟踪轨迹信息。
进一步的，所述装置还包括：类型判断模块；所述类型判断模块配置成：如果图像包含有异物，提取图像标注的异物区域，得到异物区域子图像；将异物区域子图像输入图像识别模型，得到异物的异物类型；异物类型包括活体类型和非活体类型。
进一步的，所述装置还包括：异物标注模块；所述异物标注模块配置成：如果图像包含异物，定位异物对应的目标位置；以预设标注方式在显示行驶区域的界面上标注目标位置。
本申请实施例还提供一种仓库管理系统，系统包括：服务器、摄像装置和目标机器人；服务器分别与摄像装置、目标机器人通信连接；服务器配置成执行如第一方面所述的仓库管理方法的步骤。
进一步的，上述至少一个摄像装置安装于目标机器人的指定位置，以对目标机器人对应的行驶区域进行图像采集；和/或，至少一个摄像装置安装于仓库的指定位置，以对目标机器人对应的行驶区域进行图像采集。
进一步的，上述仓库为立体仓库，目标机器人为堆垛机，至少一个摄像装置安装于堆垛机的两侧轮组上方；和/或，至少一个摄像装置安装于立体仓库的巷道两侧。
本申请实施例还提供一种电子设备，包括处理器和存储器，存储器存储有能够被处理器执行的计算机可执行指令，处理器执行计算机可执行指令以实现上述仓库管理方法。
本申请实施例还提供一种计算机可读存储介质，计算机可读存储介质存储有计算机可执行指令，计算机可执行指令在被处理器调用和执行时，计算机可执行指令促使处理器实现上述仓库管理方法。
附图说明
为了更清楚地说明本申请的技术方案，下面将对其中所需要使用的附图作简单地介绍，应当理解，以下附图仅示出了本申请的某些实现方式，因此不应被看作是对范围的限定，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它相关的附图。
图1为本申请实施例提供的一种电子设备的结构示意图；
图2为本申请实施例提供的一种仓库管理方法的流程图；
图3为本申请实施例提供的一种异物检测方法的流程图；
图4为本申请实施例提供的另一种仓库管理方法的流程图；
图5为本申请实施例提供的一种仓库管理方法的工作流程图；
图6为本申请实施例提供的一种仓库管理装置的结构框图；
图7为本申请实施例提供的另一种仓库管理装置的结构框图；
图8为本申请实施例提供的一种仓库管理系统的结构框图。
具体实施方式
下面将结合实施例对本申请的技术方案进行清楚、完整地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。
相关技术中所采用的立体仓库，堆垛机需要通过运行轨道将货物运送到货架，从而完成货物入库，或者通过运行轨道将货架上的货物运送至其它地方，从而完成出库。因此，运行轨道的异物检测对保证堆垛机的安全稳定运行尤为重要。相关技术中主要采用红外检测方式实现轨道异物检测功能，但是红外检测精度不够，导致轨道异物不能被及时发现，很有可能会引起堆垛机的设备损坏，造成经济损失和仓库管理系统的局部失效。基于此，本申请实施例提供一种仓库管理方法、装置、系统及电子设备，为便于理解，以下对本申请实施例进行详细介绍。
首先，参照图1来描述用于实现本申请实施例的仓库管理方法、装置及系统的示例电子设备100。
如图1所示的一种电子设备的结构示意图，电子设备100包括一个或多个处理器102、一个或多个存储装置104、输入装置106、输出装置108以及图像采集装置110，这些组件通过总线系统112和/或其它形式的连接机构(未示出)互连。应当注意，图1所示的电子设备100的组件和结构只是示例性的，而非限制性的，根据需要，所述电子设备可以具有图1示出的部分组件，也可以具有图1未示出的其他组件和结构。
所述处理器102可以是中央处理单元(CPU)或者具有数据处理能力和/或指令执行能力的其它形式的处理单元，并且可以控制所述电子设备100中的其它组件以执行期望的功能。
所述存储装置104可以包括一个或多个计算机程序产品，所述计算机程序产品可以包括各种形式的计算机可读存储介质，例如易失性存储器和/或非易失性存储器。所述易失性存储器例如可以包括随机存取存储器(RAM)和/或高速缓冲存储器(cache)等。所述非易失性存储器例如可以包括只读存储器(ROM)、硬盘、闪存等。在所述计算机可读存储介质上可以存储一个或多个计算机程序指令，处理器102可以运行所述程序指令，以实现下文所述的本申请实施例中(由处理器实现)的客户端功能以及/或者其它期望的功能。在所述计算机可读存储介质中还可以存储各种应用程序和各种数据，例如所述应用程序使用和/或产生的各种数据等。
所述输入装置106可以是用户用来输入指令的装置，并且可以包括键盘、鼠标、麦克 风和触摸屏等中的一个或多个。
所述输出装置108可以向外部(例如，用户)输出各种信息(例如，图像或声音)，并且可以包括显示器、扬声器等中的一个或多个。
所述图像采集装置110可以是全景摄像机，也可以是普通摄像机，可以是安装于仓库中巷道的指定位置的多个摄像机，也可以是安装于仓库中的目标机器人的指定位置的摄像机。本申请实施例中的图像采集装置能够拍摄仓库中用户期望得到的图像(例如各个巷道、各个货架、各个目标机器人行驶区域的照片、视频等)，并且将所拍摄的图像存储在所述存储装置104中以供其它组件使用。
示例性地，用于实现根据本申请实施例的一种仓库管理方法、装置及系统的示例电子设备可以被实现为诸如服务器、智能手机、平板电脑、计算机等终端上。
图2为本申请实施例提供的一种仓库管理方法，该方法应用于服务器，通常为仓库管理系统中的服务器，该服务器与至少一个摄像装置通信连接；该摄像装置可以是多个设置于仓库中指定位置的摄像机，其安装数量与安装位置与仓库的内部结构有关，可以在仓库内的多个位置分别安装摄像机，以采集仓库内各区域的图像，或者也可以是安装于机器人上的摄像机。上述摄像装置可以对仓库中的机器人的行驶区域进行图像采集。
上述仓库管理方法主要包括以下步骤S202～步骤S206：
步骤S202，在目标机器人运行过程中，接收目标摄像装置采集到的目标机器人当前对应的行驶区域的图像。
上述目标摄像装置为上述至少一个摄像装置之一，其拍摄视角指向目标机器人当前对应的行驶区域，因而可以采集到该目标机器人对应的行驶区域的图像。上述目标摄像装置的安装位置可以有多种，在一种实施方式中，目标摄像装置可以安装于目标机器人上，其拍摄视角指向该目标机器人对应的行驶区域，随着目标机器人的移动而移动，从而可以连续拍摄该目标机器人对应的行驶区域的图像。在另一种实施方式中，上述目标摄像装置还可以安装于仓库中的指定位置，比如，各个巷道的两侧或者货架上某个位置，其拍摄视角指向巷道，即可拍摄经过该巷道的机器人的图像。应理解，目标机器人在不同的行驶区域运行时，可以从不同的摄像装置接收其采集到的图像。
上述目标机器人可以是不同类型的搬运机器人，比如堆垛机，翻板式机器人、辊筒式机器人、顶升式机器人、牵引式机器人和叉车等。
具体实施时，首先控制上述目标机器人启动运行，在运行过程中，通过上述目标摄像装置实时采集该目标机器人的行驶区域的图像，该行驶区域可以是目标堆垛机可行驶的轨道所在区域。
步骤S204，检测图像是否包含异物；其中，异物为行驶区域中的非固定物。
在通过上述目标摄像装置实时采集到目标机器人的行驶区域的图像后，快速进行图像检测分析，判断该图像中是否存在异物。具体的检测方式有多种，比如，通过图像与预先保存的正常的没有异物的图像进行比对，诸如，实际应用中可以预先存储每个没有异物的行驶区域的正常图像，通过比对采集到的图像和与其同属一个行驶区域的预存正常图像进行相似度比对，从而判别采集到的图像是否包含异物；或者也可以通过预设训练好的模型对图像进行目标检测，该训练好的模型可以是基于深度学习算法训练出的模型，比如，通过对CNN神经网络进行训练，训练所得的模型能够准确判别图像中是否存在异物，输出异物判别结果，还可以在图像中标记异物位置。在实际应用中，可以通过标注有真实的异物检测结果(诸如是否包含异物、异物位置等)的区域图像作为训练样本，通过训练样本对上述模型进行训练，基于预设的损失函数以及反向传播算法调整模型参数，直至模型能够输出符合预期的异物检测结果时确定训练结束。
目标机器人的行驶区域一般来说为巷道或轨道，巷道或轨道上的物品都将阻碍机器人行驶，因此都可以被视为异物，该异物即为行驶区域中的非固定物，如掉落的货物、零件、活物等。
步骤S206，如果图像中包含异物，控制目标机器人停止运行。
当检测出图像中包含异物时，及时控制目标机器人暂停运行，有多种实现方式，比如，可以直接向目标机器人发送关闭指令，以使目标机器人停止运行，或者可以向目标机器人发送暂停指定时长的控制指令，以使目标机器人暂停运行，待工作人员对异物进行处理后，再继续行驶。其中，该指定时长可以根据需求而灵活设置，通常基于平时工作人员的常规异物处理时间而定，诸如大于常规的异物处理时间。
本申请实施例提供的仓库管理方法，能够在目标机器人运行的过程中，接收到目标摄像装置采集的目标机器人当前对应的行驶区域的图像；并对该目标机器人对应的行驶区域的图像进行异物检测，确定该图像中包含异物时，及时控制目标机器人暂停运行。本申请实施例提供的上述方法可以及时发现机器人运行路径上的异物，从而有效地避免设备损坏或其它经济损失。
为了能够对目标机器人的行驶区域进行实时全面地监控，便于对后续检测出的异物进行位置追踪，本申请实施例中，上述至少一个摄像装置可分别安装于仓库的指定位置，比如，仓库中巷道的两侧或者货架的某个指定位置，能够对仓库中所有的巷道、货架等各区域进行图像采集。
基于上述摄像装置的布设方式，上述服务器还预存有至少一个摄像装置与行驶区域的对应关系，比如，摄像装置1对应行驶区域A，摄像装置2对应行驶区域B，依此类推。 上述接收目标摄像装置采集到的目标机器人当前对应的行驶区域的图像的步骤可以通过以下方式实现：
根据上述对应关系和目标机器人当前对应的行驶区域，确定目标摄像装置；获取目标摄像装置采集的图像。
比如，当目标机器人在运行过程中，来到了行驶区域B，那么可以基于上述的摄像装置与行驶区域的对应关系，确定出目标摄像装置为摄像装置2，然后获取摄像装置2采集的图像，即可获取到目标机器人当前对应的行驶区域的图像。
为了提升异物检测的效率和精准度，本申请实施例的服务器中预先配置有异物检测模型，该异物检测模型是通过大量训练样本对神经网络进行训练得到的，其可以用于进行图像的异物检测和异物标注。本申请实施例中，上述检测图像是否包含异物的步骤可参考图3所示的异物检测方法的流程图实现，具体参照如下步骤S302～步骤S304：
步骤S302，将图像输入异物检测模型。
步骤S304，如果异物检测模型输出的图像标注有异物区域，确定图像包含有异物。
由于上述异物检测模型是运用深度学习的算法，事先采集数据训练所得到的一个能检测和分割出异物的模型。因此，运用这个模型可以得到异物所在图像的确切像素区域，而且会在输出的图像中标注有异物区域的标识，比如，长方形框、正方形框等，或者，也可以直接输出异物在图像中的像素点坐标，通过坐标来标记异物所在的位置区域。
在将上述图像输入异物检测模型后，异物检测模型会输出一个图像，如果异物检测模型输出的图像中标注有异物区域，则可以确定该图像包含有异物。
通过上述方式进行异物的检测，可以在保证图像检测效率的同时，提高检测的精准度，从而更好地保证目标机器人稳定安全地工作。
另外，在上述检测到图像包含异物时，还可以向目标摄像装置发送关闭指令，以使目标摄像装置停止工作。这样可以避免该目标摄像装置继续采集有异物的图像，避免浪费功耗且进行无意义的图像采集。
在检测出图像中包含异物后，为了进一步提高异物处理效率，使工作人员能够快速进行异物处理，进而使目标机器人快速恢复正常工作，本申请实施例还可以在确定有异物之后进行报警。示例地，通过图4所示的方法的流程图中的步骤S402～步骤S410，可以实现对上述异物的报警过程：
步骤S402，如果图像包含有异物，提取图像标注的异物区域，得到异物区域子图像；
步骤S404，将异物区域子图像输入图像识别模型，得到异物的异物类型；异物类型包括活体类型和非活体类型。
上述图像识别模型可以是通过深度学习训练得到的、用于识别对象的类型的模型；该 模型所包含的神经网络结构可以有多种形式，在此不做具体限定。
在将上述异物区域对应的子图像输入到图像识别模型后，可以得到图像中上述异物的异物类型，即可确定该异物是活体类型还是非活体类型。
步骤S406，获取目标摄像装置采集的图像之前指定时长内的视频帧序列。
上述指定时长为预设的一个时间长度，对于非活体类型的异物，该指定时长可基于仓库内货架高度对应的物体自由落体时间确定的。这样，可以保证在指定时长内的视频帧序列中，可以找到掉落的物体的初始存放位置的图像，也就可以确定出图像中异物的初始存放位置。
而对于活体类型的异物，该指定时长可以是另一种设定值，可以相对时间长一些，以便确定出该活体的追踪轨迹信息，进而确定出该活体是如何进入仓库的。
步骤S408，如果异物的异物类型为非活体类型，基于视频帧序列确定异物的初始存放位置，并向第一指定终端发送第一通知信息；第一通知信息携带有初始存放位置对应的提示信息；
在仓库中，非活体类型一般指货物，但也可能是仓库中的零部件或手机之类的遗失物，如果在目标机器人的行驶区域的图像中检测出非活体类型的异物，说明可能是有货物或零件从货架上滑落，或某人的手机掉落等。进一步通过上述指定时长内的视频帧序列，查找到该异物的初始存放位置。
服务器确定出上述异物的初始存放位置后，就可以向第一指定终端发送第一通知信息，第一指定终端可以是负责整个仓库的管理的工作人员对应的智能终端，也可以是负责该行驶区域的货架管理的工作人员对应的智能终端，第一通知信息可以是短信消息，或者也可以是邮件信息，还可以是通过即时通讯软件发送的消息等，信息通知形式在此不进行限制。该第一通知信息中还可以携带包含该异物的图像，以方便工作人员进行处理。
由于第一通知信息中包含该异物的初始存放位置，因此，可以使异物处理的工作人员第一时间快速地将该货物放回其初始存放位置，或者在货物存在损坏问题时进行新的货物补充，或者对零部件或手机进行相应的处理，从而避免商品交付时造成额外的双方损失，另外，还可以加快目标机器人恢复正常工作的进程。
步骤S410，如果异物的异物类型为活体类型，基于视频帧序列确定异物的跟踪轨迹信息，并向第二指定终端发送第二通知信息；第二通知信息携带有异物的跟踪轨迹信息。
在仓库中，如果有活体类型的异物，有可能是闯入仓库的小动物，更有可能是在目标机器人启动运行时，还有人员在仓库中工作或逗留，这种情况下，可以向第二指定终端发送第二通知信息，该第二通知信息中包含有异物的追踪轨迹信息，可以是该人员的定位信息，以便使进行异物处理的工作人员根据该定位信息及时找到该异物并进行处理。该第二 通知信息中还可以携带包含该异物的图像，以方便工作人员进行处理。
需要说明的是，上述第一指定终端和第二指定终端可以是同一个终端，也可以是不同的终端，可根据实际情况进行不同的设定。
对于上述异物类型是活体类型的情况，如果目标机器人在行驶过程中撞到该活体类型的异物，很可能会出现安全事故，因此，这种情况下，紧急程度相对高一些，需要及时地进行报警操作，本申请实施例中，上述服务器还连接有报警装置，报警装置包括：语音报警器和/或信号灯报警器；如果服务器确定异物类型为活体类型，直接触发报警装置进行报警。上述报警装置的报警方式可以包括：语音报警和/或信号灯报警。
在实施例中，如果服务器检测出目标机器人的行驶区域的图像中有异物时，可以直接控制目标机器人停止运行，为了进一步缩短目标机器人恢复正常工作的时间，可以通过以下两种方式(1)和(2)之一控制目标机器人停止运行：
(1)向目标机器人发送关闭指令，以使目标机器人停止运行。
当服务器检测出目标机器人的行驶区域的图像中有异物时，直接向目标机器人发送关闭指令，以使该目标机器人停止工作。这种情况下，在异物处理结束后，需要服务器再次向目标机器人发送开启指令，从而恢复正常运行，或者通过人工手动方式开启该机器人。这种方式的好处在于可以及时地使目标机器人重新开始工作，缩短目标机器人恢复正常工作的时间。
(2)向目标机器人发送暂停指定时长的控制指令，以使目标机器人暂停运行。
为了避免异物还未处理完就使目标机器人恢复正常工作的情况发生，上述暂停指定时长相对来说会设置较长一些，通常而言可以预先统计常规情况下工作人员的异物处理时间，然后设定暂停指定时长比异物处理时间更长即可，以确保异物完全处理结束之后使目标机器人再自动开启运行。这种方式的好处是减少一次服务器重新发送开启指令的步骤，或者减少一次人工开启机器人的操作，减少了服务器的功耗，提高服务器的处理效率。
另外，为了进一步提高异物处理效率，还可以对异物的位置进行终端界面显示，具体实施中，上述方法还可以包括：如果图像包含异物，定位异物对应的目标位置；以预设标注方式在显示行驶区域的界面上标注目标位置。
上述预设标注方式包括以下至少之一：指定颜色框标注、加粗框标注和闪烁框标注。通过采用上述标注方式，可以清楚地在终端界面上显示出异物位置，以便于相关工作人员在第一时间定位异物并及时处理。
为了方便工作人员区分异物类型，以采取不同的处理手段，本申请实施例还可以通过以下方式进行异物标注：首先确定异物的异物类型；然后根据异物类型确定标注方式；以确定的标注方式在显示行驶区域的界面上标注目标位置。比如，对于活体类型的异物，用 红框圈出来，对于非活体类型的异物，通过绿色框圈出来等。
下面以堆垛机为例，对本申请实施例的仓库管理方法的流程进行详细说明：
参见图5所示的工作流程图，具体包括以下步骤S1～步骤S15：
S1：服务器向堆垛机发送开机指令；
S2：堆垛机根据上述开启指令正常启动运行；
S3：堆垛机正常启动运行后，服务器向摄像机发送开机指令；
S4：摄像机执行上述开机指令，摄像机开启，进行图像采集；
S5：摄像机持续扫描，实时传送图像到服务器；
S6：服务器存储图像A并实时对图像A进行异物检测；
S7：服务器判断有异物，向堆垛机发送暂停指令；
S8：服务器判断有异物，向摄像机发送关机指令；
S9：服务器对图像A进行图像识别，判断异物类型；
S10：服务器确定异物类型为活体类型，向堆垛机发送预警指令；
S11：堆垛机根据预警指令通过其上的报警装置进行报警；
S12：服务器对活体类型的异物进行定位，将定位信息发送至用户端；
S13：服务器确定异物类型为非活体类型，获取相关视频进行分析，确定货物的初始存放位置，将相关视频和货物的初始存放位置发送至用户端；
S14：通过用户端接收到通知的工作人员进行异物处理；
S15：人工处理完异物后，向服务器发送确认信息；
后续服务器继续执行步骤S1，重新向堆垛机发送开启指令，以使堆垛机继续运行。需要说明的是，上述只是一种具体的实施过程，其中有些步骤可以改变或调整。
本申请实施例提供的仓库管理方法，相比于现有的红外检测方式，可以准确地判断目标机器人的行驶区域是否存在异物，有效地防止堆垛机的设备损坏，从而避免经济损失和仓库管理系统的局部失效；能够准确地识别该异物的类型，并针对不同的类型进行不同的预警提醒，提醒信息中包括有异物的定位信息或者初始存放位置，以便工作人员可以及时、方便地进行异物处理，提高工作人员的异物处理效率，并且缩短目标机器人恢复正常工作的时间。另外，本申请实施例无需设置高密度的红外检测装备，成本较低。
基于上述方法实施例，本申请实施例还提供一种仓库管理装置，该装置应用于服务器；服务器与至少一个摄像装置通信连接；参见图6所示，该装置包括图像接收模块602、异物检测模块604和机器人控制模块606：
图像接收模块602，配置成在目标机器人运行过程中，接收目标摄像装置采集到的目 标机器人当前对应的行驶区域的图像，目标摄像装置的拍摄视角指向目标机器人当前对应的行驶区域；目标摄像装置为至少一个摄像装置之一；异物检测模块604，配置成检测图像是否包含异物；其中，异物为行驶区域中的非固定物；机器人控制模块606，配置成在异物检测模块的检测结果为是时，控制目标机器人停止运行。
进一步的，上述至少一个摄像装置分别安装于仓库的指定位置；上述服务器预存有至少一个摄像装置与行驶区域的对应关系；上述图像接收模块602还配置成：根据对应关系和目标机器人当前对应的行驶区域，确定目标摄像装置；获取目标摄像装置采集的图像。
参见图7所示的另一种仓库管理装置的结构框图，仓库管理装置还包括：异物报警模块608；该异物报警模块608配置成：获取目标摄像装置采集的图像之前指定时长内的视频帧序列；如果异物的异物类型为非活体类型，基于视频帧序列确定异物的初始存放位置，并向第一指定终端发送第一通知信息；第一通知信息携带有初始存放位置对应的提示信息；如果异物的异物类型为活体类型，基于视频帧序列确定异物的跟踪轨迹信息，并向第二指定终端发送第二通知信息；第二通知信息携带有异物的跟踪轨迹信息。
进一步的，上述服务器还连接有报警装置，报警装置包括：语音报警器和/或信号灯报警器；上述异物报警模块608还配置成：如果异物的异物类型为活体类型，触发报警装置进行报警。
进一步的，上述服务器预先配置有异物检测模型；上述异物检测模块604还配置成：将图像输入异物检测模型；如果异物检测模型输出的图像标注有异物区域，确定图像包含有异物。
进一步的，上述装置还包括类型判断模块610，该类型判断模块610配置成：如果图像包含有异物，提取图像标注的异物区域，得到异物区域子图像；将异物区域子图像输入图像识别模型，得到异物的异物类型；异物类型包括活体类型和非活体类型。
进一步的，上述机器人控制模块606，还配置成：向目标机器人发送关闭指令，以使目标机器人停止运行；向目标机器人发送暂停指定时长的控制指令，以使目标机器人暂停运行。
进一步的，上述装置还包括异物标注模块612；该异物标注模块612配置成：如果图像包含异物，定位异物对应的目标位置；以预设标注方式在显示行驶区域的界面上标注目标位置。
进一步的，上述异物标注模块612还配置成：确定异物的异物类型；根据异物类型确定标注方式；以确定的标注方式在显示行驶区域的界面上标注目标位置。
进一步的，上述目标机器人为目标堆垛机，行驶区域为目标堆垛机可行驶的轨道所在区域。
本申请实施例提供的仓库管理装置，其实现原理及产生的技术效果和前述仓库管理方法实施例相同，为简要描述，仓库管理装置的实施例部分未提及之处，可参考前述仓库管理方法实施例中相应内容。
基于上述方法实施例，本申请实施例还提供一种仓库管理系统，参见图8所示的仓库管理系统的结构框图，该系统包括：服务器82、摄像装置84和目标机器人86；服务器82分别与摄像装置84、目标机器人86通信连接；服务器82用于执行上述方法实施例中的仓库管理方法的步骤。
进一步的，上述至少一个摄像装置84安装于目标机器人86的指定位置，以对目标机器人86对应的行驶区域进行图像采集；和/或，至少一个摄像装置84安装于仓库的指定位置，以对目标机器人86对应的行驶区域进行图像采集。
在一种可能的实施方式中，上述仓库为立体仓库，上述目标机器人86为堆垛机，至少一个摄像装置84安装于堆垛机的两侧轮组上方；和/或，至少一个摄像装置84安装于立体仓库的巷道两侧。
在另一种可能的实施方式中，上述系统还包括用户端，比如服务器中预存手机号对应的终端、预存邮件对应的终端等。
需要说明的是，摄像装置的安装位置包括但不限于上述提供的安装位置。
本申请实施例提供的仓库管理系统，其实现原理及产生的技术效果和前述仓库管理方法实施例相同，为简要描述，仓库管理系统的实施例部分未提及之处，可参考前述仓库管理方法实施例中相应内容。
本申请实施例还提供了一种计算机可读存储介质，该计算机可读存储介质存储有计算机可执行指令，该计算机可执行指令在被处理器调用和执行时，该计算机可执行指令促使处理器实现上述仓库管理方法，具体实现可参见前述方法实施例，在此不再赘述。
本申请实施例所提供的仓库管理方法、装置、系统和电子设备的计算机程序产品，包括存储了程序代码的计算机可读存储介质，所述程序代码包括的指令可用于执行前面方法实施例中所述的方法，具体实现可参见方法实施例，在此不再赘述。
除非另外具体说明，否则在这些实施例中阐述的部件和步骤的相对步骤、数字表达式和数值并不限制本申请的范围。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个处理器可执行的非易失的计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的 形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。
在本申请的描述中，需要说明的是，术语“中心”、“上”、“下”、“左”、“右”、“竖直”、“水平”、“内”、“外”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本申请和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本申请的限制。此外，术语“第一”、“第二”、“第三”仅用于描述目的，而不能理解为指示或暗示相对重要性。
最后应说明的是：以上所述实施例，仅为本申请的具体实施方式，用以说明本申请的技术方案，而非对其限制，本申请的保护范围并不局限于此，尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，其依然可以对前述实施例所记载的技术方案进行修改或可轻易想到变化，或者对其中部分技术特征进行等同替换；而这些修改、变化或者替换，并不使相应技术方案的本质脱离本申请实施例技术方案的精神和范围，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应所述以权利要求的保护范围为准。
工业实用性
本申请提出的技术方案中，在目标机器人运行过程中，可以接收到拍摄视角指向该目标机器人当前的行驶区域的目标摄像装置所采集到的图像，进而对该目标机器人当前对应的行驶区域的图像进行检测，判断该图像中是否包含异物；该异物为行驶区域中的非固定物；如果检测到图像中有异物，则及时控制目标机器人停止运行，可以有效地避免设备损坏或其它经济损失。

Claims (19)

  1. 一种仓库管理方法，其特征在于，所述方法应用于服务器；所述服务器与至少一个摄像装置通信连接；所述方法包括：
    在目标机器人运行过程中，接收目标摄像装置采集到的所述目标机器人当前对应的行驶区域的图像，所述目标摄像装置的拍摄视角指向所述目标机器人当前对应的行驶区域；所述目标摄像装置为所述至少一个摄像装置之一；
    检测所述图像是否包含异物；其中，所述异物为所述行驶区域中的非固定物；
    如果是，控制所述目标机器人停止运行。
  2. 根据权利要求1所述的仓库管理方法，其特征在于，所述至少一个摄像装置分别安装于所述仓库的指定位置；所述服务器预存有所述至少一个摄像装置与行驶区域的对应关系；
    所述接收所述目标摄像装置采集到的所述目标机器人当前对应的行驶区域的图像，包括：
    根据所述对应关系和所述目标机器人当前对应的行驶区域，确定所述目标摄像装置；
    获取所述目标摄像装置采集的图像。
  3. 根据权利要求1或2所述的仓库管理方法，其特征在于，所述方法还包括：
    获取所述目标摄像装置采集的所述图像之前指定时长内的视频帧序列；
    如果所述异物的异物类型为非活体类型，基于所述视频帧序列确定所述异物的初始存放位置，并向第一指定终端发送第一通知信息；所述第一通知信息携带有所述初始存放位置对应的提示信息；
    如果所述异物的异物类型为活体类型，基于所述视频帧序列确定所述异物的跟踪轨迹信息，并向第二指定终端发送第二通知信息；所述第二通知信息携带有所述异物的跟踪轨迹信息。
  4. 根据权利要求1至3中任一项所述的仓库管理方法，其特征在于，所述服务器还连接有报警装置，所述报警装置包括：语音报警器和/或信号灯报警器；
    所述方法还包括：
    如果所述异物的异物类型为活体类型，触发所述报警装置进行报警。
  5. 根据权利要求1至4中任一项所述的仓库管理方法，其特征在于，所述服务器预先配置有异物检测模型；
    所述检测所述图像是否包含异物，包括：
    将所述图像输入所述异物检测模型；
    如果所述异物检测模型输出的所述图像标注有异物区域，确定所述图像包含有异物。
  6. 根据权利要求5所述的仓库管理方法，其特征在于，所述方法还包括：
    如果所述图像包含有异物，提取所述图像标注的异物区域，得到异物区域子图像；
    将所述异物区域子图像输入图像识别模型，得到所述异物的异物类型；所述异物类型包括活体类型和非活体类型。
  7. 根据权利要求1至6中任一项所述的仓库管理方法，其特征在于，控制所述目标机器人停止运行的步骤，包括以下之一：
    向所述目标机器人发送关闭指令，以使所述目标机器人停止运行；
    向所述目标机器人发送暂停指定时长的控制指令，以使所述目标机器人暂停运行。
  8. 根据权利要求1至7中任一项所述的仓库管理方法，其特征在于，所述方法还包括：
    如果所述图像包含异物，定位所述异物对应的目标位置；
    以预设标注方式在显示所述行驶区域的界面上标注所述目标位置。
  9. 根据权利要求8所述的仓库管理方法，其特征在于，所述以预设标注方式在显示所述行驶区域的界面上标注所述目标位置，包括：
    确定所述异物的异物类型；
    根据所述异物类型确定标注方式；
    以确定的所述标注方式在显示所述行驶区域的界面上标注所述目标位置。
  10. 根据权利要求1至9中任一项所述的仓库管理方法，其特征在于，所述目标机器人为目标堆垛机，所述行驶区域为所述目标堆垛机可行驶的轨道所在区域。
  11. 一种仓库管理装置，其特征在于，所述装置应用于服务器；所述服务器与至少一个摄像装置通信连接；所述仓库管理装置包括：
    图像接收模块，配置成在目标机器人运行过程中，接收目标摄像装置采集到的所述目标机器人当前对应的行驶区域的图像，所述目标摄像装置的拍摄视角指向所述目标机器人当前对应的行驶区域；所述目标摄像装置为所述至少一个摄像装置之一；
    异物检测模块，配置成检测所述图像是否包含异物；其中，所述异物为所述行驶区域中的非固定物；
    机器人控制模块，配置成在所述异物检测模块的检测结果为是时，控制所述目标机器人停止运行。
  12. 根据权利要求11所述的仓库管理装置，其特征在于，所述装置还包括：异物报警模块；
    所述异物报警模块配置成：
    获取目标摄像装置采集的图像之前指定时长内的视频帧序列；如果异物的异物类型为 非活体类型，基于视频帧序列确定异物的初始存放位置，并向第一指定终端发送第一通知信息；第一通知信息携带有初始存放位置对应的提示信息；如果异物的异物类型为活体类型，基于视频帧序列确定异物的跟踪轨迹信息，并向第二指定终端发送第二通知信息；第二通知信息携带有异物的跟踪轨迹信息。
  13. 根据权利要求11或12所述的仓库管理装置，其特征在于，所述装置还包括：类型判断模块；
    所述类型判断模块配置成：如果图像包含有异物，提取图像标注的异物区域，得到异物区域子图像；将异物区域子图像输入图像识别模型，得到异物的异物类型；异物类型包括活体类型和非活体类型。
  14. 根据权利要求11至13任一项所述的仓库管理装置，其特征在于，所述装置还包括：异物标注模块；
    所述异物标注模块配置成：如果图像包含异物，定位异物对应的目标位置；以预设标注方式在显示行驶区域的界面上标注目标位置。
  15. 一种仓库管理系统，其特征在于，所述仓库管理系统包括：服务器、摄像装置和目标机器人；
    所述服务器分别与所述摄像装置、所述目标机器人通信连接；
    所述服务器配置成执行如权利要求1-10任一项所述的仓库管理方法的步骤。
  16. 根据权利要求15所述的仓库管理系统，其特征在于，至少一个摄像装置安装于所述目标机器人的指定位置，以对所述目标机器人对应的行驶区域进行图像采集；和/或，所述至少一个摄像装置安装于所述仓库的指定位置，以对所述目标机器人对应的行驶区域进行图像采集。
  17. 根据权利要求16所述的仓库管理系统，其特征在于，所述仓库为立体仓库，所述目标机器人为堆垛机，所述至少一个摄像装置安装于所述堆垛机的两侧轮组上方；和/或，所述至少一个摄像装置安装于所述立体仓库的巷道两侧。
  18. 一种电子设备，其特征在于，包括处理器和存储器，所述存储器存储有能够被所述处理器执行的计算机可执行指令，所述处理器执行所述计算机可执行指令以实现权利要求1至10任一项所述的仓库管理方法。
  19. 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机可执行指令，所述计算机可执行指令在被处理器调用和执行时，计算机可执行指令促使处理器实现权利要求1至10任一项所述的仓库管理方法。
PCT/CN2021/102441 2020-10-30 2021-06-25 仓库管理方法、装置、系统及电子设备 WO2022088716A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011194452.XA CN112394690B (zh) 2020-10-30 2020-10-30 仓库管理方法、装置、系统及电子设备
CN202011194452.X 2020-10-30

Publications (1)

Publication Number Publication Date
WO2022088716A1 true WO2022088716A1 (zh) 2022-05-05

Family

ID=74598614

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/102441 WO2022088716A1 (zh) 2020-10-30 2021-06-25 仓库管理方法、装置、系统及电子设备

Country Status (2)

Country Link
CN (1) CN112394690B (zh)
WO (1) WO2022088716A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115180364A (zh) * 2022-08-01 2022-10-14 宁夏广天夏科技股份有限公司 基于gmi磁传感器的矿用输送带异物监测装置及方法
CN117854211A (zh) * 2024-03-07 2024-04-09 南京奥看信息科技有限公司 一种基于智能视觉的目标对象识别方法及装置

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112394690B (zh) * 2020-10-30 2022-05-17 北京旷视机器人技术有限公司 仓库管理方法、装置、系统及电子设备
CN113145468A (zh) * 2021-03-18 2021-07-23 浙江博城机器人科技有限公司 基于多目垃圾分类机器人的生活垃圾分拣方法
CN113651059A (zh) * 2021-07-28 2021-11-16 北京旷视机器人技术有限公司 异常处理方法、装置、物料传输系统和机器人
CN114422704A (zh) * 2022-01-27 2022-04-29 深圳市商汤科技有限公司 电子设备的控制方法、装置、设备及介质
CN115659452B (zh) * 2022-09-21 2024-04-12 联通数字科技有限公司 一种智能巡视方法、智能巡视系统及计算机可读存储介质

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012053838A (ja) * 2010-09-03 2012-03-15 Hitachi Plant Technologies Ltd 無人搬送車および走行制御方法
CN104787084A (zh) * 2015-04-16 2015-07-22 北京交通大学 一种铁路异物侵限检测系统及检测方法
CN106695792A (zh) * 2017-01-05 2017-05-24 中国计量大学 基于机器视觉的码垛机器人跟踪监控系统及方法
CN107229903A (zh) * 2017-04-17 2017-10-03 深圳奥比中光科技有限公司 机器人避障的方法、装置及存储装置
CN107885206A (zh) * 2017-11-07 2018-04-06 中国科学院福建物质结构研究所 一种机器人避障方法及应用其的机器人
CN109409238A (zh) * 2018-09-28 2019-03-01 深圳市中电数通智慧安全科技股份有限公司 一种障碍物检测方法、装置及终端设备
CN109607031A (zh) * 2019-01-14 2019-04-12 青岛舍科技有限公司 基于无人机全景的智能仓储系统及方法
CN110765922A (zh) * 2019-10-18 2020-02-07 华南理工大学 一种agv用双目视觉物体检测障碍物系统
CN112394690A (zh) * 2020-10-30 2021-02-23 北京旷视机器人技术有限公司 仓库管理方法、装置、系统及电子设备

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105799740B (zh) * 2016-03-08 2017-06-16 浙江大学 一种基于物联网技术的轨道异物入侵自动检测与预警方法
US10339497B2 (en) * 2016-07-21 2019-07-02 Ebay Inc. System and method for dynamic inventory management
US10579949B2 (en) * 2016-11-16 2020-03-03 Wipro Limited Methods and systems for facilitating transportation of articles in warehouses using video analytics
CN108202965A (zh) * 2016-12-16 2018-06-26 东莞市海柔智能科技有限公司 自动化仓储管理方法、装置和系统
CN108876251A (zh) * 2018-07-05 2018-11-23 北京智行者科技有限公司 一种物流派送作业方法
CN110824498A (zh) * 2018-08-07 2020-02-21 杭州海康机器人技术有限公司 障碍物检测方法、装置及系统
CN110303500B (zh) * 2019-07-10 2020-09-29 中信梧桐港供应链管理有限公司 一种仓库机器人控制系统及方法
CN110852244A (zh) * 2019-11-06 2020-02-28 深圳创维数字技术有限公司 车辆控制方法、装置及计算机可读存储介质
CN111680609B (zh) * 2020-06-03 2023-02-07 合肥中科类脑智能技术有限公司 基于图像配准和目标检测的异物识别系统及其方法

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012053838A (ja) * 2010-09-03 2012-03-15 Hitachi Plant Technologies Ltd 無人搬送車および走行制御方法
CN104787084A (zh) * 2015-04-16 2015-07-22 北京交通大学 一种铁路异物侵限检测系统及检测方法
CN106695792A (zh) * 2017-01-05 2017-05-24 中国计量大学 基于机器视觉的码垛机器人跟踪监控系统及方法
CN107229903A (zh) * 2017-04-17 2017-10-03 深圳奥比中光科技有限公司 机器人避障的方法、装置及存储装置
CN107885206A (zh) * 2017-11-07 2018-04-06 中国科学院福建物质结构研究所 一种机器人避障方法及应用其的机器人
CN109409238A (zh) * 2018-09-28 2019-03-01 深圳市中电数通智慧安全科技股份有限公司 一种障碍物检测方法、装置及终端设备
CN109607031A (zh) * 2019-01-14 2019-04-12 青岛舍科技有限公司 基于无人机全景的智能仓储系统及方法
CN110765922A (zh) * 2019-10-18 2020-02-07 华南理工大学 一种agv用双目视觉物体检测障碍物系统
CN112394690A (zh) * 2020-10-30 2021-02-23 北京旷视机器人技术有限公司 仓库管理方法、装置、系统及电子设备

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115180364A (zh) * 2022-08-01 2022-10-14 宁夏广天夏科技股份有限公司 基于gmi磁传感器的矿用输送带异物监测装置及方法
CN115180364B (zh) * 2022-08-01 2024-02-27 宁夏广天夏科技股份有限公司 基于gmi磁传感器的矿用输送带异物监测装置及方法
CN117854211A (zh) * 2024-03-07 2024-04-09 南京奥看信息科技有限公司 一种基于智能视觉的目标对象识别方法及装置
CN117854211B (zh) * 2024-03-07 2024-05-28 南京奥看信息科技有限公司 一种基于智能视觉的目标对象识别方法及装置

Also Published As

Publication number Publication date
CN112394690B (zh) 2022-05-17
CN112394690A (zh) 2021-02-23

Similar Documents

Publication Publication Date Title
WO2022088716A1 (zh) 仓库管理方法、装置、系统及电子设备
CN110974088B (zh) 扫地机器人控制方法、扫地机器人及存储介质
TWI438702B (zh) 影像環境邊界之動態設定方法及人員活動內容之即時判定方法
JP6900918B2 (ja) 学習装置及び学習方法
CN111597879A (zh) 一种基于监控视频的姿态检测方法、装置及系统
CN103093177A (zh) 人脸辨识监控管理方法
CN115600953A (zh) 仓库库位的监控方法、装置、计算机设备和存储介质
KR102233679B1 (ko) Ess 침입자 및 화재 감지 장치 및 방법
CN114332925A (zh) 电梯内宠物检测方法、系统、装置及计算机可读存储介质
CN111325937B (zh) 翻越行为检测方法、装置和电子系统
JP2012198802A (ja) 侵入物検出システム
US20230368177A1 (en) Graphic code display method, terminal and storage medium
CN113505704A (zh) 图像识别的人员安全侦测方法、系统、设备及存储介质
US20120026292A1 (en) Monitor computer and method for monitoring a specified scene using the same
JP2019102941A (ja) 画像処理装置、画像処理装置の制御方法
KR102451479B1 (ko) 타워크레인의 연결 상태를 감지하는 스마트 안전 시스템 및 체결 관리 장치의 제어 방법
WO2022198507A1 (zh) 障碍物检测方法、装置、设备和计算机存储介质
JP2020037464A (ja) 判定装置および判定方法
Pietrini et al. Embedded Vision System for Real-Time Shelves Rows Detection for Planogram Compliance Check
CN114666476B (zh) 机器人智能录像方法、装置、设备及存储介质
CN116269336A (zh) 一种跌倒检测方法、装置、控制器及存储介质
KR102439548B1 (ko) 발전소 감시 시스템 및 방법
KR102586132B1 (ko) 산업현장의 쓰러짐 감지 감시카메라 시스템 및 이의 동작 방법
CN113645449B (zh) 工业互联网的数据采集方法、系统及计算机可读存储介质
CN112541452A (zh) 一种针对可疑人员的管控方法、系统及设备

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: 21884457

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21884457

Country of ref document: EP

Kind code of ref document: A1