A forklift task scheduling management method and system

Abstract

The present application provides a method and system for scheduling and managing forklift tasks, and relates to the technical field of task scheduling. The method includes: registering and fusing warehouse image information and warehouse point cloud data information, generating a warehouse visualization model, and extracting storage features from task demand information; performing task matching identification on the warehouse visualization model, and standardizing multiple forklift work data streams; scheduling and analyzing task storage allocation information, multiple standard forklift work data streams, and multiple forklift performance parameters based on task scheduling rules, and generating multiple task scheduling plans; performing differentiated evaluation and optimization, and performing task scheduling management. The present application can solve the technical problem of low efficiency of task scheduling in the prior art due to poor real-time response to changes in the storage environment and changes in task demand, achieve the technical goal of improving warehouse management efficiency, and achieve the technical effect of reducing logistics costs.

Description

Technical Field

[0001] The present application relates to the technical field of task scheduling, and in particular to a method and system for managing forklift task scheduling. Background Art

[0002] Forklift is a widely used equipment in warehousing, logistics and production. It is used to transport, stack and store goods. It has the functions of lifting, moving forward and backward, and turning left and right. It can carry out efficient and flexible cargo handling operations in warehouses, cargo yards, docks and other places. The efficiency of forklift task scheduling directly affects the operating efficiency of the entire warehousing system.

[0003] At present, in the existing technology, the forklift task scheduling method mostly relies on manual experience, and the real-time response to changes in the warehouse environment and task requirements is poor, resulting in low task scheduling efficiency and high logistics costs.

[0004] In summary, the existing technology has technical problems such as low efficiency in task scheduling and high logistics costs due to poor real-time response to changes in the warehousing environment and changes in task requirements. Summary of the invention

[0005] The purpose of this application is to provide a forklift task scheduling management method and system to solve the technical problems in the prior art that the real-time response to changes in the storage environment and changes in task requirements is poor, resulting in low task scheduling efficiency and high logistics costs.

[0006] In view of the above problems, the present application provides a forklift task scheduling management method and system.

[0007] In the first aspect, the present application provides a method for scheduling and managing forklift tasks, which is implemented through a forklift task scheduling and management system, wherein the method comprises: S1: acquiring warehouse image information and warehouse point cloud data information through visual sensing equipment and laser scanners, aligning and fusing the warehouse image information and warehouse point cloud data information, generating a warehouse visualization model, and generating task requirement information through a warehouse management system at the same time, performing warehouse feature extraction on the task requirement information, and obtaining task warehouse multidimensional feature information, wherein the task warehouse multidimensional feature information comprises task type, task priority, location and quantity; S2: performing task matching identification on the warehouse visualization model based on the task warehouse multidimensional feature information, obtaining task warehouse allocation information, and installing a signal on the forklift. The sensing module integrates a GPS locator, a speed sensor, a load sensor, and an energy consumption monitor. The signal sensing module collects and obtains multiple working data streams of multiple forklifts, and standardizes the multiple forklift working data streams to obtain multiple standard forklift working data streams; S3: obtaining multiple forklift performance parameters of multiple forklifts, defining task scheduling rules, and performing scheduling analysis on the task storage allocation information, the multiple standard forklift working data streams, and the multiple forklift performance parameters based on the task scheduling rules to generate multiple task scheduling schemes; S4: performing differentiated evaluation and optimization based on the multiple task scheduling schemes, determining the target task scheduling scheme, and synchronizing the target task scheduling scheme to the forklift scheduling platform for task scheduling management.

[0008] In the second aspect, the present application also provides a forklift task scheduling and management system for executing a forklift task scheduling and management method as described in the first aspect, wherein the system comprises: a task warehouse multidimensional feature information acquisition module, the task warehouse multidimensional feature information acquisition module is used to collect and acquire warehouse image information and warehouse point cloud data information through visual sensing equipment and laser scanners, align and fuse the warehouse image information and warehouse point cloud data information, generate a warehouse visualization model, and at the same time generate task requirement information through a warehouse management system, perform warehouse feature extraction on the task requirement information, and obtain task warehouse multidimensional feature information, the task warehouse multidimensional feature information includes task type, task priority, location and quantity; a standard forklift work data stream acquisition module, the standard forklift work data stream acquisition module is used to perform task matching identification on the warehouse visualization model based on the task warehouse multidimensional feature information, and obtain task warehouse allocation information, in A signal sensing module is installed on the forklift, and the signal sensing module integrates a GPS locator, a speed sensor, a load sensor, and an energy consumption monitor. The signal sensing module collects and obtains multiple working data streams of multiple forklifts, and the multiple forklift working data streams are standardized to obtain multiple standard forklift working data streams; a task scheduling scheme generation module, the task scheduling scheme generation module is used to obtain multiple forklift performance parameters of multiple forklifts, define task scheduling rules, and perform scheduling analysis on the task storage allocation information, the multiple standard forklift working data streams and the multiple forklift performance parameters based on the task scheduling rules to generate multiple task scheduling schemes; a target task scheduling scheme determination module, the target task scheduling scheme determination module is used to perform differentiated evaluation and optimization based on the multiple task scheduling schemes, determine the target task scheduling scheme, and synchronize the target task scheduling scheme to the forklift scheduling platform for task scheduling management.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] 1. Collect warehouse image information and warehouse point cloud data information through visual sensing equipment and laser scanners, align and fuse the warehouse image information and warehouse point cloud data information to generate a warehouse visualization model, and generate task demand information through the warehouse management system, extract warehouse features from the task demand information, and obtain task warehouse multidimensional feature information, which includes task type, task priority, location and quantity; perform task matching identification on the warehouse visualization model based on the task warehouse multidimensional feature information to obtain task warehouse allocation information, and install a signal sensing module on the forklift, which integrates a GPS locator, a speed sensor, a load sensor, and an energy consumption monitor. The module collects and obtains multiple working data streams of multiple stackers, standardizes the multiple stacker working data streams, and obtains multiple standard stacker working data streams; obtains multiple stacker performance parameters of multiple stackers, defines task scheduling rules, and performs scheduling analysis on the task storage allocation information, the multiple standard stacker working data streams, and the multiple stacker performance parameters based on the task scheduling rules to generate multiple task scheduling schemes; performs differentiated evaluation and optimization based on the multiple task scheduling schemes, determines the target task scheduling scheme, and synchronizes the target task scheduling scheme to the stacker scheduling platform for task scheduling management, that is, collects the stacker's working status information, task requirement information, and storage environment information through sensors and information systems. Then, the collected data is modeled and trained using a deep learning algorithm to generate an intelligently optimized task scheduling model. Then, according to the task scheduling model, the stacker is tasked and scheduled. Finally, through the real-time feedback mechanism, the model is adjusted and optimized in real time according to the actual operation of the stacker and the completion of the task, and finally the technical goal of improving the efficiency of warehouse management is achieved, and the technical effect of reducing logistics costs is achieved.

[0011] 2. By improving the operational management efficiency of the warehousing system, the technical goal of improving the quality of logistics services is achieved.

[0012] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0014] Figure 1 A flowchart of a method for dispatching and managing forklift tasks is provided for this application;

[0015] Figure 2 This is a structural schematic diagram of a forklift task scheduling management system for this application.

[0016] Description of reference numerals:

[0017] Task storage multi-dimensional feature information acquisition module 11, standard stacker work data flow acquisition module 12, task scheduling plan generation module 13, target task scheduling plan determination module 14. DETAILED DESCRIPTION

[0018] This application provides a method and system for scheduling and managing forklift tasks, which solves the technical problem in the prior art that the task scheduling efficiency is low and the logistics cost is high due to the poor real-time response to changes in the storage environment and task requirements. The technical goal of improving the efficiency of storage management is achieved, and the technical effect of reducing logistics costs is achieved.

[0019] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0020] Embodiment 1

[0021] Please see attached Figure 1 The present application provides a method for scheduling and managing a forklift task, wherein the method is applied to a forklift task scheduling and management system, and the method specifically comprises the following steps:

[0022] S1: Warehouse image information and warehouse point cloud data information are acquired through visual sensing equipment and laser scanners, and the warehouse image information and warehouse point cloud data information are aligning and fused to generate a warehouse visualization model. At the same time, task requirement information is generated through a warehouse management system, and warehouse features are extracted from the task requirement information to obtain task warehouse multidimensional feature information, which includes task type, task priority, location, and quantity.

[0023] Specifically, the warehouse image information is acquired by collecting images of the warehouse through visual sensing equipment. The warehouse point cloud data information is acquired by collecting point cloud data of the warehouse through a laser scanner. Furthermore, the warehouse image information and the warehouse point cloud data information are feature registered and data fusion rendered to generate a warehouse visualization model. At the same time, the task requirement information is generated through the warehouse management system. Among them, the warehouse management system is a management system for warehouse scheduling tasks. The task requirement information is the warehouse scheduling task to be executed. Furthermore, the task requirement information is subjected to warehouse feature extraction, that is, the requirements of the warehouse task are extracted to obtain the task warehouse multi-dimensional feature information. Among them, the task warehouse multi-dimensional feature information includes task type, task priority, location and quantity. Furthermore, the task type is the type of warehousing, storage and outbound in the warehouse. The task priority is the level of priority execution required for the task. The location is the storage location. The quantity is the storage quantity.

[0024] Step 2: S2: Based on the multi-dimensional feature information of the task warehouse, the warehouse visualization model is task matched and identified to obtain the task warehouse allocation information, and a signal sensing module is installed on the stacker. The signal sensing module integrates a GPS locator, a speed sensor, a load sensor, and an energy consumption monitor. The signal sensing module is used to collect and obtain multiple working data streams of multiple stackers, and the multiple stacker working data streams are standardized to obtain multiple standard stacker working data streams;

[0025] Specifically, the multidimensional characteristic information of task storage refers to the characteristic information of the storage task type, task priority, location and quantity. The storage matching strategy refers to parameters such as storage environment adaptability and in-and-out warehouse priority. Based on the multidimensional characteristic information of task storage and the storage matching strategy, the matching cargo location is task-identified in the warehouse visualization model to obtain the task storage allocation information. Further, a signal sensing module is installed on the stacker, wherein the signal sensing module is an integrated module of a GPS locator, a speed sensor, a load sensor and an energy consumption monitor. Further, multiple working data streams of multiple stackers are collected and obtained through the signal sensing module, wherein the multiple working data streams include the positioning data stream collected by the GPS locator in a continuous time, the speed data stream collected by the speed sensor, the load data stream collected by the load sensor, and the energy consumption data stream collected by the energy consumption monitor. The working data streams of multiple stackers are standardized, that is, the positioning data stream, the speed data stream, the load data stream and the energy consumption data stream are de-unitized and aligned, thereby obtaining multiple standard stacker working data streams.

[0026] S3: Acquire multiple stacker performance parameters of multiple stackers, define task scheduling rules, perform scheduling analysis on the task storage allocation information, the multiple standard stacker work data streams and the multiple stacker performance parameters based on the task scheduling rules, and generate multiple task scheduling schemes;

[0027] Specifically, multiple stacker performance parameters of multiple stackers are collected and obtained. For example, stacker performance parameters include lifting height, driving speed, load, energy consumption and other parameters. Task scheduling rules are defined according to stacker performance parameters, and task scheduling rules are mapped according to task storage allocation information, multiple standard stacker work data streams and multiple stacker performance parameters to obtain task rule scheduling parameter thresholds. Multiple task scheduling schemes are obtained according to task rule scheduling parameter thresholds.

[0028] S4: Perform differentiated evaluation and optimization based on the multiple task scheduling schemes, determine a target task scheduling scheme, and synchronize the target task scheduling scheme to the forklift scheduling platform for task scheduling management.

[0029] Specifically, the feedback parameters of the scheduling control of multiple task scheduling schemes are obtained, and the optimal scheduling effect is extracted from the feedback parameters, and then the task scheduling scheme with the best scheduling effect is determined as the target task scheduling scheme. Furthermore, the target task scheduling scheme is synchronized to the forklift scheduling platform for task scheduling management, which is used to execute the forklift scheduling.

[0030] The method for task scheduling and management of a forklift is applied to a task scheduling and management system of a forklift, and can collect the working status information, task requirement information and storage environment information of the forklift through sensors and information systems. Then, the collected data is modeled and trained using a deep learning algorithm to generate an intelligently optimized task scheduling model. Then, according to the task scheduling model, the forklift is assigned and scheduled. Finally, through a real-time feedback mechanism, the model is adjusted and optimized in real time according to the actual operation of the forklift and the completion of the task, and finally the technical goal of improving the efficiency of warehouse management is achieved, and the technical effect of reducing logistics costs is achieved.

[0031] Furthermore, the present application also includes the following steps:

[0032] S11: obtaining a warehouse feature label library through warehouse attribute information, wherein the warehouse feature label library is a type label library of each warehouse entity, and constructing a warehouse semantic segmentation network based on the warehouse feature label library;

[0033] S12: performing feature segmentation on the warehouse image information using the warehouse semantic segmentation network to obtain a warehouse image semantic segmentation entity set;

[0034] S13: performing data dimensionality reduction processing on the warehouse image information and the warehouse point cloud data information based on the warehouse image semantic segmentation entity set to obtain warehouse dimensionality reduction image information and warehouse dimensionality reduction point cloud data information;

[0035] S14: performing feature registration and data fusion rendering on the warehouse dimension reduction image information and the warehouse dimension reduction point cloud data information to generate the warehouse visualization model.

[0036] Specifically, the warehouse attribute information is the attribute type of the warehouse constituent materials, such as shelves, channels, walls, etc. A warehouse feature label library is constructed according to different attribute types of warehouse constituent materials, such as a warehouse feature label library including physical labels such as shelves, channels, walls, etc. Among them, the warehouse feature label library includes warehouse entities corresponding to each attribute type. Further, a warehouse semantic segmentation network is constructed, and multiple labels in the warehouse feature label library are used as warehouse feature input data and input into the warehouse semantic segmentation network. The warehouse feature input data is divided to obtain warehouse feature training data and warehouse feature verification data, wherein the division ratio of the warehouse feature training data and the warehouse feature verification data is obtained by a technician in this field according to the actual situation. For example, the division ratio of the warehouse feature training data and the warehouse feature verification data is 7:3. Further, the warehouse feature training data is input into the warehouse semantic segmentation network. When the output data of the warehouse semantic segmentation network tends to be stable, the warehouse semantic segmentation network is verified by the warehouse feature verification data. When the output accuracy of the warehouse semantic segmentation network meets the output accuracy threshold of the warehouse semantic segmentation network, the warehouse semantic segmentation network is trained. Among them, the output accuracy threshold of the warehouse semantic segmentation network is customized by technical personnel in this field according to actual conditions. For example, the output accuracy threshold of the warehouse semantic segmentation network is 90%.

[0037] Then, the warehouse image information collected is segmented using the warehouse semantic segmentation network to obtain a warehouse image semantic segmentation entity set. For example, the warehouse image semantic segmentation entity set includes shelves, aisles, and walls.

[0038] Next, each warehouse entity in the warehouse image semantic segmentation entity set is extracted, and the warehouse image information and warehouse point cloud data information corresponding to each warehouse entity are subjected to data dimensionality reduction processing to obtain warehouse dimensionality reduction image information and warehouse dimensionality reduction point cloud data information. After dimensionality reduction processing, the warehouse image information and warehouse point cloud data information have lower dimensions, but still retain the main structure and features of the original data, which can be used for subsequent analysis and visualization.

[0039] Next, extract key features from the warehouse reduced dimension image information after dimensionality reduction, such as edges, corners, textures, etc. Extract structural features from the warehouse reduced dimension point cloud data information after dimensionality reduction, such as planes, edges, corners, etc. Use a feature-based registration algorithm, for example, the registration algorithm is an iterative closest point algorithm, etc., to align the features in the warehouse reduced dimension image information and the features in the warehouse reduced dimension point cloud data information, and then map the features in the warehouse reduced dimension image information to the features in the warehouse reduced dimension point cloud data information to generate a warehouse visualization model.

[0040] The warehouse visualization model is generated by registering and fusing the warehouse image information and the warehouse point cloud data information, thereby improving the accuracy of generating the warehouse visualization model.

[0041] Furthermore, the present application also includes the following steps:

[0042] Performing a criticality evaluation on each warehouse entity type label in the warehouse feature label library to obtain criticality information of the warehouse entity type label;

[0043] Obtaining the precision of the target model, performing hierarchical analysis on the criticality information of the warehouse entity type label based on the precision of the target model, and determining the data density dimensionality reduction hierarchical information;

[0044] Mapping and matching the data density dimensionality reduction level information with the warehouse image semantic segmentation entity set to obtain a warehouse entity level dimensionality reduction rule;

[0045] Based on the warehouse entity hierarchical dimensionality reduction rule, the warehouse image information and the warehouse point cloud data information are respectively subjected to hierarchical dimensionality reduction processing to obtain the warehouse reduced dimensionality image information and the warehouse reduced dimensionality point cloud data information.

[0046] Specifically, the criticality of each warehouse entity type label in the warehouse feature label library is evaluated, wherein the higher the application degree of each warehouse entity, the higher the criticality obtained by the evaluation, and vice versa, the lower the criticality, and then the criticality information of the warehouse entity type label is obtained. For example, the higher the criticality of the shelf and channel, the higher the criticality information of the shelf and channel label.

[0047] Then, the precision of the target model is the precision of the hierarchical division of the warehouse entity type labels according to the warehouse entity type label criticality information. For example, the shelves and channels are hierarchically divided according to the precision of the target model. When the criticality of the shelf relative to the channel is higher, the level of the shelf relative to the channel is higher. Among them, the higher the precision of the target model, the higher the level of the division, and vice versa. Further, when the level is higher, the degree of data dimensionality reduction is lower, and vice versa. Further, the criticality information of the warehouse entity type labels is hierarchically analyzed through the obtained target model precision to determine the data density dimensionality reduction level information for data dimensionality reduction.

[0048] Next, the data density dimensionality reduction level information is mapped and matched with the warehouse image semantic segmentation entity set, and the mapping relationship between each warehouse entity in the warehouse image semantic segmentation entity set and the data density dimensionality reduction level information is matched, and then the warehouse entity level dimensionality reduction rule is obtained by combination. Among them, the warehouse entity level dimensionality reduction rule is to reduce the dimension of the image data of each mapped warehouse entity according to the data density dimensionality reduction level information.

[0049] Next, according to the warehouse entity hierarchical dimensionality reduction rules, the warehouse image information and warehouse point cloud data information mapped to each warehouse entity are respectively subjected to dimensionality reduction processing at the corresponding level to obtain warehouse dimensionality reduction image information and warehouse dimensionality reduction point cloud data information.

[0050] By performing dimensionality reduction processing on the warehouse image information and warehouse point cloud data information, the warehouse image information and warehouse point cloud data information after dimensionality reduction processing have lower dimensions, but still retain the main structure and features of the original data, which can be used for subsequent analysis.

[0051] Furthermore, the present application also includes the following steps:

[0052] S21: assigning a decision order to each feature type of the task warehouse multidimensional feature information, and arranging and determining the task warehouse multidimensional feature decision information;

[0053] S22: Acquire cargo location distribution characteristic information through the WMS warehouse system, mark the cargo location distribution characteristic information with remaining space, and acquire remaining space storage characteristic information;

[0054] S23: Obtaining a warehouse matching strategy, wherein the warehouse matching strategy includes warehouse applicability and warehouse in / out characteristics, and performing a warehouse matching analysis on the task warehouse multidimensional feature decision information and the remaining warehouse information based on the warehouse matching strategy to obtain remaining warehouse matching feature parameter information;

[0055] S24: Identify the storage task based on the remaining storage matching feature parameter information and the warehouse visualization model to obtain the task storage allocation information.

[0056] Specifically, the task type, task priority, location, and quantity in each feature type of the task storage multidimensional feature information are arranged in order of decision priority, and the task storage multidimensional features after the order is arranged are obtained as the task storage multidimensional feature decision information. Among them, the decision order allocation, that is, the method of decision priority arrangement is obtained by a person skilled in the art through customized settings according to actual conditions.

[0057] Then, the WMS warehouse system is a warehouse management system used to efficiently manage information, resources, behaviors, inventory and distribution operations in the warehouse. The WMS warehouse system obtains information such as cargo space layout, shelf type, shelf arrangement, load-bearing capacity, storage information, etc. as cargo space distribution characteristic information. Furthermore, the remaining space is the vacant cargo space in the cargo space distribution characteristic information. The cargo space distribution characteristic information is marked with remaining space to obtain the remaining space storage characteristic information.

[0058] Next, obtain the storage applicability and in-and-out characteristics. Among them, storage applicability refers to the type of goods, size and weight, and storage environment adaptability. The in-and-out characteristics refer to the in-and-out frequency, in-and-out priority, etc. Further, obtain the storage matching strategy based on the storage applicability and in-and-out characteristics. The storage matching strategy refers to the strategy for the warehousing to be performed. Match the storage matching strategy of the remaining storage information with the multi-dimensional feature decision information of the task storage to obtain the remaining storage matching feature parameter information. For example, the higher the storage environment adaptability and the higher the in-and-out priority in the storage matching strategy of the remaining storage information, the higher the task priority in the multi-dimensional feature decision information of the task storage.

[0059] Next, the remaining storage matching feature parameter information is used to identify storage tasks in the warehouse visualization model, that is, the task type, task priority, location and quantity of the remaining storage matching are identified to obtain the task storage allocation information.

[0060] By matching and identifying tasks in the warehouse visualization model, task warehouse allocation information can be obtained, thereby further improving the accuracy of the task warehouse allocation information.

[0061] Furthermore, the present application also includes the following steps:

[0062] S31: feature mapping the task storage allocation information, the plurality of standard stacker work data streams and the plurality of stacker performance parameters with the task scheduling rules to obtain task characteristic rule parameters and stacker state rule parameters;

[0063] S32: performing task scheduling association on the task characteristic rule parameter and the forklift state rule parameter to obtain a task rule scheduling parameter threshold;

[0064] S33: randomly generating a plurality of scheduling parameter schemes based on the task rule scheduling parameter threshold, and dividing and clustering the plurality of scheduling parameter schemes to generate the plurality of task scheduling schemes.

[0065] Specifically, the task scheduling rules include task characteristic rules and forklift status rules. The task storage allocation information is feature mapped with the task scheduling rules to obtain the task characteristic rule parameters. Multiple standard forklift work data streams and multiple forklift performance parameters are feature mapped with the task scheduling rules to obtain the forklift status rule parameters.

[0066] Then, the forklift state rule parameters are the current scheduling rule parameters of the forklift. The task characteristic rule parameters and the forklift state rule parameters are associated with task scheduling, and the scheduling scheme parameter adjustment range obtained based on the association analysis of the current scheduling rule parameters of the forklift, including the task list, execution order, path planning, etc. of each forklift, is used as the task rule scheduling parameter threshold.

[0067] Next, multiple scheduling parameter schemes are randomly generated based on the scheduling parameter threshold of the task rule, including scheduling parameter schemes with different execution orders and path planning. Multiple scheduling parameter schemes are divided and clustered, and scheduling parameter schemes with relatively close scheduling parameters are clustered to generate multiple task scheduling schemes. For example, a scheduling parameter threshold is set, and task scheduling schemes whose scheduling parameters meet the scheduling parameter threshold are clustered.

[0068] The task scheduling is associated with the forklift performance parameters and the forklift state rule parameters, and then a task scheduling plan is generated, thereby further improving the accuracy of obtaining the task scheduling plan.

[0069] Furthermore, the present application also includes the following steps:

[0070] Obtaining scheduling effect evaluation indicators, wherein the scheduling effect evaluation indicators include reliability, task time, and energy consumption cost, and constructing a scheduling effect objective function according to the scheduling effect evaluation indicators;

[0071] The plurality of scheduling parameter schemes are evaluated and calculated by using the scheduling effect objective function to obtain a plurality of scheduling effect fitnesses;

[0072] Based on the multiple scheduling effect fitnesses, the multiple scheduling parameter schemes are proportionally divided and clustered to obtain multiple scheduling parameter scheme clusters;

[0073] The multiple scheduling parameter scheme clusters are iteratively optimized, and the fitness sum of the optimized multiple scheduling parameter scheme clusters is calculated to screen and obtain the multiple task scheduling schemes.

[0074] Specifically, indicators such as reliability, task time, and energy consumption cost are obtained, and the scheduling effect evaluation indicators are combined to obtain the scheduling effect objective function. The scheduling effect evaluation indicator is fitted to construct the scheduling effect objective function, that is, the scheduling effect evaluation indicator is integrated to construct the scheduling effect objective function.

[0075] Then, multiple scheduling parameter schemes are evaluated and calculated through the scheduling effect objective function, that is, the indicator parameters corresponding to multiple indicators are extracted to calculate the fitness of multiple indicator parameters, and multiple scheduling effect fitness are obtained. Among them, the better the effect of multiple scheduling parameter schemes, the stronger the scheduling effect fitness is, and vice versa.

[0076] Next, the multiple scheduling effect fitness degrees are divided according to different proportions, wherein the higher the scheduling effect fitness degree, the higher the allocation ratio of the multiple scheduling effect fitness degrees, and vice versa. For example, the higher the scheduling effect fitness degree, the higher the concentration of the multiple scheduling effect fitness degrees. Further, according to the allocation ratio and the corresponding scheduling effect fitness degree, the corresponding scheduling parameter scheme is divided and clustered to obtain the scheduling parameter scheme cluster corresponding to the multiple concentrated scheduling effect fitness degrees.

[0077] Next, multiple scheduling parameter scheme clusters are iteratively optimized, that is, the scheduling parameter schemes in each scheduling parameter scheme cluster are iteratively optimized to increase the diversity of scheduling parameter schemes. The scheduling effect fitness is calculated for each of the optimized scheduling parameter scheme clusters, and the total sum of the multiple scheduling effect fitnesses in the scheduling parameter scheme cluster is calculated to screen out multiple task scheduling schemes in the scheduling parameter scheme cluster with the highest scheduling effect fitness.

[0078] By obtaining the scheduling effect evaluation index, multiple scheduling effect fitnesses are obtained, and then multiple task scheduling schemes are screened to further improve the efficiency of obtaining multiple task scheduling schemes.

[0079] Furthermore, the present application also includes the following steps:

[0080] S41: Selecting the optimal task scheduling scheme according to the multiple task scheduling schemes to perform forklift scheduling control and obtain forklift scheduling feedback parameters;

[0081] S42: If the forklift scheduling feedback parameter does not reach the preset scheduling effect, simulate the remaining task scheduling schemes in the multiple task scheduling schemes to obtain the remaining multiple task scheduling simulation effects;

[0082] S43: traverse and select the remaining multiple task scheduling simulation effects, and reversely match to determine the target task scheduling solution.

[0083] Specifically, the optimal task scheduling scheme is selected from multiple task scheduling schemes. For example, multiple forklift scheduling feedback parameters are obtained by simulating and executing multiple task scheduling schemes, and the corresponding fitness is sorted. The better the forklift scheduling feedback parameter, the higher the corresponding fitness, and vice versa. The task scheduling scheme with the highest fitness is sorted according to the fitness, and the forklift scheduling control is actually performed to obtain the actual forklift scheduling feedback parameters.

[0084] Then, a preset scheduling effect is obtained, wherein the preset scheduling effect is an ideal scheduling effect of the forklift scheduling feedback parameters obtained by the forklift scheduling control, that is, the fitness of the forklift scheduling feedback parameters in the ideal scheduling effect satisfies the fitness threshold and thus achieves the preset scheduling effect, and the fitness threshold is obtained by a technician in this field according to the actual situation. Furthermore, if the task scheduling simulation effect of the forklift scheduling feedback parameters does not reach the preset scheduling effect, it means that the forklift scheduling feedback parameters need to be optimized. Simulate the remaining task scheduling schemes in the multiple task scheduling schemes to obtain the remaining multiple task scheduling simulation effects, which are used to determine whether the preset scheduling effect is achieved.

[0085] Next, multiple task scheduling simulation effects are sorted in descending order to obtain the optimal results of the remaining multiple task scheduling simulation effects. The optimal results of the remaining multiple task scheduling simulation effects are accessed in sequence to extract the task scheduling scheme that achieves the preset scheduling effect, and then reverse matching is performed to determine the target task scheduling scheme.

[0086] By judging the effect of task scheduling simulation on the forklift scheduling feedback parameters, the target task scheduling plan is determined, and the accuracy of obtaining the target task scheduling plan is further improved.

[0087] In summary, the forklift task scheduling management method provided by the present application has the following technical effects:

[0088] 1. Collect warehouse image information and warehouse point cloud data information through visual sensing equipment and laser scanners, align and fuse the warehouse image information and warehouse point cloud data information to generate a warehouse visualization model, and generate task demand information through the warehouse management system, extract warehouse features from the task demand information, and obtain task warehouse multidimensional feature information, which includes task type, task priority, location and quantity; perform task matching identification on the warehouse visualization model based on the task warehouse multidimensional feature information to obtain task warehouse allocation information, and install a signal sensing module on the forklift, which integrates a GPS locator, a speed sensor, a load sensor, and an energy consumption monitor. The module collects and obtains multiple working data streams of multiple stackers, standardizes the multiple stacker working data streams, and obtains multiple standard stacker working data streams; obtains multiple stacker performance parameters of multiple stackers, defines task scheduling rules, and performs scheduling analysis on the task storage allocation information, the multiple standard stacker working data streams, and the multiple stacker performance parameters based on the task scheduling rules to generate multiple task scheduling schemes; performs differentiated evaluation and optimization based on the multiple task scheduling schemes, determines the target task scheduling scheme, and synchronizes the target task scheduling scheme to the stacker scheduling platform for task scheduling management, that is, collects the stacker's working status information, task requirement information, and storage environment information through sensors and information systems. Then, the collected data is modeled and trained using a deep learning algorithm to generate an intelligently optimized task scheduling model. Then, according to the task scheduling model, the stacker is tasked and scheduled. Finally, through the real-time feedback mechanism, the model is adjusted and optimized in real time according to the actual operation of the stacker and the completion of the task, and finally the technical goal of improving the efficiency of warehouse management is achieved, and the technical effect of reducing logistics costs is achieved.

[0089] 2. By improving the operational management efficiency of the warehousing system, the technical goal of improving the quality of logistics services is achieved.

[0090] Embodiment 2

[0091] Based on the forklift task scheduling management method in the aforementioned embodiment, the present application also provides a forklift task scheduling management system, see the attached Figure 2 , the system comprising:

[0092] The task warehouse multi-dimensional feature information acquisition module 11 is used to collect warehouse image information and warehouse point cloud data information through visual sensing equipment and laser scanners, align and fuse the warehouse image information and warehouse point cloud data information to generate a warehouse visualization model, and at the same time generate task demand information through a warehouse management system, extract warehouse features from the task demand information, and obtain task warehouse multi-dimensional feature information, wherein the task warehouse multi-dimensional feature information includes task type, task priority, location, and quantity;

[0093] A standard stacker work data stream acquisition module 12, the standard stacker work data stream acquisition module 12 is used to perform task matching identification on the warehouse visualization model based on the multi-dimensional feature information of the task warehouse, obtain task warehouse allocation information, install a signal sensing module on the stacker, the signal sensing module integrates a GPS locator, a speed sensor, a load sensor, and an energy consumption monitor, collect and obtain multiple work data streams of multiple stackers through the signal sensing module, standardize the multiple stacker work data streams, and obtain multiple standard stacker work data streams;

[0094] A task scheduling scheme generating module 13, the task scheduling scheme generating module 13 is used to obtain multiple stacker performance parameters of multiple stackers, define task scheduling rules, perform scheduling analysis on the task storage allocation information, the multiple standard stacker work data streams and the multiple stacker performance parameters based on the task scheduling rules, and generate multiple task scheduling schemes;

[0095] The target task scheduling scheme determining module 14 is used to perform differentiated evaluation and optimization based on the multiple task scheduling schemes, determine the target task scheduling scheme, and synchronize the target task scheduling scheme to the forklift scheduling platform for task scheduling management.

[0096] Furthermore, the task storage multi-dimensional feature information acquisition module 11 in the system is also used for:

[0097] Obtain a warehouse feature label library through warehouse attribute information, wherein the warehouse feature label library is a type label library of each warehouse entity, and train and construct a warehouse semantic segmentation network based on the warehouse feature label library;

[0098] Using the warehouse semantic segmentation network to perform feature segmentation on the warehouse image information to obtain a warehouse image semantic segmentation entity set;

[0099] Performing data dimensionality reduction processing on the warehouse image information and the warehouse point cloud data information based on the warehouse image semantic segmentation entity set to obtain warehouse dimensionality reduction image information and warehouse dimensionality reduction point cloud data information;

[0100] The warehouse dimension reduction image information and the warehouse dimension reduction point cloud data information are subjected to feature registration and data fusion rendering to generate the warehouse visualization model.

[0101] Furthermore, the task storage multi-dimensional feature information acquisition module 11 in the system is also used for:

[0102] Performing a criticality evaluation on each warehouse entity type label in the warehouse feature label library to obtain criticality information of the warehouse entity type label;

[0103] Obtaining the precision of the target model, performing hierarchical analysis on the criticality information of the warehouse entity type label based on the precision of the target model, and determining the data density dimensionality reduction hierarchical information;

[0104] Mapping and matching the data density dimensionality reduction level information with the warehouse image semantic segmentation entity set to obtain a warehouse entity level dimensionality reduction rule;

[0105] Based on the warehouse entity hierarchical dimensionality reduction rule, the warehouse image information and the warehouse point cloud data information are respectively subjected to hierarchical dimensionality reduction processing to obtain the warehouse reduced dimensionality image information and the warehouse reduced dimensionality point cloud data information.

[0106] Furthermore, the standard stacker operation data stream acquisition module 12 in the system is also used for:

[0107] Allocating a decision order for each feature type of the task storage multidimensional feature information, and arranging and determining the task storage multidimensional feature decision information;

[0108] Acquire cargo location distribution characteristic information through the WMS warehouse system, mark the cargo location distribution characteristic information with remaining space, and acquire remaining space storage characteristic information;

[0109] Obtaining a warehouse matching strategy, the warehouse matching strategy includes warehouse applicability and warehouse-in and -out characteristics, and performing a warehouse matching analysis on the task warehouse multidimensional feature decision information and the remaining warehouse information based on the warehouse matching strategy to obtain remaining warehouse matching feature parameter information;

[0110] The storage task is identified based on the remaining storage matching feature parameter information and the warehouse visualization model to obtain the task storage allocation information.

[0111] Furthermore, the task scheduling scheme generating module 13 in the system is also used for:

[0112] Feature mapping is performed on the task storage allocation information, the plurality of standard stacker work data streams and the plurality of stacker performance parameters with the task scheduling rules to obtain task characteristic rule parameters and stacker state rule parameters;

[0113] Associating the task characteristic rule parameters with the forklift state rule parameters with task scheduling to obtain a task rule scheduling parameter threshold;

[0114] A plurality of scheduling parameter schemes are randomly generated based on the task rule scheduling parameter threshold, and the plurality of scheduling parameter schemes are divided and clustered to generate the plurality of task scheduling schemes.

[0115] Furthermore, the task scheduling scheme generating module 13 in the system is also used for:

[0116] Obtaining scheduling effect evaluation indicators, wherein the scheduling effect evaluation indicators include reliability, task time, and energy consumption cost, and constructing a scheduling effect objective function according to the scheduling effect evaluation indicators;

[0117] The plurality of scheduling parameter schemes are evaluated and calculated by using the scheduling effect objective function to obtain a plurality of scheduling effect fitnesses;

[0118] Based on the multiple scheduling effect fitnesses, the multiple scheduling parameter schemes are proportionally divided and clustered to obtain multiple scheduling parameter scheme clusters;

[0119] The multiple scheduling parameter scheme clusters are iteratively optimized, and the fitness sum of the optimized multiple scheduling parameter scheme clusters is calculated to screen and obtain the multiple task scheduling schemes.

[0120] Furthermore, the target task scheduling scheme determination module 14 in the system is also used for:

[0121] Selecting the optimal task scheduling scheme according to the multiple task scheduling schemes to perform forklift scheduling control and obtain forklift scheduling feedback parameters;

[0122] If the forklift scheduling feedback parameter does not reach the preset scheduling effect, simulating the remaining task scheduling schemes in the multiple task scheduling schemes to obtain the remaining multiple task scheduling simulation effects;

[0123] The remaining multiple task scheduling simulation effects are traversed and optimized, and the target task scheduling scheme is determined by reverse matching.

[0124] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The forklift task scheduling management method and specific examples in the aforementioned embodiment 1 are also applicable to a forklift task scheduling management system in this embodiment. Through the aforementioned detailed description of a forklift task scheduling management method, those skilled in the art can clearly understand a forklift task scheduling management system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0125] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0126] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A method for scheduling and managing forklift tasks, characterized in that: The method comprises: S1: Warehouse image information and warehouse point cloud data information are acquired through visual sensing equipment and laser scanners, and the warehouse image information and warehouse point cloud data information are aligning and fused to generate a warehouse visualization model. At the same time, task requirement information is generated through a warehouse management system, and warehouse features are extracted from the task requirement information to obtain task warehouse multidimensional feature information, which includes task type, task priority, location, and quantity. S2: Based on the multi-dimensional feature information of the task warehouse, the warehouse visualization model is task matched and identified to obtain the task warehouse allocation information, and a signal sensing module is installed on the stacker, wherein the signal sensing module integrates a GPS locator, a speed sensor, a load sensor, and an energy consumption monitor, and multiple working data streams of multiple stackers are collected and obtained through the signal sensing module, and the multiple stacker working data streams are standardized to obtain multiple standard stacker working data streams; S3: Acquire multiple stacker performance parameters of multiple stackers, define task scheduling rules, perform scheduling analysis on the task storage allocation information, the multiple standard stacker work data streams and the multiple stacker performance parameters based on the task scheduling rules, and generate multiple task scheduling schemes; S4: performing differentiated evaluation and optimization based on the multiple task scheduling schemes, determining a target task scheduling scheme, and synchronizing the target task scheduling scheme to the forklift scheduling platform for task scheduling management; The step of generating a warehouse visualization model in S1 includes: S11: obtaining a warehouse feature label library through warehouse attribute information, wherein the warehouse feature label library is a type label library of each warehouse entity, and constructing a warehouse semantic segmentation network based on the warehouse feature label library; S12: performing feature segmentation on the warehouse image information using the warehouse semantic segmentation network to obtain a warehouse image semantic segmentation entity set; S13: performing data dimensionality reduction processing on the warehouse image information and the warehouse point cloud data information based on the warehouse image semantic segmentation entity set to obtain warehouse dimensionality reduction image information and warehouse dimensionality reduction point cloud data information; S14: performing feature registration and data fusion rendering on the warehouse dimension reduction image information and the warehouse dimension reduction point cloud data information to generate the warehouse visualization model; The warehouse dimension reduction image information and warehouse dimension reduction point cloud data information obtained in S13 include: Performing a criticality evaluation on each warehouse entity type label in the warehouse feature label library to obtain criticality information of the warehouse entity type label; Obtaining the precision of the target model, performing hierarchical analysis on the criticality information of the warehouse entity type label based on the precision of the target model, and determining the data density dimensionality reduction hierarchical information; Mapping and matching the data density dimensionality reduction level information with the warehouse image semantic segmentation entity set to obtain a warehouse entity level dimensionality reduction rule; Based on the warehouse entity hierarchical dimensionality reduction rule, the warehouse image information and the warehouse point cloud data information are respectively subjected to hierarchical dimensionality reduction processing to obtain the warehouse reduced dimensionality image information and the warehouse reduced dimensionality point cloud data information.

2. The method according to claim 1, characterized in that The task storage allocation information obtained in S2 includes: S21: assigning a decision order to each feature type of the task warehouse multidimensional feature information, and arranging and determining the task warehouse multidimensional feature decision information; S22: Acquire cargo location distribution characteristic information through the WMS warehouse system, mark the cargo location distribution characteristic information with remaining space, and acquire remaining space storage characteristic information; S23: Obtaining a warehouse matching strategy, wherein the warehouse matching strategy includes warehouse applicability and warehouse in / out characteristics, and performing a warehouse matching analysis on the task warehouse multidimensional feature decision information and the remaining warehouse information based on the warehouse matching strategy to obtain remaining warehouse matching feature parameter information; S24: Perform storage task identification based on the remaining storage matching feature parameter information and the warehouse visualization model to obtain the task storage allocation information.

3. The method according to claim 1, characterized in that In S3, multiple task scheduling schemes are generated, including: S31: feature mapping the task storage allocation information, the plurality of standard stacker work data streams and the plurality of stacker performance parameters with the task scheduling rules to obtain task characteristic rule parameters and stacker state rule parameters; S32: performing task scheduling association on the task characteristic rule parameter and the forklift state rule parameter to obtain a task rule scheduling parameter threshold; S33: randomly generating a plurality of scheduling parameter schemes based on the task rule scheduling parameter threshold, and dividing and clustering the plurality of scheduling parameter schemes to generate the plurality of task scheduling schemes.

4. The method according to claim 3, characterized in that Determining the target control parameter set in S33 includes: Obtaining scheduling effect evaluation indicators, wherein the scheduling effect evaluation indicators include reliability, task time, and energy consumption cost, and constructing a scheduling effect objective function according to the scheduling effect evaluation indicators; The plurality of scheduling parameter schemes are evaluated and calculated by using the scheduling effect objective function to obtain a plurality of scheduling effect fitnesses; Based on the multiple scheduling effect fitnesses, the multiple scheduling parameter schemes are proportionally divided and clustered to obtain multiple scheduling parameter scheme clusters; The multiple scheduling parameter scheme clusters are iteratively optimized, and the fitness sum of the optimized multiple scheduling parameter scheme clusters is calculated to screen and obtain the multiple task scheduling schemes.

5. The method according to claim 1, characterized in that Determining the target task scheduling scheme in S4 includes: S41: Selecting the optimal task scheduling scheme according to the multiple task scheduling schemes to perform forklift scheduling control and obtain forklift scheduling feedback parameters; S42: If the forklift scheduling feedback parameter does not reach the preset scheduling effect, simulate the remaining task scheduling schemes in the multiple task scheduling schemes to obtain the remaining multiple task scheduling simulation effects; S43: traverse and select the remaining multiple task scheduling simulation effects, and reversely match to determine the target task scheduling solution.

6. A forklift task scheduling management system, characterized in that: For implementing the steps of the method according to any one of claims 1 to 5, the system comprises: A task warehouse multi-dimensional feature information acquisition module, which is used to collect warehouse image information and warehouse point cloud data information through visual sensing equipment and laser scanners, align and fuse the warehouse image information and warehouse point cloud data information, generate a warehouse visualization model, and generate task demand information through a warehouse management system, extract warehouse features from the task demand information, and obtain task warehouse multi-dimensional feature information, which includes task type, task priority, location, and quantity; A standard stacker work data stream acquisition module, which is used to perform task matching identification on the warehouse visualization model based on the multi-dimensional feature information of the task warehouse, and obtain task warehouse allocation information. A signal sensing module is installed on the stacker, and the signal sensing module integrates a GPS locator, a speed sensor, a load sensor, and an energy consumption monitor. The signal sensing module is used to collect and obtain multiple work data streams of multiple stackers, and the multiple stacker work data streams are standardized to obtain multiple standard stacker work data streams. A task scheduling scheme generating module, the task scheduling scheme generating module is used to obtain multiple stacker performance parameters of multiple stackers, define task scheduling rules, perform scheduling analysis on the task storage allocation information, the multiple standard stacker work data streams and the multiple stacker performance parameters based on the task scheduling rules, and generate multiple task scheduling schemes; The target task scheduling scheme determination module is used to perform differentiated evaluation and optimization based on the multiple task scheduling schemes, determine the target task scheduling scheme, and synchronize the target task scheduling scheme to the forklift scheduling platform for task scheduling management.

Images