Book packaging method and device based on readable field, server and medium

Abstract

The application provides a book packaging method and device based on a suitable reading field, a server and a medium, comprising: receiving book order information; identifying the suitable reading field corresponding to each book category in the to-be-packaged book list; generating a first book allocation strategy based on the suitable reading field and the in-stored book list corresponding to each book warehouse; for each target book warehouse, dividing the to-be-packaged book sub-list corresponding to the target book warehouse into multiple picking lists based on the suitable reading field, generating a second book allocation strategy based on the number range of schedulable robots and the current state information of each book sorting robot in the target book warehouse; and controlling the target book sorting robot to perform a book packaging action according to the corresponding picking list according to the second book allocation strategy. The application can significantly improve the efficiency and automation of book sorting, and also helps to improve the experience of sellers, buyers and warehouse management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of book packaging, and in particular to a book packaging method and device based on a suitable reading field, a server and a medium. BACKGROUND

[0002] The traditional book management software system has the deficiency in the classification of the new curriculum standard extracurricular reading system. Specifically, with the implementation of the new curriculum standard, the classification of the extracurricular reading system has gradually become an important part of primary and secondary education. However, the traditional book management software system has not been updated in time and has not covered this new classification system, resulting in that the book goods cannot be accurately identified as suitable for the age group in the system. This defect brings many inconveniences to the seller, the buyer and the warehouse management. In addition, the common means of book sorting at present is that the warehouse management personnel manually sort and package the corresponding books according to the sales order, resulting in low efficiency and low automation of book sorting. SUMMARY

[0003] Therefore, the purpose of the present application is to provide a book packaging method and device based on a suitable reading field, a server and a medium, which can significantly improve the efficiency and automation of book sorting and also help to improve the experience of the seller, the buyer and the warehouse management.

[0004] In a first aspect, the present application provides a book packaging method based on a suitable reading field. The method is applied to a server in a book storage system, the book storage system further comprising a book sorting robot deployed in each book warehouse, the server being in communication connection with the book sorting robot, and the method comprising:

[0005] receiving book order information, the book order information at least comprising a to-be-packaged book list, the to-be-packaged book list being used to describe the book category and the quantity information of each age group corresponding to the book category;

[0006] identifying the suitable reading field corresponding to each book category in the to-be-packaged book list, the suitable reading field being used to describe the suitability of the book category with respect to the age group to which it belongs;

[0007] based on the suitable reading field and the in-warehouse book list corresponding to each book warehouse, taking the minimum number of allocated book warehouses and the minimum number of book sorting robots dispatched in each book warehouse as the target, generating a first book allocation strategy, the first book allocation strategy being used to describe at least one target book warehouse, the corresponding to-be-packaged book sub-list and the number range of the schedulable robots;

[0008] For each target warehouse, based on the readability field, the to-be-packed book sub-list corresponding to the target warehouse is divided into multiple picking orders, and based on the number of schedulable robots and the current state information of each book sorting robot in the target warehouse, a second book allocation strategy is generated, which is used to describe the target book sorting robot and its corresponding picking order in the target warehouse;

[0009] According to the second book allocation strategy, the target book sorting robot is controlled to perform a book packaging action according to the corresponding picking order.

[0010] In an embodiment, based on the readability field and the in-warehouse book list corresponding to each warehouse, a first book allocation strategy is generated with the goal of minimizing the number of allocated warehouses and the number of book sorting robots scheduled in each warehouse, including:

[0011] According to the readability constraint, based on the readability field corresponding to each book category in the to-be-packed book list and the in-warehouse book list corresponding to each warehouse, an initial book allocation strategy set is generated, and the readability constraint is that the book categories corresponding to the same readability field and the same age group are preferentially allocated to the same warehouse.

[0012] For any candidate book allocation strategy in the book allocation strategy set, a fitness value corresponding to the candidate book allocation strategy is generated with the goal of minimizing the number of warehouses, minimizing the number of book sorting robots scheduled in each warehouse, and maximizing the concentration of readability fields.

[0013] Based on the fitness value, multiple candidate book allocation strategies are selected from the book allocation strategy set, and a cross and mutation operation is performed on the selected book allocation strategies to obtain new candidate book allocation strategies to form a new book allocation strategy set. When the iteration stopping condition is met, the candidate book allocation strategy corresponding to the maximum fitness value in the book allocation strategy set at this time is selected as the first book allocation strategy.

[0014] In an embodiment, the mechanical arm of the book sorting robot is equipped with a whole book gripper and a single book gripper at the end; based on the number of schedulable robots and the current state information of each book sorting robot in the target warehouse, a second book allocation strategy is generated, including:

[0015] The book packaging task of the picking order is published to the task pool;

[0016] For any one book packaging task in the task pool, a competition request initiated by each book sorting robot in the target book warehouse for the book packaging task is received, the task cost of each book sorting robot is determined based on the current state information carried by the competition request, and the target book sorting robot corresponding to the picking list of the book packaging task is determined according to the task cost of each book sorting robot and the number range of schedulable robots, and the book packaging task is eliminated from the task pool; wherein the current state information includes the distance between the current position of the book sorting robot and the position information corresponding to the book packaging task, the current use of the book sorting robot The type of grab, and the power information of the book sorting robot;

[0017] Continue to determine the target book sorting robot corresponding to the picking list of the next book packaging task until the task pool is empty, and obtain a second book allocation strategy.

[0018] In an embodiment, according to the second book allocation strategy, the target book sorting robot is controlled to perform a book packaging action according to the picking list corresponding thereto, comprising:

[0019] Obtain the warehouse grid model corresponding to the target book warehouse, and the warehouse grid model describes the occupancy state of each grid in the target book warehouse by specifying a marker;

[0020] Based on the warehouse grid model, the current position of the target book sorting robot and the picking list corresponding thereto, the picking path corresponding to the target book sorting robot is planned;

[0021] Based on the picking path, the time for the target book sorting robot to reach each grid in the target book warehouse is predicted to determine whether there are multiple target book sorting robots simultaneously occupying the same grid;

[0022] If so, the target book sorting robots simultaneously occupying the same grid are controlled to pause or avoid obstacles, so that the target book sorting robots successfully reach the grid corresponding to the picking list and perform the book packaging action according to the picking list corresponding thereto.

[0023] In an embodiment, the book sorting robot is equipped with a vision device; performing a book packaging action according to the picking list corresponding thereto, comprising:

[0024] After the target book sorting robot successfully reaches the grid corresponding to the picking list, the shelf image data collected by the vision device is obtained;

[0025] Based on the shelf image data and the position information described in the picking list, the book to be grabbed is located, and whether the grab type currently used by the target book sorting robot matches the book to be grabbed is identified based on the shelf image data;

[0026] If no, switch the current gripper type used by the target book sorting robot, and perform the picking action and the book packing action on the to-be-picked book using the switched gripper.

[0027] In an embodiment, the book storage system further comprises a book monitoring device communicatively connected to the server, the book monitoring device comprising a network of pressure sensors and a network of cameras deployed on each shelf in the book warehouse, the network of cameras comprising a top layer fisheye lens, a middle layer wide-angle lens, and a bottom layer depth camera; the method further comprises:

[0028] sensing, by the network of pressure sensors, pressure changes corresponding to each shelf board in the shelf to monitor whether a book-in event or a book-out event occurs in the shelf based on the pressure changes;

[0029] If yes, locate the target shelf and shelf board where the book-in event or the book-out event occurs, activate the network of cameras corresponding to the target shelf and shelf board, acquire a set of book images collected by the activated network of cameras for the target shelf and shelf board, the set of book images comprising an occupancy heat map collected by the top layer fisheye lens, book image data collected by the middle layer wide-angle lens, and book point cloud data collected by the bottom layer depth camera;

[0030] updating the in-stored book list corresponding to the book warehouse based on the set of book images.

[0031] In an embodiment, updating the in-stored book list corresponding to the book warehouse based on the set of book images comprises:

[0032] In the case of a book-in event, the following operations are performed:

[0033] identifying, based on the occupancy heat map, an in-stored event area of the target shelf and shelf board where the book-in event occurs;

[0034] cutting, from the book image data, first sub-image data corresponding to the in-stored event area, and identifying, based on the first sub-image data, book spine line data and book ISBN data corresponding to the newly in-stored book;

[0035] determining, based on the book spine line data and the book point cloud data, position information corresponding to the newly in-stored book;

[0036] updating the in-stored book list based on the book ISBN data and the position information corresponding to the newly in-stored book;

[0037] In the case of a book-out event, the following operations are performed:

[0038] identifying, based on the occupancy heat map, an out-stored event area of the target shelf and shelf board where the book-out event occurs;

[0039] cutting out a second sub-image region corresponding to the library event region from the library image data, and identifying, based on the second sub-image region and a previous frame second sub-image region, book ISBN data and location information corresponding to the book that has been checked out;

[0040] updating the list of books that have been checked in based on the book ISBN data and the location information corresponding to the book that has been checked out.

[0041] In a second aspect, the present application further provides a book packaging device based on a suitable reading field, which is applied to a server in a book storage system, the book storage system further comprising a book sorting robot arranged in each book room, and the server is in communication connection with the book sorting robot, and the device comprises:

[0042] an order receiving module, configured to receive book order information, the book order information at least comprising a list of books to be packaged, the list of books to be packaged being used to describe a book category corresponding to each age group and quantity information thereof;

[0043] a suitable reading field identifying module, configured to identify a suitable reading field corresponding to each book category in the list of books to be packaged, the suitable reading field being used to describe the suitability of the book category relative to the age group to which the book category belongs;

[0044] a first allocation module, configured to generate a first book allocation strategy based on the suitable reading field and a list of books that have been checked in corresponding to each book room, with the goal of minimizing the number of book rooms to be allocated and the number of book sorting robots to be dispatched in each book room, the first book allocation strategy being used to describe at least one target book room, a sub-list of books to be packaged corresponding to the target book room, and a range of the number of schedulable robots;

[0045] a second allocation module, configured to, for each target book room, divide the sub-list of books to be packaged corresponding to the target book room into a plurality of picking orders based on the suitable reading field, and generate a second book allocation strategy based on the range of the number of schedulable robots and current state information of each book sorting robot in the target book room, the second book allocation strategy being used to describe a target book sorting robot to be dispatched in the target book room and a picking order corresponding to the target book sorting robot;

[0046] a book packaging module, configured to control the target book sorting robot to perform a book packaging action according to the picking order corresponding to the target book sorting robot according to the second book allocation strategy.

[0047] In a third aspect, the present application further provides a server, comprising a processor and a memory, the memory storing computer executable instructions capable of being executed by the processor, and the processor executes the computer executable instructions to implement the method of any one of the first aspect.

[0048] In a fourth aspect, the present application also provides a computer readable storage medium, which stores computer executable instructions, and the computer executable instructions, when invoked and executed by a processor, cause the processor to implement any of the methods provided in the first aspect.

[0049] The present application provides a book packaging method and device based on a suitable reading field, a server and a medium, which are applied to a server in a book storage system, the book storage system also comprises a book sorting robot arranged in each book warehouse, and the server is in communication connection with the book sorting robot. First, book order information is received, and the book order information at least comprises a to-be-packaged book list, the to-be-packaged book list is used to describe the book categories corresponding to each age group and the quantity information thereof; then, the suitable reading field corresponding to each book category in the to-be-packaged book list is identified, and the suitable reading field is used to describe the suitability of the book category relative to the age group to which the book category belongs; on this basis, based on the suitable reading field and the in-warehouse book list corresponding to each book warehouse, a first book allocation strategy is generated, the first book allocation strategy is used to describe at least one target book warehouse, the to-be-packaged book sub-list corresponding to the target book warehouse and the number range of the schedulable robot; then, for each target book warehouse, the to-be-packaged book sub-list corresponding to the target book warehouse is divided into a plurality of picking orders based on the suitable reading field, and a second book allocation strategy is generated based on the number range of the schedulable robot and the current state information of each book sorting robot in the target book warehouse, the second book allocation strategy is used to describe the target book sorting robot scheduled in the target book warehouse and the picking order corresponding to the target book sorting robot; finally, the target book sorting robot is controlled to perform a book packaging action according to the picking order corresponding to the target book sorting robot according to the second book allocation strategy. The above method identifies the suitable reading field corresponding to each book category in the to-be-packaged book list, and then generates the first book allocation strategy based on the suitable reading field, taking the book warehouse as the research object, generates the second book allocation strategy based on the book sorting robot as the research object, and controls the book sorting robot to perform the book packaging action according to the picking order. The present application not only significantly improves the efficiency and automation degree of book sorting, but also helps to improve the experience of the seller, the buyer and the warehouse management on the basis of the suitable reading field.

[0050] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application will be realized and achieved by the structure particularly pointed out in the description, claims and drawings.

[0051] In order to make the above objectives, characteristics and advantages of the present application more apparent, comprehensible and easier to understand, the following preferred embodiments are specifically described below with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to make the above objectives, characteristics and advantages of the present application more apparent, comprehensible and easier to understand, the following preferred embodiments are specifically described below with reference to the accompanying drawings.

[0053] Figure 1 A flowchart of a book packaging method based on a suitable reading field provided by an embodiment of the present application;

[0054] Figure 2 A core picking strategy diagram of a book packaging method based on a suitable reading field provided by an embodiment of the present application;

[0055] Figure 3 A structure diagram of a book packaging device based on a suitable reading field provided by an embodiment of the present application;

[0056] Figure 4 A structure diagram of a server provided by an embodiment of the present application. DETAILED DESCRIPTION

[0057] In order to make the above objectives, characteristics and advantages of the present application more apparent, comprehensible and easier to understand, the following preferred embodiments are specifically described below with reference to the accompanying drawings.

[0058] (1) Inconvenience of salesperson: Salesperson cannot accurately divide the suitable reading section for each grade of primary and secondary schools, so as to accurately recommend books for each age group. Therefore, there are the following problems: Inaccurate recommendation: When recommending books to schools, libraries or parents, the salesperson cannot provide accurate recommendations according to the specific age group of students, resulting in poor recommendation effect; Low sales efficiency: Due to the lack of clear suitable reading section information, the salesperson needs to spend more time on manual screening and recommendation, which reduces the sales efficiency; Customer satisfaction decreases: Inaccurate recommendation may lead customers to question the professional ability of the salesperson, affecting customer satisfaction and long-term cooperation.

[0059] (2) The inconvenience of the purchaser: the purchaser (such as a school, a library, or a parent) cannot accurately know which age group the book is suitable for, and cannot accurately know which books the current age group should read. Therefore, the following problems exist: selection difficulty: when parents choose books for their children, they cannot accurately determine whether the books are suitable for the age group of the children, which may lead to improper selection and affect the children's reading interest and effect; management confusion: after a school or a library purchases a batch of books, it needs to distinguish and screen each book variety, so as to allocate the books to corresponding grades or borrowing shelves, which increases the management difficulty and workload; resource waste: due to inaccurate classification, some books may be incorrectly allocated to an unsuitable age group, causing resource waste.

[0060] (3) The inconvenience of the book warehouse: the warehouse cannot be packaged according to the requirements of the school or the library. Therefore, the following problems exist: low packaging efficiency: when the warehouse packages books, it cannot quickly and accurately package the books according to the requirements of the school or the library, which increases the packaging time and cost; distribution error: due to the lack of clear reading field information, books may be incorrectly distributed to unsuitable schools or grades, affecting the distribution efficiency and customer satisfaction; complex inventory management: it is impossible to manage the inventory according to the reading field classification, which increases the complexity and error rate of inventory management.

[0061] Based on this, the present application provides a book packaging method, device, server and medium based on a reading field, which can significantly improve the efficiency and automation of book sorting, and also helps to improve the experience of the seller, the purchaser and the warehouse management.

[0062] In order to facilitate the understanding of the present embodiment, first, a book packaging method based on a reading field is introduced in detail, which is applied to a server in a book storage system. The book storage system also includes a book sorting robot deployed in each book warehouse. The server is in communication connection with the book sorting robot. Referring to the flowchart of the book packaging method based on the reading field shown in Figure 1 The method mainly includes the following steps S102 to S110:

[0063] Step S102, receiving book order information.

[0064] The book order information at least includes a to-be-packed book list, and the to-be-packed book list is used to describe the book category corresponding to each age group and the quantity information thereof. Books with the same name and the same publisher can be identified as belonging to the same book category. For example, a book category is "How Steel Was Made" published by People's Education Press. In addition, the book order information can also include customer information (such as enterprise name, contact person, telephone, address, credit limit, total arrears, etc.), auxiliary information (such as packing requirements, transportation mode, etc.), platform information (such as platform store name, platform order number, etc.).

[0065] In step S104, an appropriate reading field corresponding to each book category in the to-be-packed book list is identified.

[0066] The appropriate reading field is used to describe the appropriateness of the book category with respect to the age group to which it belongs. In an example, the server pre-stores an appropriate reading field corresponding to each book category. By reading the book category contained in the to-be-packed book list, the appropriate reading field corresponding to each book category in the to-be-packed book list can be identified. For example, the appropriate reading field can be divided into first appropriate reading, second appropriate reading, third appropriate reading, teacher appropriate reading, etc. Different appropriate reading fields can be set based on actual needs, and the embodiments of the present application do not limit this.

[0067] In step S106, based on the appropriate reading field and the in-stored book list corresponding to each book warehouse, a first book allocation strategy is generated, with the number of allocated book warehouses being the smallest and the number of book sorting robots dispatched in each book warehouse being the smallest as the target.

[0068] The in-stored book list is used to describe the book category and the quantity information and the position information of the book category that have been stored in the book warehouse. The first book allocation strategy is used to describe at least one target book warehouse and the corresponding to-be-packed book sub-list and the number range of the schedulable robots. The to-be-packed book sub-list is also the book category and the quantity information of the book category that need to be executed by the target book warehouse. The number range of the schedulable robots is also the number range of the book sorting robots that can be scheduled by the target book warehouse when executing the book packing action.

[0069] In an example, an initial book allocation strategy set can be generated based on the corresponding suitability field of each book category in the to-be-packed book list and the corresponding in-stocked book list of each book warehouse according to the pre-configured suitability constraint, the book allocation strategy set including a plurality of candidate book allocation strategies, the candidate book allocation strategy being used to describe at least one candidate book warehouse and a corresponding to-be-packed book sub-list and a number range of schedulable robots; a fitness function considering the number of book warehouses, the number of scheduled book sorting robots in each book warehouse, and the suitability field concentration is used to calculate the fitness value corresponding to each candidate book allocation strategy, the suitability field concentration being used to describe whether the book categories corresponding to the same age group and the same suitability field are allocated to as few book warehouses as possible, and the smaller the number of allocated book warehouses, the higher the value of the suitability field concentration; the candidate book allocation strategies are screened, crossed, and mutated based on the fitness value to form a new book allocation strategy set; after the candidate book allocation strategies in the new book allocation strategy set that do not meet the suitability constraint are removed, the fitness value corresponding to each candidate book allocation strategy in the new book allocation strategy set is recalculated, and the process is repeated until the iteration stopping condition (such as reaching a preset iteration stopping number) is met, and the candidate book allocation strategy with the highest fitness value in the book allocation strategy set at this time is taken as the first book allocation strategy.

[0070] In step S108, for each target book warehouse, the to-be-packed book sub-list corresponding to the target book warehouse is divided into a plurality of picking orders based on the suitability field, and a second book allocation strategy is generated based on the number range of schedulable robots and the current state information of each book sorting robot in the target book warehouse.

[0071] The current state information includes the distance between the current position of the book sorting robot and the position information corresponding to the book packing task, the type of the current gripper used by the book sorting robot, and the power information of the book sorting robot. The second book allocation strategy is used to describe the target book sorting robot scheduled in the target book warehouse and the corresponding picking order of the target book sorting robot. The picking order is used to describe the number of book categories to be sorted by the target book sorting robot and the position of the book categories. Specifically, the position of the book categories can be described by the shelf, the level in the shelf, and the specific coordinates in the level.

[0072] In an example, for each target book warehouse, the same book category is preferentially divided into the same picking order, and if the number of a certain book category is greater than a preset packing number, the book category can be divided into two or more picking orders; if the number of a certain book category is less than the preset packing number, the book categories of the same age group and the same suitability field can be divided into the same picking order.

[0073] On this basis, the book sorting task corresponding to the picking list can be published to a task pool, for each book sorting task in the task pool, a competition request uploaded by each book sorting robot in the target book warehouse for the book sorting task can be received, and then a corresponding target book sorting robot for the book sorting task is allocated according to the current state information of each book sorting robot, the book sorting task is eliminated from the task pool, the target book sorting robot corresponding to the next book sorting task in the task pool is determined, and the process is repeated until the task pool is empty, that is, the second book allocation strategy is obtained.

[0074] In step S110, the target book sorting robot is controlled to perform the book packaging action according to the corresponding picking list according to the second book allocation strategy.

[0075] In an example, each target book sorting robot moves in the target book warehouse according to the corresponding picking list, collision detection is performed during the movement, and in the case that there is a collision between two or more target book sorting robots, the related target book sorting robot is controlled to pause or detour, so that each target book sorting robot safely reaches the position described in the corresponding picking list; after the target book sorting robot reaches the position described in the corresponding picking list, the vision device carried thereon can be used to realize positioning of the to-be-grabbed book, determining whether to switch the currently used grab, etc., so as to realize grabbing and packaging of the to-be-grabbed book.

[0076] The book packaging method based on the read field provided by the embodiment of the application, by identifying the read field corresponding to each book category in the to-be-packaged book list, and then generating a first book allocation strategy based on the read field, taking the book warehouse as the research object, generating a second book allocation strategy taking the book sorting robot as the research object, and controlling the book sorting robot to perform the book packaging action according to the picking list, the application not only significantly improves the efficiency and automation degree of book sorting, but also helps to improve the experience of the seller, the buyer and the warehouse management.

[0077] On the basis of the foregoing embodiment, the embodiment of the application provides a core picking strategy of the book packaging method based on the read field, referring to Figure 2The core picking strategy diagram of the book packing method based on the readable field includes: (1) determining the target book warehouse according to the specified range priority order of the book warehouse and whether the inventory of the book warehouse meets the requirements; (2) splitting the picking list according to the readable field (such as the first / second / third readable field and the teacher applicable field) under a certain target book warehouse; and (3) calculating the optimal picking path of the target book warehouse based on the picking list, including each target book picking robot and the corresponding picking list and picking path thereof. When the number of book categories is too large, 1 picking list is formed for every 200 categories to facilitate parallel picking.

[0078] Based on the above core picking strategy, the embodiment of the application provides a specific implementation of the book packing method based on the readable field.

[0079] In an embodiment, the book storage system further includes a book monitoring device in communication connection with the server, and the book monitoring device includes a network of pressure sensors and a network of cameras deployed on each shelf in the book warehouse, and the network of cameras includes a top layer fisheye lens, a middle layer wide-angle lens and a bottom layer depth camera.

[0080] In an example, 3-5 sensors can be installed at the bottom of each shelf of each shelf, and the distance between two adjacent sensors can be less than or equal to 30 cm, and the sensors can be high-precision thin-film pressure sensors.

[0081] In an example, the "top layer", "middle layer" and "bottom layer" in the top layer fisheye lens, the middle layer wide-angle lens and the bottom layer depth camera do not describe the height of each device, but describe the logical order of the data collected by each device in the process of updating the in-warehouse book list. For example, the specific area is first located by using the image collected by the top layer fisheye lens, then the book spine data and the book ISBN (International Standard Book Number) data are identified by using the middle layer wide-angle lens, and finally the position information of the book is determined by using the bottom layer depth camera. In a specific embodiment, the top layer fisheye lens can be a fisheye lens of model Dahua DH-IPC-EW4831-AS, the middle layer wide-angle lens can be a wide-angle lens of model Hikvision DS-2CD3347G2, and the bottom layer depth camera can be a depth camera of model Intel RealSense L515.

[0082] Based on the structure, the embodiment of the application provides a specific implementation of updating the in-warehouse book list, as follows (1.1) to (1.3):

[0083] (1.1) The pressure sensor network is used to perceive the pressure change of each shelf in the bookshelf, so as to monitor whether the bookshelf has a book-in event or a book-out event based on the pressure change.

[0084] In one example, for any pressure sensor in the pressure sensor network, the pressure change at the position of the pressure sensor can be perceived by the pressure sensor, and if the pressure change value is greater than a preset pressure threshold (for example, the weight of the removed book is less than 50g), it is determined that the bookshelf has a book-in event or a book-out event, and the pressure sensor sends a signal to the server. The positive and negative of the pressure change value can be used to describe the event type, for example, the pressure change value is positive, which means that a book-in event occurs; otherwise, the pressure change value is negative, which means that a book-out event occurs.

[0085] (1.2) If yes, the target bookshelf and shelf where the book-in event or the book-out event occurs are located, the camera network corresponding to the target bookshelf and shelf is activated, and the book image set collected by the activated camera network for the target bookshelf and shelf is obtained.

[0086] The book image set includes an occupancy heat map collected by a top fisheye lens, book image data collected by a middle wide-angle lens, and book point cloud data collected by a bottom depth camera.

[0087] In one example, the server can identify the specific position (i.e., the target bookshelf and shelf) of the pressure sensor in the book warehouse according to the identification of the pressure sensor, and then only activate the camera network corresponding to the target bookshelf and shelf to trigger the camera network corresponding to the target bookshelf and shelf to perform the image collection action, so as to reduce the computing load.

[0088] Optionally, one camera network can be set for each bookshelf, and when a book-in event or a book-out event occurs at a certain shelf, the camera network corresponding to the bookshelf to which the shelf belongs can be activated to collect the book image set; one camera network can also be set for each or multiple shelves in a bookshelf, and when a book-in event or a book-out event occurs at a certain shelf, the corresponding camera network can be activated to collect the book image set.

[0089] (1.3) Update the in-stored book list corresponding to the book warehouse according to the book image set. In specific implementation, an embodiment for updating the in-stored book list can be provided for the book-in event and the book-out event respectively.

[0090] (1.31) In the case of a book-in event, the following operations are performed:

[0091] The book-in event area of the target shelf and the layer plate where the book-in event occurs is identified based on the occupancy heat map; first sub-image data corresponding to the book-in event area is intercepted from the book image data, and book spine line data and book ISBN data corresponding to the newly-in book are identified based on the first sub-image data; the position information corresponding to the newly-in book is determined according to the book spine line data and the book point cloud data; and the in-stored book list is updated based on the book ISBN data and the position information corresponding to the newly-in book.

[0092] In a specific implementation, the top fisheye lens: quickly scans the whole row of shelves to generate an occupancy heat map, and marks the area that may change (for example, from 'high occupancy' to 'low occupancy'). The middle wide-angle lens: high-resolution scanning of the positioning layer plate, YOLOv8-Tiny model detects the book spine line position, CRNN-OCR identifies the book spine text. The bottom depth camera: scanning the stacked books to generate a three-dimensional point cloud, extracting the contour through the Hough algorithm, calculating the three-dimensional coordinates of each book, and then updating the in-stored book list in the book-in scenario.

[0093] (1.32) In the case of a book-out event, the following operations are performed:

[0094] The book-out event area of the target shelf and the layer plate where the book-out event occurs is identified based on the occupancy heat map; the second sub-image area corresponding to the book-out event area is intercepted from the book image data, and the book ISBN data and the position information corresponding to the out-stored book are identified based on the second sub-image area and the previous frame second sub-image area; and the in-stored book list is updated based on the book ISBN data and the position information corresponding to the out-stored book. In a specific implementation, the top fisheye lens: quickly scans the whole row of shelves to generate an occupancy heat map, and marks the area that may change (for example, from 'high occupancy' to 'low occupancy'). The middle wide-angle lens: high-resolution scanning of the positioning layer plate, comparing the obtained second sub-image area with the previous frame second sub-image area, identifying the book ISBN data and the position information corresponding to the book that has been stored in the area, and then updating the in-stored book list in the book-out scenario.

[0095] Further, after updating the in-stored book list, the above book image set can be sent to the associated terminal to calibrate and supplement the updated in-stored book list through the graphical user interface of the associated terminal. Calibration means calibrating the updated book information, and supplement means supplementing the unpacking condition (whole package or single book) of the updated book, which is helpful for subsequent automatic grabbing by a book sorting robot.

[0096] On the basis of the foregoing embodiments, the foregoing steps S102 to S110 are further explained and described.

[0097] For the foregoing step S102, the book order information is received, including the list of books to be packed, customer information, auxiliary information, platform information, etc.

[0098] For the foregoing step S104, the corresponding suitable reading field of each book category in the list of books to be packed is identified. The book category contained in the list of books to be packed can be directly read. For each book category, the corresponding book category is determined based on the preset mapping relationship between the book category and the suitable reading field. Optionally, the mapping relationship between the book category and the suitable reading field can be determined based on the after-school reading guidance list provided by the relevant department. For the book categories not contained in the after-school reading guidance list, the neural network can be trained using the after-school reading guidance list to enable the neural network to learn the mapping relationship between the age group, the suitable reading field, and the book category, and then output the corresponding suitable reading field of the book category not contained in the after-school reading guidance list using the trained neural network model.

[0099] For the foregoing step S106, based on the suitable reading field and the in-stock book list corresponding to each book warehouse, the specific process of generating the first book allocation strategy is as follows:

[0100] (2.1) According to the suitability constraint, based on the suitable reading field corresponding to each book category in the list of books to be packed, and the in-stock book list corresponding to each book warehouse, an initial set of book allocation strategies is generated.

[0101] Among them, the suitability constraint is: the book categories corresponding to the same age group and the same suitable reading field are preferentially allocated to the same book warehouse. In specific implementation, the same warehouse is preferentially allocated, that is, if a certain book warehouse has allocated a certain book category under the same "age group-suitable field", then other book categories in the same group are preferentially allocated to the warehouse, for example, if warehouse 1 has allocated book A of "grade 3-textbook", then book C of the same "grade 3-textbook" should also be preferentially allocated to warehouse 1. Further, for each (grade, suitable field, category) combination, a list of candidate book warehouses with sufficient inventory to cover demand is screened, and preferably, if a certain candidate book warehouse has allocated other book categories of the same "grade-suitable field", then its priority is promoted to the top of the feasible list.

[0102] In one case, each individual (i.e. alternative book allocation strategy) is represented as an array of length N, N being the total number of (grade, suitability field, book category) combinations, each element representing the warehouse ID to which the combination is allocated, and must belong to its feasible warehouse set. Example: combination 1 (grade 3 - textbook - book A) is allocated to warehouse 1; combination 2 (grade 3 - supplementary - book B) is allocated to warehouse 2. Based on this, the initial set of book allocation strategies can be generated according to the following process: first, group each (grade, suitability field), try to allocate all book categories within the group to the same book warehouse (if the book warehouse has enough inventory); if not, randomly select from the feasible book warehouses, but prefer to select the book warehouse that already contains other categories of the same group. Example: if book A and book C of "grade 3 - textbook" need to be allocated to the same book warehouse, and book warehouse 1 can meet the needs of both, then both will be allocated to book warehouse 1 in the initial solution.

[0103] (2.2) For any one of the alternative book allocation strategies in the set of book allocation strategies, the fitness value corresponding to the alternative book allocation strategy is generated, aiming to minimize the number of book warehouses, minimize the number of book sorting robots dispatched in each book warehouse, and maximize the concentration of suitability field groups.

[0104] In specific implementation, the embodiment of the present application mainly considers the following objectives in the process of determining the fitness value corresponding to the alternative book allocation strategy:

[0105] Objective 1: Minimize the number of warehouses (unique warehouse number);

[0106] Objective 2: Minimize the number of robots (i.e. total number of packages, number of packages = ⌈demand / 200⌉);

[0107] Objective 3: Maximize the concentration of "age group - suitability field" groups in the warehouse (reduce the number of dispersed warehouses within the group).

[0108] Based on the above objectives, the embodiment of the present application provides an expression of a fitness function:

[0109] Fitness function = warehouse number * W1 + robot number * W2 + dispersion penalty term * W3;

[0110] Wherein, the dispersion penalty term is: count the number of different warehouses to which the categories within each "age group - suitability field" group are allocated, the greater the total, the higher the penalty. Weight setting: W1 >> W3 > W2, to ensure that the number of warehouses is prioritized, followed by the concentration within the group, and finally the number of robots.

[0111] (2.3) Selecting multiple candidate book allocation strategies from the book allocation strategy set based on the fitness value, and performing crossover and mutation operations on the selected book allocation strategies to obtain new candidate book allocation strategies, to form a new book allocation strategy set, until the iteration stopping condition is met, the candidate book allocation strategy corresponding to the maximum fitness value in the book allocation strategy set at this time is selected as the first book allocation strategy.

[0112] Optionally, select: select multiple optimal candidate book allocation strategies from the candidate book allocation strategies in order from high to low fitness value. Group crossover: cut and color the chromosome according to the "age range - moderate field" group, exchange the warehouse allocation of the same group of parents, and keep the consistency within the group. For example: Parent 1 allocates the "Grade 3 - Textbook" group to Book Warehouse 1, and Parent 2 allocates it to Book Warehouse 2. After exchange, the offspring may inherit the allocation of the entire group. Mutation: In-group mutation: With a certain probability, all book categories of a certain "age range - moderate field" group are migrated to another feasible book warehouse; single-point mutation: randomly select a certain (grade, moderate field, category) combination and switch to another book warehouse in its feasible warehouse list (preferably a book warehouse that already contains books of the same group).

[0113] The new book allocation strategy set is formed through the above selection, crossover and mutation operations, and the process is repeated until the iteration stopping condition is met, and the candidate book allocation strategy corresponding to the maximum fitness value is selected as the first book allocation strategy.

[0114] For example, the first book allocation strategy is as follows:

[0115] (I) Target warehouse list: all target book warehouses allocated tasks;

[0116] (II) Target book warehouse and its corresponding to-be-packed book sub-list and number of schedulable robots: for each target book warehouse, list its allocated (grade, moderate field, category) combination, quantity required, and number of packages (number of robots). For example:

[0117] Warehouse 1:

[0118] Grade 3 | Moderate field "First Readable" | Book A (500 copies -> 3 packages);

[0119] Grade 3 | Moderate field "First Readable" | Book C (200 copies -> 1 package);

[0120] Number of schedulable robots = 4 ( ).

[0121] Further, the mechanical arm end of the book sorting robot is equipped with a whole book grabbing clamp and a single book grabbing clamp, wherein the whole book grabbing clamp is suitable for grabbing unopened whole books (for example, 10 books per package), and the single book grabbing clamp is suitable for grabbing opened single books. Based on this, for the foregoing step S108, based on the schedulable robot quantity range and the current state information of each book sorting robot in the target book warehouse, the specific process of generating a second book allocation strategy is as follows:

[0122] (3.1) Publish the book packaging task of the picking list to the task pool, and mark each book packaging task as "unallocated". The book packaging task describes the book category (book name, publisher) corresponding to the picking list, the shelf location (shelf and level in the warehouse), and the required grabbing type (whole book grabbing or single book grabbing).

[0123] (3.2) For any book packaging task in the task pool, receive the competition request initiated by each book sorting robot in the target book warehouse for the book packaging task, determine the task cost of each book sorting robot based on the current state information carried by the competition request, and determine the target book sorting robot corresponding to the picking list of the book packaging task according to the task cost of each book sorting robot and the schedulable robot quantity range, and eliminate the book packaging task from the task pool.

[0124] In specific implementation, the book category (book name, publisher) described by the book packaging task, the shelf location (shelf and level in the warehouse), and the required grabbing type (whole book grabbing or single book grabbing) need to be obtained, and the current state information of each book sorting robot also needs to be obtained, which includes the distance between the current position of the book sorting robot and the position information corresponding to the book packaging task, the grabbing type currently used by the book sorting robot, and the power information of the book sorting robot.

[0125] Based on the above information, the task cost of each book sorting robot for the book packaging task is determined, and the lower the task cost, the more suitable the book sorting robot is for executing the book sorting task. The calculation formula of the task cost is as follows:

[0126] Task cost = path length weight × path distance + switching clamp weight × switching times + power penalty term;

[0127] Wherein, path distance: the total distance from the current position of the robot to the shelf to the packaging area; switching times: if the required grabbing clamp is different from the current one, then the number +1, otherwise 0; power penalty term: if the path distance exceeds the maximum distance supported by the remaining power, the cost is set to infinity (prohibit allocation).

[0128] In a specific implementation, the book sorting robot with the lowest task cost can be bound to the book sorting task, and the book sorting task can be marked as "assigned" or removed from the task pool.

[0129] (3.3) Continue to determine the target book sorting robot corresponding to the picking list of the next book packing task until the task pool is empty, and obtain a second book allocation strategy. In a specific implementation, the foregoing (3.2) is repeated until all book sorting tasks in the task pool are marked as "assigned", or the task pool does not contain a book sorting task, and the final second book allocation strategy is obtained.

[0130] For example, it is assumed that the task pool includes: picking list 1: book A (whole package gripper, shelf S1, path distance 30 meters). Picking list 2: book B (single book gripper, shelf S2, path distance 50 meters). Wherein, the robot state is: book sorting robot R1 supports 100 meters of power, the current gripper is a whole package, located at the entrance of the warehouse, book sorting robot R2 supports 80 meters of power, the current gripper is a single book, located in the middle of the shelf area. The bidding process is as follows: picking list 1: R1 cost = 30 x 1 (path) + 0 (no need to switch) = 30; R2 cost = 40 x 1 (path) + 1 x 10 (switch gripper) = 50, that is, book sorting robot R1 wins. Picking list 2: R1 cost = 55 x 1 (path) + 1 x 10 (switch gripper) = 65; R2 cost = 50 x 1 (path) + 0 = 50, that is, book sorting robot R2 wins. Based on this, the following second book allocation strategy is output: the book sorting task bound to the book sorting robot R1 is the picking list 1 (book A, whole package gripper); the book sorting task bound to the book sorting robot R2 is the picking list 2 (book B, single book gripper).

[0131] For the foregoing step S110, according to the second book allocation strategy, the target book sorting robot performs the specific process of the book packing action according to the picking list corresponding thereto as follows:

[0132] (4.1) Obtain the warehouse grid model corresponding to the target book warehouse.

[0133] Wherein, the warehouse grid model describes the occupancy state of each grid in the target book warehouse by specifying a label. For example, for the grid where the shelf is located, the grid is permanently labeled as "1", that is, the grid is occupied; for other grids, if a book sorting robot is passing through the grid, the grid is labeled as "1", that is, the grid is occupied, if no book sorting robot is passing through the grid at this time, the grid is labeled as "0", that is, the grid is idle (passable).

[0134] (4.2) Based on the warehouse grid model, the current position of the target book sorting robot and its corresponding picking list, the target book sorting robot's corresponding picking path is planned.

[0135] For example, the planning algorithm of the picking path can adopt A* algorithm. The specific process is as follows:

[0136] First, define the parameters of A* algorithm parameters, including: define node: each grid unit is a node, node attributes include: coordinates (x, y); actual cost g: the moving cost from the starting point to the current node (such as 1 per grid); heuristic cost h: the estimated cost of the current node to the end point (manhattan distance or euclidean distance); total cost f = g + h. Define the movement rule: allow the robot to move in the four directions (four-neighborhood) or increase the diagonal movement (eight-neighborhood); the cost of diagonal movement is set to √2 (closer to the actual distance); prohibit entering the grid marked as 1 (permanent or temporary occupation).

[0137] Then, the picking path is planned:

[0138] (I) Input parameters: starting point (i.e. the grid coordinates corresponding to the current position of the robot), end point (i.e. the grid coordinates where the target shelf is located), real-time warehouse grid model (including all occupation marks).

[0139] (II) Initialize data structure: open list (Open List): store nodes to be explored, sorted in ascending order of f value; closed list (Closed List): store explored nodes; parent node mapping (Parent Map): record the optimal predecessor node of each node, used for path backtracking.

[0140] (III) Algorithm execution: add the starting point to the open list, set its g = 0, h is calculated according to the heuristic, and f = g + h. Loop the following steps: a. Select the current node: take out the node with the smallest f value from the open list. b. Termination condition: if the current node is the end point, exit the loop and backtrack the path; if the open list is empty, the path does not exist. c. Generate adjacent nodes: check the four-neighborhood / eight-neighborhood grid of the current node. d. Filter invalid nodes: ignore the grid marked as 1 (occupied); ignore the nodes already in the closed list. e. Update node cost: calculate the g value of the adjacent node (current node g + moving cost); if the adjacent node is not in the open list, or the new g value is lower, update its g, h, f, and set the current node as the parent node; add the adjacent node to the open list (if it does not exist). f. Move the current node to the closed list.

[0141] (IV) Path backtracking: trace back the parent nodes from the end node to the starting point to generate the final path coordinate sequence.

[0142] It should be noted that the above-mentioned picking path planning algorithm is only an example, and those skilled in the art can select other path planning algorithms to generate a picking path corresponding to each target book picking robot, and the embodiments of the present application do not limit this.

[0143] (4.3) Based on the picking path, the time for the target book picking robot to reach each grid in the target book warehouse is predicted to determine whether there are multiple target book picking robots simultaneously occupying the same grid.

[0144] (4.4) If so, the target book picking robots simultaneously occupying the same grid are controlled to pause or avoid obstacles to enable the target book picking robots to successfully reach the grid corresponding to the picking order and perform book packaging actions according to the corresponding picking order.

[0145] In the case of multiple target book picking robots simultaneously occupying the same grid, the priority of each target book picking robot can be determined to ensure that the target book picking robot with the highest priority passes through the grid to successfully reach the grid corresponding to the picking order. In specific implementation, the priority of multiple target book picking robots simultaneously occupying the same grid can be determined according to the following formula:

[0146] Priority = Task Remaining Time Urgency × Package Priority Coefficient + Robot Remaining Power Weight;

[0147] Wherein, the remaining time urgency = 1 / task remaining time (the shorter the time, the higher the priority); the package priority coefficient is defined by business rules (for example, the first package is limited).

[0148] In one specific embodiment, for a target book picking robot with lower priority, it can be controlled to pause, and after the target book picking robot with higher priority passes through the grid, the target book picking robot with lower priority is controlled to pass through the grid. In another specific embodiment, the target book picking robot with lower priority can also be re-planned to avoid obstacles, and the path planning process can refer to the aforementioned (4.2), and the embodiments of the present application do not repeat the description.

[0149] Further, the book picking robot is equipped with a vision device. On this basis, the embodiments of the present application further provide specific implementations for performing book packaging actions according to the corresponding picking order, including:

[0150] (I) After the target book picking robot successfully reaches the grid corresponding to the picking order, the vision device collects the shelf image data. In one example, the book picking robot can be equipped with a vision device (such as a camera, etc.), and the vision device is used to collect the shelf image data.

[0151] (II) Based on the location information described in the picking list and the shelf image data, the to-be-grabbed book is located, and based on the shelf image data, it is identified whether the current grabbing type used by the target book sorting robot matches the to-be-grabbed book. In an example, image processing can be performed on the shelf image data to extract the book bounding box contained in the picking list, and then the rough book position is obtained, and the specific operation position of the book sorting robot performing the grabbing action is determined in combination with the specific location information described in the picking list. In addition, the recognition result of the shelf image data can be combined with the unpacking state (single or whole package) of the book described in the in-warehouse book list (or picking list) to determine whether the current grabbing type used by the target book sorting robot matches.

[0152] (III) If not, the current grabbing type used by the target book sorting robot is switched, and the to-be-grabbed book is executed by the switched grabbing to perform the grabbing action and the book packaging action.

[0153] Further, the book sorting robot end can be installed with a rubber wheel type grabbing structure. In the library, books are usually arranged in rows on the bookshelf, and books are placed closely together, so the mechanical claw cannot be directly used to clamp a single book. Considering that the force control of the mechanical arm in the design of the human hand cannot achieve biological level precision, and considering that the hook will damage the book, the embodiment of the application changes the thinking and uses a rubber wheel structure. A round plate with a rubber ring is installed at the head of the mechanical arm to form an extraction structure, which can accurately separate the target book from the book pile and minimize damage to the book. The setting of the rotation angle of the servo motor is the setting of the farthest and closest distance of the book extraction, which facilitates the subsequent clamping action of the mechanical claw.

[0154] Further, in order to reduce the friction between the book and the shelf, the shelf can be made of a smoother material to facilitate the rubber wheel type grabbing structure to extract the book more easily. Further, in order to avoid the book from falling off the shelf due to too low friction between the book and the shelf, a high-transparency material partition can be provided on the shelf, and when it is detected that the book sorting robot is about to perform the grabbing action, the partition is controlled to retract or lift to expose the book on the shelf, so that the book sorting robot uses the rubber wheel type grabbing structure to extract the book and uses the corresponding type of grabbing to grab the book.

[0155] In summary, the book packaging method based on the readable field provided by the embodiment of the application makes the book commodity data information more complete, can obtain the specific number of books through the number of packages in the single preparation link, and combines the package specification adjustment sheet to facilitate the query of the package inventory of various package specifications of the library book commodity, facilitates the quick query of the number of packages without unpacking, and provides great convenience for the sales single preparation and other links. In addition, through multi-algorithm collaborative optimization, combined with business rules and automatic scheduling strategies, under the constraint of the readable field (grade-textbook type), the automation degree and efficiency of book sorting are significantly improved, at least with the following characteristics:

[0156] (1) Accurate sorting and resource concentration driven by readable field: the priority allocation mechanism based on the genetic algorithm of the embodiment of the application preferentially concentrates the books of the same readable field (such as "fifth grade-Chinese textbook") to the same book warehouse, reduces the cross-warehouse sorting demand, and reduces the subsequent robot scheduling complexity. Example: all fifth grade textbook books of a primary school are allocated to warehouse A to avoid the path intersection and coordination cost caused by being scattered to multiple warehouses. (2) Full-process automation and intelligent decision-making: the embodiment of the application automatically processes from warehouse selection to package generation (every 200 books as a package) without human intervention. The genetic algorithm and the bidding algorithm are linked to realize the optimal matching of two-level resources of warehouse-robot. Further, the embodiment of the application is based on the conflict detection algorithm of the space-time grid, which can predict path conflicts in real time, and low-priority robots can automatically bypass or pause, combined with the dynamic redistribution mechanism, to ensure continuous execution of tasks in high-load scenarios. (3) Efficiency improvement and cost optimization: the shortest sorting path, the minimum number of grabbing and clamping switching, the maximum resource utilization rate, the minimum number of warehouses, and the minimum number of robots comprehensively improve the book sorting efficiency and reduce the book sorting cost. (4) Dynamic adaptability and robustness enhancement: in the embodiment of the application, the robot power is real-time included in the bidding cost calculation, and the low-power robot is automatically allocated to a short-distance task to improve the dynamic adaptability and robustness.

[0157] On the basis of the foregoing embodiment, the embodiment of the application provides a book packaging device based on a readable field, which is applied to a server in a book storage system, the book storage system further comprising a book sorting robot deployed in each book warehouse, the server being in communication connection with the book sorting robot, and referring to a structure schematic diagram of a book packaging device based on a readable field shown in Figure 3 The device mainly comprises the following parts:

[0158] An order receiving module 302 is configured to receive book order information, and the book order information at least comprises a list of books to be packaged, the list of books to be packaged being configured to describe the book category and quantity information corresponding to each age group;

[0159] The suitable reading identification module 304 is used for identifying a suitable reading field corresponding to each book category in the to-be-packed book list, and the suitable reading field is used for describing the suitability of the book category relative to the age group to which the book category belongs.

[0160] The first distribution module 306 is used for generating a first book distribution strategy based on the suitable reading field and the in-stored book list corresponding to each book warehouse, with the number of distributed book warehouses being minimum and the number of dispatched book sorting robots in each book warehouse being minimum as targets, and the first book distribution strategy is used for describing at least one target book warehouse and a corresponding to-be-packed book sub-list and a range of the number of schedulable robots.

[0161] The second distribution module 308 is used for, for each target book warehouse, dividing the to-be-packed book sub-list corresponding to the target book warehouse into a plurality of picking orders based on the suitable reading field, and generating a second book distribution strategy based on the range of the number of schedulable robots and current state information of each book sorting robot in the target book warehouse, and the second book distribution strategy is used for describing a target book sorting robot dispatched in the target book warehouse and a corresponding picking order.

[0162] The book packing module 310 is used for controlling the target book sorting robot to perform a book packing action according to the corresponding picking order according to the second book distribution strategy.

[0163] The book packing device based on the suitable reading field provided by the embodiment of the present application identifies the suitable reading field corresponding to each book category in the to-be-packed book list, and then generates a first book distribution strategy taking the book warehouse as the research object and a second book distribution strategy taking the book sorting robot as the research object based on the suitable reading field, so as to control the book sorting robot to perform a book packing action according to the picking order. The present application not only significantly improves the efficiency and automation degree of book sorting, but also helps to improve the experience of the seller, the buyer and the warehouse management on the basis of the suitable reading field.

[0164] In an implementation manner, the first distribution module 306 is specifically configured to:

[0165] According to the suitability constraint, an initial book distribution strategy set is generated based on the suitable reading field corresponding to each book category in the to-be-packed book list and the in-stored book list corresponding to each book warehouse, and the suitability constraint is that the book categories corresponding to the same age group and the same suitable reading field are preferentially distributed to the same book warehouse.

[0166] For any one of the candidate book distribution strategies in the book distribution strategy set, an adaptive value corresponding to the candidate book distribution strategy is generated with the number of book warehouses being minimized, the number of dispatched book sorting robots in each book warehouse being minimized, and the concentration of the suitable reading field being maximized as targets.

[0167] selecting a plurality of candidate book allocation strategies from the book allocation strategy set based on the fitness values, and performing crossover and mutation operations on the selected book allocation strategies to obtain new candidate book allocation strategies, to form a new book allocation strategy set, until a stopping condition is met, and the candidate book allocation strategy corresponding to the maximum fitness value in the book allocation strategy set at this time is taken as the first book allocation strategy.

[0168] In an embodiment, the mechanical arms of the book sorting robots are each equipped with a whole-book book gripper and a single-book book gripper; and the second allocation module 308 is specifically configured to:

[0169] publishing the book packing tasks of the picking lists to a task pool;

[0170] For any one book packing task in the task pool, receiving a competition request initiated by each book sorting robot in the target book warehouse for the book packing task, determining the task cost of each book sorting robot based on the current state information carried by the competition request, determining the target book sorting robot corresponding to the picking list of the book packing task according to the task cost of each book sorting robot and the range of the number of schedulable robots, and excluding the book packing task from the task pool; wherein the current state information includes the distance between the current position of the book sorting robot and the position information corresponding to the book packing task, the type of the gripper currently used by the book sorting robot, and the power information of the book sorting robot.

[0171] continuing to determine the target book sorting robot corresponding to the picking list of the next book packing task until the task pool is empty, to obtain the second book allocation strategy.

[0172] In an embodiment, the book packing module 310 is specifically configured to:

[0173] obtain a warehouse grid model corresponding to the target book warehouse, the warehouse grid model describing the occupancy state of each grid in the target book warehouse by specifying a marker;

[0174] based on the warehouse grid model, the current position of the target book sorting robot and the picking list corresponding thereto, planning a picking path corresponding to the target book sorting robot;

[0175] based on the picking path, predicting the time for the target book sorting robot to reach each grid in the target book warehouse, to determine whether there is a case where multiple target book sorting robots simultaneously occupy the same grid;

[0176] If yes, pause control or obstacle avoidance control is performed on the target book sorting robot simultaneously occupying the same grid, so that the target book sorting robot successfully reaches the grid corresponding to the order picking list and performs book packaging actions according to the corresponding order picking list.

[0177] In an embodiment, the book packaging module 310 is specifically configured to:

[0178] After the target book sorting robot successfully reaches the grid corresponding to the order picking list, obtain the shelf image data collected by the visual equipment;

[0179] Based on the shelf image data and the position information described in the order picking list, the book to be grabbed is located, and whether the current gripping type used by the target book sorting robot matches the book to be grabbed is identified based on the shelf image data;

[0180] If no, switch the current gripping type used by the target book sorting robot, and perform the grabbing action and the book packaging action on the book to be grabbed by using the switched gripping.

[0181] In an embodiment, the book storage system further comprises a book monitoring device in communication connection with the server, the book monitoring device comprising a network of pressure sensors and a network of cameras deployed on each shelf in the book warehouse, the network of cameras comprising a top layer fisheye lens, a middle layer wide-angle lens and a bottom layer depth camera; and an inventory updating module configured to:

[0182] By means of the network of pressure sensors, the pressure change corresponding to each shelf plate in the shelf is perceived, so as to monitor whether a book storage event or a book withdrawal event occurs in the shelf based on the pressure change;

[0183] If yes, the target shelf and shelf plate where the book storage event or the book withdrawal event occurs are located, the network of cameras corresponding to the target shelf and shelf plate is activated, a set of book images of the target shelf and shelf plate collected by the activated network of cameras are obtained, the set of book images comprising an occupancy heat map collected by the top layer fisheye lens, book image data collected by the middle layer wide-angle lens and book point cloud data collected by the bottom layer depth camera;

[0184] According to the set of book images, the in-stored book list corresponding to the book warehouse is updated.

[0185] In an embodiment, the inventory updating module is specifically configured to:

[0186] In the case of a book storage event, the following operations are performed:

[0187] Based on the occupancy heat map, the storage event area of the target shelf and shelf plate where the book storage event occurs is identified;

[0188] The first sub-image data corresponding to the in-storage event area is intercepted from the book image data, and book spine line data and book ISBN data corresponding to the newly-stored book are identified based on the first sub-image data;

[0189] According to the book spine line data and the book point cloud data, position information corresponding to the newly-stored book is determined;

[0190] Based on the book ISBN data and the position information corresponding to the newly-stored book, an in-storage book list is updated;

[0191] In the case of a book out-of-storage event, the following operations are performed:

[0192] Based on the occupation heat map, an out-of-storage event area of a target shelf and a layer plate where the book out-of-storage event occurs is identified;

[0193] The second sub-image area corresponding to the out-of-storage event area is intercepted from the book image data, and book ISBN data and position information corresponding to the out-of-storage book are identified based on the second sub-image area and a previous frame second sub-image area;

[0194] Based on the book ISBN data and the position information corresponding to the out-of-storage book, the in-storage book list is updated.

[0195] The device provided by the embodiment of the application has the same implementation principle and technical effects as the foregoing method embodiment, and for brevity of description, the part not mentioned in the device embodiment can be referred to the corresponding content in the foregoing method embodiment.

[0196] The embodiment of the application provides a server, and specifically, the server comprises a processor and a storage device; the storage device stores a computer program, and the computer program performs the method according to any one of the foregoing embodiments when the computer program is run by the processor.

[0197] Figure 4 A structural diagram of a server provided by the embodiment of the application is provided, and the server 100 comprises a processor 40, a memory 41, a bus 42 and a communication interface 43, the processor 40, the communication interface 43 and the memory 41 are connected through the bus 42; the processor 40 is configured to execute an executable module stored in the memory 41, for example, a computer program.

[0198] The memory 41 can include a high-speed random access memory (RAM) and can also include a non-volatile memory, for example, at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 43 (which can be wired or wireless), and the Internet, a wide area network, a local area network, a metropolitan area network and the like can be used.

[0199] Bus 42 can be an ISA bus, a PCI bus, or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one bidirectional arrow is used in the figure, but it does not mean that there is only one bus or only one type of bus.

[0200] The memory 41 is configured to store a program, and the processor 40 executes the program after receiving an execution instruction. The method executed by the device for defining the flow process disclosed in any of the embodiments of the application can be applied to the processor 40 or implemented by the processor 40.

[0201] The processor 40 can be an integrated circuit chip with a processing capability of signals. In the implementation process, each step of the above method can be completed by integrated logic circuits of hardware in the processor 40 or instructions in the form of software. The processor 40 described above can be a general-purpose processor, including a central processing unit (CPU) and a network processor (NP), etc.; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. Each method, step and logic block diagram disclosed in the embodiments of the application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiments of the application can be directly embodied as a hardware code processor for execution, or a combination of hardware and software modules in the code processor for execution. The software module can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 41, and the processor 40 reads the information in the memory 41 and combines the hardware to complete the steps of the above method.

[0202] The computer program product of the readable storage medium provided by the embodiments of the application includes a computer readable storage medium storing program codes, and the program codes include instructions for executing the method described in the foregoing method embodiments. For specific implementation, please refer to the foregoing method embodiments, which will not be described here.

[0203] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of software products. The computer software product is stored in a storage medium and includes instructions for making a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0204] Finally, it should be noted that: the above-described embodiments are only specific embodiments of the present application, which are used to illustrate the technical solutions of the present application, but not to limit them. The protection scope of the present application is not limited to this. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily think of changes to the technical solutions recorded in the foregoing embodiments within the technical range disclosed by the present application, or make equivalent replacements to some technical features. The modifications, changes or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A book packaging method based on the readability field, characterized in that, The method is applied to a server in a book storage system, which also includes book sorting robots deployed in each book warehouse. The server is communicatively connected to the book sorting robots. The method includes: Receive book order information, which includes at least a list of books to be packaged, and the list of books to be packaged describes the book categories and quantities corresponding to each school age group. Identify the appropriate reading field for each book category in the list of books to be packaged, wherein the appropriate reading field is used to describe the readability of the book category relative to its corresponding school age group; Based on the readability field and the list of books already in the warehouse corresponding to each book warehouse, a first book allocation strategy is generated with the goal of minimizing the number of book warehouses allocated and minimizing the number of book sorting robots scheduled in each book warehouse. The first book allocation strategy is used to describe at least one target book warehouse and its corresponding sub-list of books to be packed and the range of the number of schedulable robots. For each target book warehouse, the list of books to be packed corresponding to the target book warehouse is divided into multiple picking orders based on the readability field. Based on the range of the number of schedulable robots and the current status information of each book sorting robot in the target book warehouse, a second book allocation strategy is generated. The second book allocation strategy is used to describe the target book sorting robots scheduled in the target book warehouse and their corresponding picking orders. According to the second book allocation strategy, the target book sorting robot is controlled to perform book packing actions according to its corresponding picking list; Based on the readability field and the list of books already in storage corresponding to each book warehouse, a first book allocation strategy is generated with the goal of minimizing the number of book warehouses allocated and the number of book sorting robots scheduled in each book warehouse. This strategy includes: According to the readability constraint, based on the readability field corresponding to each book category in the list of books to be packaged and the list of books already in the warehouse corresponding to each book warehouse, an initial set of book allocation strategies is generated. The readability constraint is: book categories corresponding to the same age group and the same readability field are preferentially allocated to the same book warehouse. For any candidate book allocation strategy in the set of book allocation strategies, the fitness value corresponding to the candidate book allocation strategy is generated with the goal of minimizing the number of book warehouses, minimizing the number of book sorting robots scheduled in each book warehouse, and maximizing the concentration of the appropriate reading fields. Based on the fitness value, multiple candidate book allocation strategies are selected from the book allocation strategy set, and crossover and mutation operations are performed on the selected book allocation strategies to obtain new candidate book allocation strategies to form a new book allocation strategy set. When the iteration stopping condition is met, the candidate book allocation strategy corresponding to the largest fitness value in the book allocation strategy set at this time is taken as the first book allocation strategy.

2. The book packaging method based on the readability field according to claim 1, characterized in that, The robotic arms of the book sorting robots are equipped with grippers for whole packages of books and grippers for single books. Based on the range of schedulable robots and the current status information of each book sorting robot in the target book warehouse, a second book allocation strategy is generated, including: The book packing task from the picking list is published to the task pool; For any book packing task in the task pool, a competition request is received from each book sorting robot in the target book warehouse for that book packing task. Based on the current status information carried in the competition request, the task cost corresponding to each book sorting robot is determined. According to the task cost corresponding to each book sorting robot and the range of the number of schedulable robots, the target book sorting robot corresponding to the picking list of the book packing task is determined, and the book packing task is removed from the task pool. The current status information includes the distance between the current position of the book sorting robot and the position information corresponding to the book packing task, the gripper type currently used by the book sorting robot, and the battery information of the book sorting robot. Continue to determine the target book sorting robot corresponding to the picking list of the next book packing task until the task pool is empty, and obtain the second book allocation strategy.

3. The book packaging method based on the readability field according to claim 1, characterized in that, According to the second book allocation strategy, the target book sorting robot is controlled to perform book packing actions according to its corresponding picking list, including: Obtain the warehouse grid model corresponding to the target book warehouse, and the warehouse grid model describes the occupancy status of each grid in the target book warehouse through specified markers; Based on the warehouse grid model, the current position of the target book sorting robot and its corresponding picking list, the picking path corresponding to the target book sorting robot is planned; Based on the picking path, the time it takes for the target book sorting robot to arrive at each grid in the target book warehouse is predicted, so as to determine whether there are multiple target book sorting robots occupying the same grid at the same time; If so, pause control or obstacle avoidance control is applied to the target book sorting robots that simultaneously occupy the same grid, so that the target book sorting robots can successfully reach the grid corresponding to the picking list and perform book packing actions according to the corresponding picking list.

4. The book packaging method based on the readability field according to claim 3, characterized in that, The book sorting robot is equipped with vision devices; it performs book packing actions according to the corresponding picking list, including: After the target book sorting robot successfully reaches the grid corresponding to the picking list, the shelf image data collected by the vision device is acquired. The book to be picked is located based on the shelf image data and the location information described in the picking list, and the type of gripper currently used by the target book sorting robot is identified based on the shelf image data to determine whether it matches the book to be picked. If not, the current gripper type used by the target book sorting robot is switched, and the switched gripper is used to perform gripping and book packing actions on the book to be grabbed.

5. The book packaging method based on the readability field according to claim 1, characterized in that, The book storage system also includes book monitoring equipment communicatively connected to the server. The book monitoring equipment includes a pressure sensor network and a camera network deployed on each shelf in the book warehouse. The camera network includes a top-level fisheye lens, a mid-level wide-angle lens, and a bottom-level depth camera. The method further includes: The pressure sensor network senses the pressure changes corresponding to each shelf in the shelf, and monitors whether a book entry or exit event occurs on the shelf based on the pressure changes. If so, locate the target shelf and shelf where the book entry event or the book exit event occurs, activate the camera network corresponding to the target shelf and shelf, and obtain the set of book images collected by the activated camera network for the target shelf and shelf. The set of book images includes the occupancy heat map collected by the top fisheye lens, the book image data collected by the middle wide-angle lens, and the book point cloud data collected by the bottom depth camera. Update the list of books already in storage corresponding to the book warehouse based on the book image set.

6. The book packaging method based on the readability field according to claim 5, characterized in that, Updating the list of books already in storage corresponding to the book warehouse based on the book image set includes: In the event of the aforementioned book entry event, the following operations are performed: Based on the occupancy heatmap, identify the target shelves and shelves where the book entry event occurred; Extract the first sub-image data corresponding to the entry event area from the book image data, and identify the book ridge data and book ISBN data corresponding to the newly entered book based on the first sub-image data; Based on the book ridge data and the book point cloud data, determine the location information corresponding to the newly added book; The list of books already in the warehouse is updated based on the ISBN data and location information of the newly added books. In the event of the aforementioned book being shipped out, the following operations will be performed: Based on the occupancy heatmap, identify the target shelves and shelves where the book outbound event occurred; Extract the second sub-image region corresponding to the outbound event region from the book image data. Based on the second sub-image region and the second sub-image region of the previous frame, identify the book ISBN data and the location information of the outbound book. The list of books already in stock is updated based on the ISBN data of the books that have been shipped out and the location information.

7. A book packaging device based on a readability field, characterized in that, The device is applied to a server in a book storage system, which also includes book sorting robots deployed in each book warehouse. The server is communicatively connected to the book sorting robots. The device includes: The order receiving module is used to receive book order information, which includes at least a list of books to be packaged, and the list of books to be packaged describes the book categories and quantities corresponding to each school age group. The appropriateness identification module is used to identify the appropriateness field corresponding to each book category in the list of books to be packaged. The appropriateness field is used to describe the appropriateness of the book category relative to its corresponding school age group. The first allocation module is used to generate a first book allocation strategy based on the readability field and the list of books already in the warehouse corresponding to each book warehouse, with the goal of minimizing the number of book warehouses allocated and the number of book sorting robots scheduled in each book warehouse. The first book allocation strategy is used to describe at least one target book warehouse and its corresponding sub-list of books to be packed and the range of the number of schedulable robots. The second allocation module is used to divide the list of books to be packed corresponding to each target book warehouse into multiple picking orders based on the readability field, and generate a second book allocation strategy based on the range of the number of schedulable robots and the current status information of each book sorting robot in the target book warehouse. The second book allocation strategy is used to describe the target book sorting robots scheduled in the target book warehouse and their corresponding picking orders. The book packing module is used to control the target book sorting robot to perform book packing actions according to the corresponding picking list, in accordance with the second book allocation strategy. The first allocation module is specifically used for: According to the readability constraint, based on the readability field corresponding to each book category in the list of books to be packaged and the list of books already in the warehouse corresponding to each book warehouse, an initial set of book allocation strategies is generated. The readability constraint is: book categories corresponding to the same age group and the same readability field are preferentially allocated to the same book warehouse. For any candidate book allocation strategy in the set of book allocation strategies, the fitness value corresponding to the candidate book allocation strategy is generated with the goal of minimizing the number of book warehouses, minimizing the number of book sorting robots scheduled in each book warehouse, and maximizing the concentration of the appropriate reading fields. Based on the fitness value, multiple candidate book allocation strategies are selected from the book allocation strategy set, and crossover and mutation operations are performed on the selected book allocation strategies to obtain new candidate book allocation strategies to form a new book allocation strategy set. When the iteration stopping condition is met, the candidate book allocation strategy corresponding to the largest fitness value in the book allocation strategy set at this time is taken as the first book allocation strategy.

8. A server, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 6.

Images