Unmanned intelligent handling system and method for chemical raw material storage warehouses

Abstract

This invention provides an unmanned intelligent handling system and method for chemical raw material storage warehouses. The system includes: a vehicle status information acquisition module, used to acquire the status information of unmanned handling vehicles within the chemical raw material storage warehouse when a chemical raw material handling task is received; a task execution vehicle selection module, used to select a task execution vehicle from the unmanned handling vehicles based on the vehicle status information and the chemical raw material handling task; and a chemical raw material handling task allocation module, used to allocate the chemical raw material handling task to the task execution vehicle for execution. This unmanned intelligent handling system and method for chemical raw material storage warehouses eliminates the need for personnel to operate forklifts / manually to handle chemical raw materials, reducing labor costs. Even if a handling accident occurs during the handling of chemical raw materials, personnel injury or death will not occur, thus improving the safety of chemical raw material handling in and out of the warehouse to a certain extent.

Description

Technical Field

[0001] This invention relates to the field of intelligent warehousing technology, and in particular to an unmanned intelligent handling system and method for chemical raw material storage warehouses. Background Technology

[0002] Currently, most chemical production workshops are equipped with a nearby chemical raw material storage warehouse to store the chemical raw materials needed for chemical production.

[0003] When chemical production workshops need to replenish chemical raw materials, or when newly ordered chemical raw materials are transported from manufacturers, the chemical raw material storage warehouse needs to handle the corresponding inbound and outbound operations. This requires warehouse staff to use forklifts or manually to move the chemical raw materials, resulting in significant labor costs. Furthermore, chemical raw materials are often harmful to humans, such as hazardous liquids and gases. Since the handling of chemical raw materials involves human intervention (e.g., chemical raw materials leaking due to impact), accidents during handling can directly cause injury, and in severe cases, even death. Secondly, human intervention introduces many uncertainties during the handling process, such as staff inattention and limited daily energy levels for forklift / manual handling, which reduces the safety of chemical raw material handling.

[0004] Therefore, a solution is urgently needed. Summary of the Invention

[0005] One of the objectives of this invention is to provide an unmanned intelligent handling system for chemical raw material storage warehouses. This system eliminates the need for staff to operate forklifts or manually handle chemical raw materials, reducing labor costs. Even if a handling accident occurs during the handling of chemical raw materials, no personnel will be injured or killed, thus improving the safety of handling chemical raw materials in and out of the warehouse to a certain extent.

[0006] The unmanned intelligent handling system for chemical raw material storage warehouses provided in this invention includes:

[0007] The vehicle status information acquisition module is used to acquire the vehicle status information of the unmanned handling vehicle in the chemical raw material storage warehouse when a chemical raw material handling task is received.

[0008] The task execution vehicle selection module is used to select a task execution vehicle from the unmanned transport vehicles based on the vehicle status information and the chemical raw material handling task.

[0009] The chemical raw material handling task allocation module is used to assign chemical raw material handling tasks to task execution vehicles.

[0010] Preferably, the task execution vehicle selection module selects a task execution vehicle from the unmanned transport vehicles based on the vehicle status information and the chemical raw material handling task, and performs the following operations:

[0011] Extract the remaining handling capacity, remaining work route, and first task completion time limit of the unmanned transport vehicle from the vehicle status information;

[0012] Extract the pickup quantity, pickup location, delivery location, and second task completion deadline from the chemical raw material handling task;

[0013] The unmanned transport vehicle with a remaining handling capacity greater than or equal to the picking capacity is designated as the first target vehicle.

[0014] The remaining working route of the first target vehicle is used as the first simulation route;

[0015] Based on the preset driving speed of the first target vehicle, the simulation time is calculated when the first target vehicle stops at the nearest pickup location to pick up the goods and continues to drive along the first simulated route until the end.

[0016] The first future time after the current time, with the first duration determined;

[0017] When the first future time is before the completion deadline of the first task, the corresponding first target car will be used as the second target car, and the first future time will be associated with the second target car.

[0018] Based on the preset warehouse map corresponding to the chemical raw material storage warehouse, a driving route from the end position of the first simulated route to the delivery location is planned and used as the second simulated route.

[0019] Based on the speed of the second target car, simulate the second time it takes for the second target car to travel along the second simulated route to the end;

[0020] The second future time after determining the first future time associated with the second target vehicle, and the second future time after the second time spent;

[0021] When the second future time is before the second task completion deadline, determine the first time difference between the second future time and the second task completion deadline, and determine the second time difference between the first future time associated with the second target vehicle and the first task completion deadline.

[0022] The second target vehicle with the largest difference between the first and second time differences is selected as the task execution vehicle.

[0023] Preferably, the unmanned intelligent handling system for chemical raw material storage warehouses also includes:

[0024] The trolley anomaly monitoring module is used to detect anomalies in the trolley when it is performing a chemical raw material handling task.

[0025] The backup trolley dispatch module is used to dispatch the nearest backup unmanned transport trolley in the chemical raw material storage warehouse to take over the chemical raw material handling task when an anomaly is detected.

[0026] Preferably, the vehicle anomaly monitoring module detects anomalies in the task-executing vehicle and performs the following operations:

[0027] Update the anomaly detection experience base;

[0028] Get the vehicle attributes of the task execution vehicle;

[0029] Based on the vehicle's attributes, generate anomaly detection experience retrieval criteria;

[0030] Retrieve target anomaly detection experience that meets the anomaly detection experience retrieval criteria from the anomaly detection experience database;

[0031] Obtain the vehicle operation information of the task execution vehicle;

[0032] Based on experience in target anomaly detection, anomaly detection is performed on the task execution vehicle according to the vehicle's operation information.

[0033] Preferably, the vehicle anomaly monitoring module updates the anomaly detection experience base by performing the following operations:

[0034] Retrieve the new experience files generated by the default anomaly detection experience generation platform after the last update of the anomaly detection experience library;

[0035] Verify the primary credibility of the party that produced the experience document;

[0036] When the verification is successful, the experience document is simplified and then stored in the anomaly detection experience library.

[0037] Otherwise, verify the second credibility of the experience document;

[0038] When the verification is successful, the experience document is simplified and then stored in the anomaly detection experience library.

[0039] Preferably, the vehicle anomaly monitoring module verifies the credibility of the party that generated the experience document by performing the following operations:

[0040] Acquire the associated library of the platform that generates anomaly detection experience;

[0041] Determine the producer and association weight of the producer association from the producer association database;

[0042] Acquire anomaly detection experience to generate a reputation value database for the platform's contributors;

[0043] Determine the first reputation value corresponding to the producer and the second reputation value corresponding to the associated producer from the producer reputation value database;

[0044] Assign a preset calculation weight to the first reputation value to obtain the first target value;

[0045] Assign a corresponding association weight to the second reputation value to obtain the second target value;

[0046] The sum of the first target value and the second target value is used as the first confidence level.

[0047] When the first level of credibility is greater than or equal to the preset first level of credibility threshold, the verification is successful.

[0048] Preferably, the vehicle anomaly monitoring module verifies the second credibility of the experience document by performing the following operations:

[0049] Determine if the interactive area of ​​the experience files in the anomaly detection experience generation platform is empty;

[0050] If so, verification failed;

[0051] Otherwise, retrieve the interaction records from the interaction area;

[0052] Semantic clustering of interaction records yields a set of interaction records with multiple semantic categories;

[0053] Determine the semantic category value from a pre-defined value database;

[0054] When the value score is greater than or equal to the preset value score threshold, the corresponding semantic category will be used as the target semantic category.

[0055] Count the number of elements in the set of interaction records for the target semantic category;

[0056] The number weight of the set elements is determined from the preset number weight library;

[0057] Determine the confidence value of the semantic category from the preset confidence value library;

[0058] Assign weights to the confidence values ​​to obtain the third target value;

[0059] The sum of the accumulated third target values ​​is used as the second confidence level;

[0060] When the second confidence level is greater than or equal to the preset second confidence level threshold, the verification is successful.

[0061] Preferably, the vehicle anomaly monitoring module simplifies the experience files and performs the following operations:

[0062] The experience documents are expanded chronologically to obtain a sequence of experience processes;

[0063] Semantic extraction is performed on each empirical process in the sequence of empirical processes to obtain the first semantics;

[0064] Match the first semantic with the second semantic in the preset principle semantic library;

[0065] If a match is found, the corresponding empirical process will be used as the principle process.

[0066] Match the first semantics of the two empirical processes preceding the principle process and the empirical process following the principle process in the empirical process sequence with the third semantics in the preset judgment semantic library;

[0067] If a match is found, the corresponding empirical process will be used as the decision process;

[0068] The empirical processes other than the decision process are removed from the empirical process sequence to obtain a simplified result.

[0069] The unmanned intelligent handling method for chemical raw material storage warehouses provided in this invention is characterized by comprising:

[0070] Step S1: When a chemical raw material handling task is received, obtain the status information of the unmanned handling vehicle in the chemical raw material storage warehouse;

[0071] Step S2: Based on the status information of the trolley and the chemical raw material handling task, select the trolley to perform the task from the unmanned handling trolleys;

[0072] Step S3: Assign the chemical raw material handling task to the task execution trolley.

[0073] Preferably, step S2: Based on the vehicle status information and the chemical raw material handling task, select a task-performing vehicle from the unmanned handling vehicles, including:

[0074] Extract the remaining handling capacity, remaining work route, and first task completion time limit of the unmanned transport vehicle from the vehicle status information;

[0075] Extract the pickup quantity, pickup location, delivery location, and second task completion deadline from the chemical raw material handling task;

[0076] The unmanned transport vehicle with a remaining handling capacity greater than or equal to the picking capacity is designated as the first target vehicle.

[0077] The remaining working route of the first target vehicle is used as the first simulation route;

[0078] Based on the preset driving speed of the first target vehicle, the simulation time is calculated when the first target vehicle stops at the nearest pickup location to pick up the goods and continues to drive along the first simulated route until the end.

[0079] The first future time after the current time, with the first duration determined;

[0080] When the first future time is before the completion deadline of the first task, the corresponding first target car will be used as the second target car, and the first future time will be associated with the second target car.

[0081] Based on the preset warehouse map corresponding to the chemical raw material storage warehouse, a driving route from the end position of the first simulated route to the delivery location is planned and used as the second simulated route.

[0082] Based on the speed of the second target car, simulate the second time it takes for the second target car to travel along the second simulated route to the end;

[0083] The second future time after determining the first future time associated with the second target vehicle, and the second future time after the second time spent;

[0084] When the second future time is before the second task completion deadline, determine the first time difference between the second future time and the second task completion deadline, and determine the second time difference between the first future time associated with the second target vehicle and the first task completion deadline.

[0085] The second target vehicle with the largest difference between the first and second time differences is selected as the task execution vehicle.

[0086] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0087] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0088] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0089] Figure 1 This is a schematic diagram of an unmanned intelligent handling system for a chemical raw material storage warehouse in an embodiment of the present invention;

[0090] Figure 2This is a flowchart of an unmanned intelligent handling method for a chemical raw material storage warehouse, as described in an embodiment of the present invention. Detailed Implementation

[0091] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0092] This invention provides an unmanned intelligent handling system for chemical raw material storage warehouses, such as... Figure 1 As shown, it includes:

[0093] The vehicle status information acquisition module 1 is used to acquire the vehicle status information of the unmanned handling vehicle in the chemical raw material storage warehouse when a chemical raw material handling task is received.

[0094] Task execution vehicle selection module 2 is used to select a task execution vehicle from the unmanned handling vehicles based on the vehicle status information and the chemical raw material handling task.

[0095] The chemical raw material handling task allocation module 3 is used to allocate chemical raw material handling tasks to task execution vehicles.

[0096] The working principle and beneficial effects of the above technical solution are as follows:

[0097] Chemical raw material handling tasks can be issued by production personnel in the chemical production workshop or by drivers transporting chemical raw materials from the factory. For example, when production personnel discover that type A chemical raw material in site D is running low, they issue a task to replenish type A chemical raw material in site D. Alternatively, when a driver is transporting type A chemical raw material to a chemical raw material storage warehouse and is nearing arrival, they issue a task to receive the newly arrived type A chemical raw material in the storage warehouse. A large number of automated guided vehicles (AGVs) are installed in the chemical raw material storage warehouse. These AGVs can perform functions such as moving and lifting chemical raw material packaging boxes / bottles, which falls within the existing technology scope of the robotics field and will not be elaborated upon. Based on the vehicle status information and the chemical raw material handling task, the most suitable AGV is selected from the AGVs to perform the task.

[0098] In practical applications, the handling of chemical raw materials in and out of the chemical raw material storage warehouse is completed by unmanned handling vehicles, achieving unmanned operation.

[0099] This application eliminates the need for staff to operate forklifts or manually handle chemical raw materials, reducing labor costs. Even if a handling accident occurs during the handling of chemical raw materials, no personnel will be injured or killed, which to some extent improves the safety of handling chemical raw materials in and out of the warehouse.

[0100] In one embodiment, the task execution vehicle selection module 2 selects a task execution vehicle from the unmanned transport vehicles based on the vehicle status information and the chemical raw material handling task, and performs the following operations:

[0101] The remaining handling capacity, remaining working route, and first task completion deadline of the automated guided vehicle (AGV) are extracted from the vehicle status information. Each AGV has a maximum load capacity, and the remaining handling capacity is the difference between the maximum load capacity and the weight of the goods currently being handled by the AGV. When a handling task is assigned to an AGV, the AGV plans its own route to perform the task. The remaining working route is the route to be taken until the assigned handling task is completed. When a handling task is assigned to an AGV, a first task completion deadline is specified, meaning the assigned handling task must be completed before this deadline.

[0102] Extract the following information from the chemical raw material handling task: pickup quantity, pickup location, delivery location, and second task completion deadline. The pickup quantity is the weight of goods the automated guided vehicle (AGV) needs to pick up during the task; the pickup location is the location the AGV needs to go to for pickup; and the delivery location is the location where goods need to be delivered. For example, if new chemical raw materials are received into the warehouse, the pickup location would be the warehouse entry point, and the delivery location would be the shelf. The second task completion deadline is the latest time the AGV can complete the task.

[0103] The unmanned transport vehicle with a remaining handling capacity greater than or equal to the picking capacity is designated as the first target vehicle.

[0104] The remaining working route of the first target vehicle is used as the first simulation route;

[0105] Based on the preset driving speed of the first target vehicle, the simulation calculates the time taken for the first target vehicle to travel along the first simulated route, then proceed to the nearest pickup location to retrieve the goods, and then continue traveling along the first simulated route to the end. Here, "traveling to the nearest location" refers to the real-time planning of the route between the first target vehicle's current location and the pickup location while traveling along the first simulated route. When the route length reaches a turning point where it changes from short to long, the vehicle selects the route of the length corresponding to the turning point to travel to the pickup location. The same principle applies to continuing to travel to the nearest location.

[0106] The first future time after the current time, with the first duration determined;

[0107] When the first future time is before the completion deadline of the first task, the corresponding first target car will be used as the second target car, and the first future time will be associated with the second target car.

[0108] Based on a preset warehouse map corresponding to the chemical raw material storage warehouse, a driving route from the end of the first simulated route to the delivery location is planned and used as the second simulated route; wherein, the preset warehouse map is an internal map of the chemical raw material storage warehouse, which is marked with roads that can be passed by unmanned transport vehicles, the types of chemical raw materials placed on each shelf, etc.

[0109] Based on the speed of the second target car, simulate the second time it takes for the second target car to travel along the second simulated route to the end;

[0110] The second future time after determining the first future time associated with the second target vehicle, and the second future time after the second time spent;

[0111] When the second future time is before the second task completion deadline, determine the first time difference between the second future time and the second task completion deadline, and determine the second time difference between the first future time associated with the second target vehicle and the first task completion deadline.

[0112] The second target vehicle with the largest difference between the first and second time differences is selected as the task execution vehicle.

[0113] The working principle and beneficial effects of the above technical solution are as follows:

[0114] Because chemical production workshops operate almost continuously and the procurement of chemical raw materials is frequent, the inbound and outbound handling of chemical raw material storage warehouses is quite busy. Normally, when a chemical raw material handling task is received, it is added to a queue and only executed after all previous tasks are completed. This reduces the utilization rate of automated guided vehicles (AGVs) and can even lead to a backlog of chemical raw material handling tasks. This invention, based on AGV status information and chemical raw material handling tasks, identifies AGVs already performing tasks that can also handle chemical raw material handling, assigning them to these tasks instead of directly queuing newly received tasks. This improves the utilization rate of AGVs and is particularly suitable for chemical raw material storage warehouses with busy inbound and outbound handling. The standard for "incidental" is that the automated guided vehicle (AGV) can pick up goods from the chemical raw material handling task midway through its journey, then deliver them to the destination of the current task, and then deliver them to the delivery location of the chemical raw material handling task. This process should not delay the completion of the current task or the chemical raw material handling task. The larger the sum of the differences between the first and second time differences, the less delay there is. Furthermore, the unloading time of the AGV during the current task is not considered, because a larger sum of the differences between the first and second time differences means that even if unloading occurs during this period, it will still minimize delays if the task is completed by the AGV.

[0115] In one embodiment, the unmanned intelligent handling system for a chemical raw material storage warehouse further includes:

[0116] The trolley anomaly monitoring module is used to detect anomalies in the trolley when it is performing a chemical raw material handling task.

[0117] The backup trolley dispatch module is used to dispatch the nearest backup unmanned transport trolley in the chemical raw material storage warehouse to take over the chemical raw material handling task when an anomaly is detected.

[0118] The working principle and beneficial effects of the above technical solution are as follows:

[0119] Backup unmanned transport vehicles (ARTVs) are installed in the chemical raw material storage warehouse. When the ARTVs malfunction, the backup ARTVs take over and complete the chemical raw material handling task, improving the stability of unmanned intelligent handling within the warehouse. During the handover, the backup ARTVs first travel to the location of the ARTVs, retrieve the goods from their loading equipment, and then deliver them.

[0120] In one embodiment, the vehicle anomaly detection module performs anomaly detection on the task execution vehicle and performs the following operations:

[0121] Update the anomaly detection experience base;

[0122] Obtain the vehicle attributes of the task execution vehicle; the vehicle attributes include: vehicle model, service duration, historical faults, etc.

[0123] Based on the vehicle's attributes, anomaly detection experience retrieval conditions are generated. These anomaly detection experience retrieval conditions are used to find retrieval conditions that are similar to the attributes of the vehicle executing the task. For example, if the vehicle's attribute is that it has been in use for 300 hours, then the generated anomaly detection experience retrieval conditions would be that the vehicle being retrieved has been in use for 300 hours ± 20 hours.

[0124] Retrieve target anomaly detection experience that meets the anomaly detection experience retrieval criteria from the anomaly detection experience database; among which, the target anomaly detection experience is the experience of performing anomaly detection on other unmanned transport vehicles with similar attributes to the task execution vehicle in the past.

[0125] Obtain the vehicle operation information of the task execution vehicle; the vehicle operation information includes: operating voltage, operating current, etc.

[0126] Based on experience in target anomaly detection, anomaly detection is performed on the task-executing vehicle according to the vehicle's operational information. For example, if the target anomaly detection experience indicates that an excessively high operating voltage of the vehicle suggests a short circuit, then the system checks whether the operating voltage in the vehicle's operational information is too high. If so, it indicates that a short circuit has occurred inside the vehicle, resulting in an anomaly.

[0127] The working principle and beneficial effects of the above technical solution are as follows:

[0128] An anomaly detection experience base is introduced to improve the applicability of anomaly detection for task execution vehicles. Furthermore, anomaly detection experience retrieval conditions are introduced to ensure that the identified target anomaly detection experience is applicable to the task execution vehicle, thereby improving the accuracy of anomaly detection for the vehicle.

[0129] In one embodiment, the vehicle anomaly monitoring module updates the anomaly detection experience base by performing the following operations:

[0130] Retrieve the newly generated experience files from the default anomaly detection experience generation platform after the last update of the anomaly detection experience library; the anomaly detection experience generation platform is a platform that generates anomaly detection experience for unmanned transport vehicles, such as a forum for unmanned transport vehicle engineers to communicate, where engineers share their experience in anomaly detection for unmanned transport vehicles, and the corresponding experience files are the engineers' sharing records.

[0131] Verify the primary credibility of the party that produced the experience document;

[0132] When the verification is successful, the experience document is simplified and then stored in the anomaly detection experience library.

[0133] Otherwise, verify the second credibility of the experience document;

[0134] When the verification is successful, the experience document is simplified and then stored in the anomaly detection experience library.

[0135] The working principle and beneficial effects of the above technical solution are as follows:

[0136] Normally, anomaly detection experience is shared by engineers from the manufacturers of automated guided vehicles (AGVs) and on-site maintenance engineers of AGVs in chemical raw material storage warehouses. However, this approach has certain limitations. This invention addresses these limitations by determining anomaly detection experience based on newly generated experience files from an anomaly detection experience generation platform, updating the anomaly detection experience library, improving the comprehensiveness of the library's construction, and enhancing the ability to detect anomalies in task-execution AGVs based on this library.

[0137] In one embodiment, the vehicle anomaly monitoring module verifies the initial credibility of the party that generated the experience document by performing the following operations:

[0138] Obtain the generator association library from the anomaly detection experience generation platform; the generator association library is provided by the anomaly detection experience generation platform, and the generator is the unmanned transport vehicle engineer who registered on the forum. During registration, the engineer can be invited or otherwise, and the relationship with the inviter is filled in during registration, so the inviter is the associated generator.

[0139] The associated producers and associated weights are determined from the producer association database. Among them, the closer the relationship between the producer and the person filling in the form and the person inviting the form, the greater the credibility guarantee of the person inviting the form (guaranteeing that the person will not engage in false car anomaly detection experience sharing after registration), the greater the associated weight.

[0140] Obtain the reputation value database of the anomaly detection experience generation platform; the reputation value database is also provided by the anomaly detection experience generation platform, which contains the reputation value corresponding to different generation platforms. The more times a generation platform has shared car anomaly detection experience in the past, the more likes it has received, etc., the higher its reputation value. Conversely, if it is reported that the shared car anomaly detection experience is fake, the reputation value will be negative.

[0141] Determine the first reputation value corresponding to the producer and the second reputation value corresponding to the associated producer from the producer reputation value database;

[0142] A preset calculation weight is assigned to the first reputation value to obtain the first target value; wherein, the calculation weight can be set manually; when assigning the weight, the first reputation value is multiplied by the calculation weight to obtain the first target value;

[0143] Assign a corresponding association weight to the second reputation value to obtain the second target value; similarly, when assigning the weight, the two are multiplied.

[0144] The sum of the first target value and the second target value is used as the first confidence level.

[0145] When the first level of credibility is greater than or equal to the preset first level of credibility threshold, the verification is successful.

[0146] The working principle and beneficial effects of the above technical solution are as follows:

[0147] The primary credibility of a generator is verified by combining its own reputation (first credibility score) and the reputation of associated generators (second credibility score), thus improving the comprehensiveness and accuracy of the verification. Furthermore, when a generator has a poor reputation (low primary credibility score), it affects the primary credibility of its associated generators, indirectly increasing the cost of errors for generators sharing false vehicle anomaly detection experience on the anomaly detection experience generation platform, thereby optimizing the order of vehicle anomaly detection experience sharing within the platform.

[0148] In one embodiment, the vehicle anomaly monitoring module verifies the second credibility of the experience document by performing the following operations:

[0149] Determine whether the interactive area of ​​the experience files in the anomaly detection experience generation platform is empty; the interactive area is the comment area corresponding to the experience files shared by the anomaly detection experience generation platform creator.

[0150] If so, the verification fails; among them, the first credibility of the creator itself fails the verification. If the interaction area is empty, it means that no one has commented, so there is no basis for the second credibility verification, and the verification fails.

[0151] Otherwise, extract the interaction records from the interaction area; these interaction records are comments.

[0152] Semantic clustering is performed on the interaction records to obtain a set of interaction records with multiple semantic categories; among which, the semantic categories include: positive reviews, negative reviews, neutral reviews, irrelevant comments, etc.

[0153] The value of a semantic category is determined from a pre-defined value library. The value library contains value values ​​corresponding to different semantic categories. The higher the value value, the greater the value of the corresponding semantic category's interaction records, i.e., comments, as a reflection of the credibility of the experience document itself. For example, the value value of irrelevant comments is 0, and the value value of positive comments is 8.

[0154] When the value score is greater than or equal to the preset value score threshold, the corresponding semantic category will be used as the target semantic category.

[0155] Count the number of elements in the set of interaction records for the target semantic category;

[0156] The number weight of the set of elements is determined from the preset number weight library; the number weight library contains number weights corresponding to different number of set elements. The larger the number of set elements, the more interaction records, i.e. comments, of the same semantic category there are, and the more persuasive they are.

[0157] The credibility value of a semantic category is determined from a pre-defined credibility value library. The pre-defined credibility value library contains credibility values ​​corresponding to different semantic categories. The higher the credibility value, the more the interaction records, i.e., comments, of the semantic category can reflect the credibility of the experience document itself.

[0158] Assign weights to the confidence values ​​to obtain the third target value;

[0159] The sum of the accumulated third target values ​​is used as the second confidence level;

[0160] When the second confidence level is greater than or equal to the preset second confidence level threshold, the verification is successful.

[0161] The working principle and beneficial effects of the above technical solution are as follows:

[0162] The accuracy of verifying the second credibility of experience documents is improved by validating the comments. Furthermore, the introduction of a value library, a number weight library, and a credibility value library enhances the efficiency of verifying the second credibility of experience documents.

[0163] In one embodiment, the vehicle anomaly monitoring module simplifies the experience file and performs the following operations:

[0164] The experience documents are expanded chronologically to obtain an experience process sequence; the expanded experience process sequence is the sharing records of engineers arranged in chronological order of sharing time, i.e., the experience process.

[0165] Semantic extraction is performed on each empirical process in the sequence of empirical processes to obtain the first semantics;

[0166] The first semantic is matched with the second semantic in the preset principle semantic library; where the second semantic is the semantic that reflects the engineer's explanation of the principle, such as "because, ...".

[0167] If a match is found, the corresponding empirical process will be used as the principle process.

[0168] The first semantics of the two empirical processes preceding the principle process and the empirical process following the principle process in the empirical process sequence are matched with the third semantics in the preset judgment semantic library; where the third semantics are the semantics of the reaction engineer to judge the abnormality of the car, such as "excessive voltage" or "short circuit occurs".

[0169] If a match is found, the corresponding empirical process will be used as the decision process;

[0170] The empirical processes other than the decision process are removed from the empirical process sequence to obtain a simplified result.

[0171] The working principle and beneficial effects of the above technical solution are as follows:

[0172] Generally, when engineers share their experience in detecting anomalies in vehicles, two scenarios occur: First, they describe the judgment process first, followed by the underlying principle, for example: "The vehicle voltage is too high," "This indicates a short circuit inside the vehicle," "Because...". Second, they interweave the principle into the judgment process, for example: "The vehicle voltage is too high," "Because...", "This indicates an internal short circuit." However, vehicle anomaly detection experience only requires the judgment process, not the principle. Therefore, we first determine the principle process. The first semantics of the two empirical processes preceding the principle process and the first empirical process following the principle process are matched with the third semantics in a pre-defined judgment semantic library. This ensures that both scenarios are covered and can be identified, with the judgment process used as a simplified result. This improves the accuracy and efficiency of simplifying experience files and enhances the space utilization efficiency of the anomaly detection experience library.

[0173] This invention provides an unmanned intelligent handling method for chemical raw material storage warehouses, such as... Figure 2 As shown, it includes:

[0174] Step S1: When a chemical raw material handling task is received, obtain the status information of the unmanned handling vehicle in the chemical raw material storage warehouse;

[0175] Step S2: Based on the status information of the trolley and the chemical raw material handling task, select the trolley to perform the task from the unmanned handling trolleys;

[0176] Step S3: Assign the chemical raw material handling task to the task execution trolley.

[0177] In one embodiment, step S2: Based on the vehicle status information and the chemical raw material handling task, selecting a task-performing vehicle from the unmanned handling vehicles includes:

[0178] Extract the remaining handling capacity, remaining work route, and first task completion time limit of the unmanned transport vehicle from the vehicle status information;

[0179] Extract the pickup quantity, pickup location, delivery location, and second task completion deadline from the chemical raw material handling task;

[0180] The unmanned transport vehicle with a remaining handling capacity greater than or equal to the picking capacity is designated as the first target vehicle.

[0181] The remaining working route of the first target vehicle is used as the first simulation route;

[0182] Based on the preset driving speed of the first target vehicle, the simulation time is calculated when the first target vehicle stops at the nearest pickup location to pick up the goods and continues to drive along the first simulated route until the end.

[0183] The first future time after the current time, with the first duration determined;

[0184] When the first future time is before the completion deadline of the first task, the corresponding first target car will be used as the second target car, and the first future time will be associated with the second target car.

[0185] Based on the preset warehouse map corresponding to the chemical raw material storage warehouse, a driving route from the end position of the first simulated route to the delivery location is planned and used as the second simulated route.

[0186] Based on the speed of the second target car, simulate the second time it takes for the second target car to travel along the second simulated route to the end;

[0187] The second future time after determining the first future time associated with the second target vehicle, and the second future time after the second time spent;

[0188] When the second future time is before the second task completion deadline, determine the first time difference between the second future time and the second task completion deadline, and determine the second time difference between the first future time associated with the second target vehicle and the first task completion deadline.

[0189] The second target vehicle with the largest difference between the first and second time differences is selected as the task execution vehicle.

[0190] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An unmanned intelligent handling system for chemical raw material storage warehouses, characterized in that, include: The vehicle status information acquisition module is used to acquire the vehicle status information of the unmanned handling vehicle in the chemical raw material storage warehouse when a chemical raw material handling task is received. The task execution vehicle selection module is used to select a task execution vehicle from the unmanned transport vehicles based on the vehicle status information and the chemical raw material handling task. A chemical raw material handling task allocation module is used to allocate the chemical raw material handling task to the task execution vehicle for execution; The task execution vehicle selection module selects a task execution vehicle from the unmanned transport vehicles based on the vehicle status information and the chemical raw material handling task, and performs the following operations: Extract the remaining handling capacity, remaining working route, and first task completion time limit of the unmanned transport vehicle from the vehicle status information; Extract the picking quantity, picking location, delivery location, and second task completion time limit from the chemical raw material handling task; The unmanned transport vehicle whose remaining handling capacity is greater than or equal to the picking quantity is designated as the first target vehicle; The remaining working route of the first target vehicle is used as the first simulated route; Based on the preset driving speed of the first target vehicle, the simulation time is calculated when the first target vehicle travels along the first simulated route, goes to the nearest pickup location to pick up the goods, and then continues to travel along the first simulated route until the end. Determine the first future time after the current time, the time after which the first duration has been spent. When the first future time is before the first task completion deadline, the first target car will be designated as the second target car, and the first future time will be associated with the second target car. Based on the preset warehouse map corresponding to the chemical raw material storage warehouse, a driving route from the end position of the first simulated route to the delivery location is planned and used as the second simulated route. Based on the vehicle speed corresponding to the second target vehicle, simulate the second time taken for the second target vehicle to travel along the second simulated route to the end; The second future time after determining the first future time associated with the second target vehicle and the second time of the second duration; When the second future time is before the second task completion deadline, determine the first time difference between the second future time and the second task completion deadline, and determine the second time difference between the first future time associated with the second target vehicle and the first task completion deadline; The second target vehicle with the largest sum of the first time difference and the second time difference is selected as the task execution vehicle.

2. The unmanned intelligent handling system for chemical raw material storage warehouses as described in claim 1, characterized in that, Also includes: The trolley anomaly monitoring module is used to detect anomalies in the trolley when it is performing the chemical raw material handling task. The backup trolley scheduling module is used to dispatch the nearest backup unmanned transport trolley in the chemical raw material storage warehouse to take over the chemical raw material transport task when an anomaly is detected.

3. The unmanned intelligent handling system for chemical raw material storage warehouses as described in claim 2, characterized in that, The vehicle anomaly monitoring module performs anomaly detection on the task execution vehicle and performs the following operations: Update the anomaly detection experience base; Obtain the vehicle attributes of the task execution vehicle; Based on the vehicle attributes, generate anomaly detection experience retrieval conditions; Retrieve target anomaly detection experience that meets the anomaly detection experience retrieval conditions from the anomaly detection experience database; Obtain the vehicle operation information of the task execution vehicle; Based on the aforementioned target anomaly detection experience, and according to the vehicle operation information, anomaly detection is performed on the task execution vehicle.

4. The unmanned intelligent handling system for chemical raw material storage warehouses as described in claim 3, characterized in that, The vehicle anomaly monitoring module updates the anomaly detection experience base by performing the following operations: Obtain the newly generated experience file from the preset anomaly detection experience generation platform after the last update of the anomaly detection experience library; The credibility of the party that generated the experience document is verified. When the verification is successful, the experience document is simplified and then stored in the anomaly detection experience library. Otherwise, the second credibility of the experience document is verified; When the verification is successful, the experience document is simplified and then stored in the anomaly detection experience library.

5. The unmanned intelligent handling system for chemical raw material storage warehouses as described in claim 4, characterized in that, The vehicle anomaly monitoring module verifies the first credibility of the party that generated the experience document by performing the following operations: Obtain the associated library of the anomaly detection experience generation platform; Determine the associated producer and associated weight of the associated producer from the associated producer database; Obtain the reputation value database of the anomaly detection experience generation platform; The first reputation value corresponding to the producer and the second reputation value corresponding to the associated producer are determined from the producer reputation value database, respectively. Assign a preset calculation weight to the first reputation value to obtain the first target value; Assign the second reputation value to the associated weight to obtain the second target value; The first confidence level is calculated by summing the first target value and the second target value. When the first confidence level is greater than or equal to the preset first confidence level threshold, the verification is successful.

6. The unmanned intelligent handling system for chemical raw material storage warehouses as described in claim 4, characterized in that, The vehicle anomaly monitoring module verifies the second credibility of the experience document by performing the following operations: Determine whether the interactive area of ​​the experience file in the anomaly detection experience generation platform is empty; If so, verification failed; Otherwise, extract the interaction record from the interaction area; Semantic clustering is performed on the interaction records to obtain a set of interaction records with multiple semantic categories; The value of the semantic category is determined from a preset value database; When the value is greater than or equal to a preset value threshold, the corresponding semantic category will be used as the target semantic category. Count the number of elements in the set of interaction records for the target semantic category; The number weight of the set elements is determined from a preset number weight library; Determine the confidence value of the semantic category from a preset confidence value library; Assign the confidence value to the corresponding numerical weight to obtain the third target value; The sum of the accumulated values ​​of the third target value is used as the second confidence level; When the second confidence level is greater than or equal to the preset second confidence level threshold, the verification is successful.

7. The unmanned intelligent handling system for chemical raw material storage warehouses as described in claim 4, characterized in that, The vehicle anomaly monitoring module simplifies the experience file by performing the following operations: The experience documents are expanded chronologically to obtain a sequence of experience processes; Semantic extraction is performed on each empirical process in the sequence of empirical processes to obtain the first semantics; Match the first semantic with the second semantic in the preset principle semantic library; If a match is found, the corresponding empirical process will be taken as the principle process; The first semantics of the two empirical processes preceding the principle process and the empirical process following the principle process in the empirical process sequence are matched with the third semantics in a preset judgment semantic library. If a match is found, the corresponding empirical process will be used as the determination process; The empirical process other than the decision process is removed from the empirical process sequence to obtain a simplified result.

8. An unmanned intelligent handling method for chemical raw material storage warehouses, characterized in that, include: Step S1: When a chemical raw material handling task is received, obtain the status information of the unmanned handling vehicle in the chemical raw material storage warehouse; Step S2: Based on the vehicle status information and the chemical raw material handling task, select a task execution vehicle from the unmanned handling vehicles; Step S3: Assign the chemical raw material handling task to the task execution trolley; Step S2: Based on the vehicle status information and the chemical raw material handling task, select a task execution vehicle from the unmanned handling vehicles, including: Extract the remaining handling capacity, remaining working route, and first task completion time limit of the unmanned transport vehicle from the vehicle status information; Extract the picking quantity, picking location, delivery location, and second task completion time limit from the chemical raw material handling task; The unmanned transport vehicle whose remaining handling capacity is greater than or equal to the picking quantity is designated as the first target vehicle; The remaining working route of the first target vehicle is used as the first simulated route; Based on the preset driving speed of the first target vehicle, the simulation time is calculated when the first target vehicle travels along the first simulated route, goes to the nearest pickup location to pick up the goods, and then continues to travel along the first simulated route until the end. Determine the first future time after the current time, the time after which the first duration has been spent. When the first future time is before the first task completion deadline, the first target car will be designated as the second target car, and the first future time will be associated with the second target car. Based on the preset warehouse map corresponding to the chemical raw material storage warehouse, a driving route from the end position of the first simulated route to the delivery location is planned and used as the second simulated route. Based on the vehicle speed corresponding to the second target vehicle, simulate the second time taken for the second target vehicle to travel along the second simulated route to the end; The second future time after determining the first future time associated with the second target vehicle and the second time of the second duration; When the second future time is before the second task completion deadline, determine the first time difference between the second future time and the second task completion deadline, and determine the second time difference between the first future time associated with the second target vehicle and the first task completion deadline; The second target vehicle with the largest sum of the first time difference and the second time difference is selected as the task execution vehicle.

Images