Service method, system and device based on artificial intelligence and storage medium

Through service methods and systems based on artificial intelligence, the problem of inefficient multi-task scheduling in the existing technology is solved, and efficient communication of robot tasks and user experience is achieved.

CN119047770AInactive Publication Date: 2024-11-29HAINAN RUIXIANG SHIPPING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411147665.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-29
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The prior art is difficult to effectively schedule robots that perform multiple tasks, resulting in inefficiency and inability to adjust the task execution status in time to meet user needs.

Method used

Using artificial intelligence-based service methods and systems, the intelligent management center receives command tasks, determines the execution robot, and plans the command communication strategy, including conveying routes and time series, and responds to the robot's communication completion notification to mark the task as completed.

Benefits of technology

It improves the work efficiency of the robot, ensures timely and efficient communication of tasks, and improves the user experience and task completion speed.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119047770A_ABST
    Figure CN119047770A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides a service method, system and device based on artificial intelligence and a storage medium, and the method is executed based on an intelligent management center, and comprises the steps: receiving an instruction task issued by a manager; determining an execution robot corresponding to the instruction task based on the instruction object of the instruction task; planning and executing an instruction transmission strategy of the robot based on the instruction task and the instruction object thereof; and in response to receiving a transmission completion notification from the executing robot, marking the corresponding instruction task as completed.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of intelligent device technology, and in particular to an artificial intelligence-based service method, system, device and storage medium. Background Art

[0002] Automatic robot route planning is usually for a robot to perform a single task. If the robot has multiple tasks at the same time, it is necessary to manually preset the order of task execution to determine the robot's movement route, which is inefficient. At the same time, when the robot is performing a task, it is often unable to adjust and update the task execution status in a timely and automatic manner based on the environmental information during the task execution to meet user needs.

[0003] In response to the problem of robot scheduling, WO2018223638A1 proposes a scheduling method and device for a service robot. The application focuses on selecting the best robot according to the current task and preset scheduling rules to perform the work corresponding to the current task. However, it does not involve the reasonable scheduling of robots to perform multiple tasks.

[0004] Therefore, it is hoped to provide an artificial intelligence-based service method, system, device and storage medium to achieve intelligent scheduling of robots that need to perform multiple tasks, so that the robots can efficiently complete all tasks to be performed, improve the robot's work efficiency and enhance the user experience. Summary of the invention

[0005] One of the embodiments of the present specification provides a service method based on artificial intelligence, characterized in that the method is executed by an intelligent management center, and the method includes: receiving an instruction task issued by a management personnel; determining an execution robot corresponding to the instruction task based on the instruction object of the instruction task; planning an instruction communication strategy for the execution robot based on the instruction task and its instruction object, the instruction communication strategy at least including a communication route and a communication time sequence, the communication time sequence including a preset stay time of the execution robot at at least one preset point in the communication route; in response to receiving a communication completion notification from the execution robot, marking the corresponding instruction task as completed.

[0006] One of the embodiments of the present specification provides a service system based on artificial intelligence, characterized in that the system includes a receiving module, a determining module, a planning module and a marking module: a receiving module, used to receive an instruction task issued by a management personnel; a determining module, used to determine an execution robot corresponding to the instruction task based on the instruction object of the instruction task; a planning module, used to plan an instruction communication strategy of the execution robot based on the instruction task and its instruction object, wherein the instruction communication strategy includes at least a communication route and a communication time sequence, wherein the communication time sequence includes a preset stay time of the execution robot at at least one preset point in the communication route; a marking module, used to mark the corresponding instruction task as completed in response to receiving a communication completion notification from the execution robot.

[0007] One of the embodiments of the present specification provides an artificial intelligence-based service device, including a processor, wherein the processor is used to execute an artificial intelligence-based service method.

[0008] One of the embodiments of the present specification provides a computer-readable medium, wherein the storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a service method based on artificial intelligence.

[0009] Some embodiments of the present specification include at least the following beneficial effects: (1) by determining the execution robot corresponding to the instruction task according to the instruction object of the instruction task, an execution robot that is more suitable for executing the instruction task can be determined; by planning the instruction communication strategy of the execution robot based on the instruction task and its instruction object, the most efficient instruction communication strategy can be obtained, thereby improving the robot's work efficiency and ensuring successful communication of the instruction task; (2) by determining the corresponding execution robot according to the instruction task and the instruction object, it can be ensured that the instruction task is transmitted to the instruction object in a timely and efficient manner, thereby improving the work efficiency of the execution robot and the completion speed of the instruction task. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is an exemplary flow chart of an artificial intelligence-based service method according to some embodiments of this specification;

[0011] Figure 2 is an exemplary schematic diagram of a planning instruction delivery strategy according to some embodiments of this specification;

[0012] Figure 3 is an exemplary schematic diagram of applying a collision determination model according to some embodiments of the present specification. DETAILED DESCRIPTION

[0013] The following is a brief introduction to the drawings required for describing the embodiments. The drawings do not represent all implementation methods.

[0014] It should be understood that the "system", "device", "unit" and / or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0015] As shown in this specification and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0016] When the operations performed in the embodiments of this specification are described in steps, unless otherwise specified, the order of the steps is interchangeable, steps may be omitted, and other steps may be included in the operation process.

[0017] In response to the problem of robot scheduling, the prior art WO2018223638A1 proposes a scheduling method and device for a service robot, but the prior art does not consider intelligent scheduling when the robot needs to perform multiple tasks.

[0018] In view of this, the present invention describes some embodiments, which receive instruction tasks issued by managers, plan the instruction communication strategy of the execution robot based on the instruction tasks and their instruction objects, and update the instruction communication strategy of the execution robot through a collision determination model, determine the instruction communication strategies of multiple execution robots, thereby realizing reasonable planning of the work routes of robots that need to perform multiple tasks.

[0019] Some embodiments of this specification disclose an artificial intelligence-based service system. The artificial intelligence-based service system can realize intelligent scheduling of robots by implementing the methods and / or processes disclosed in this application.

[0020] In some embodiments, the artificial intelligence-based service system includes a receiving module, a determining module, a planning module, and a marking module.

[0021] The receiving module is used to receive instructions and tasks issued by managers.

[0022] The determination module is used to determine the execution robot corresponding to the instruction task based on the instruction object of the instruction task.

[0023] In some embodiments, the determination module can determine the effective communication range of the instruction object based on the personal data and activity range of the instruction object; in response to the presence of at least one robot in the effective communication range, one of the at least one robots is determined as the execution robot corresponding to the instruction task, and the instruction task is added to the task list of the execution robot.

[0024] The planning module is used to plan the command communication strategy of the execution robot based on the command task and its command object.

[0025] In some embodiments, the planning module can determine the task importance of the target instruction task based on the task type, instruction-associated object, and associated tasks of the target instruction task in the task list of the executing robot; determine the communication route through a preset algorithm based on the task importance of the target instruction task, the object position data of the instruction-associated object, and the robot position data of the executing robot; determine the communication time sequence based on the communication route and the task importance of the target instruction task.

[0026] In some embodiments, when the execution robot includes at least one, the planning module can also update the instruction communication strategy of at least one execution robot through the collision determination model.

[0027] The marking module is used to mark the corresponding instruction task as completed in response to receiving a communication completion notification from the execution robot.

[0028] In some embodiments, the artificial intelligence-based service system can be applied to various scenarios where robots perform tasks, such as work scenarios, warehousing sorting scenarios, assembly line operations, and other scenarios.

[0029] For more information about the above parts, see Figure 1-3 Related description.

[0030] Figure 1 is an exemplary flow chart of an artificial intelligence-based service method according to some embodiments of this specification. In some embodiments, process 100 may be executed by an intelligent management center. Figure 1 As shown, process 100 includes the following steps:

[0031] The intelligent management center can be used to manage resources and process data and / or information from at least one component or in the artificial intelligence-based service system. In some embodiments, the intelligent management center can be a single server or a server group. In some embodiments, the intelligent management center can be regional or remote. In some embodiments, the intelligent management center can be implemented on a cloud platform or provided in a virtual manner.

[0032] Step 110: receiving the instruction task issued by the management personnel.

[0033] Managers can be people who set directive tasks.

[0034] Instruction tasks refer to instructions conveyed by the robot or tasks to be performed. For example, in a work scenario, an instruction task can be a meeting notification, a commendation notice, etc. For another example, in a warehouse sorting scenario, an instruction task can be to move goods to a designated location.

[0035] Instruction tasks may include multiple types. In some embodiments, instruction tasks include notification instructions and execution instructions. Notification instructions refer to instructions that need to be conveyed by the robot, which may include positive notification instructions and negative notification instructions. For example, in a work scenario, a positive notification instruction may be a praise for an employee, and a negative notification instruction may be a criticism of an employee, etc. Execution instructions refer to instructions that contain tasks that need to be performed by the robot. For example, in a work scenario, execution instructions may be confirming whether there is someone at a certain workstation, distributing documents, etc.

[0036] In some embodiments, the administrator can input the instruction task into the intelligent management center in a variety of ways. For example, the administrator can input the instruction task into the intelligent management center by voice input, manual input, etc.

[0037] Step 120: determining an execution robot corresponding to the instruction task based on the instruction object of the instruction task.

[0038] The instruction object refers to the object to which the instruction task refers. For example, in a work scenario, when the instruction task is to notify a meeting, the corresponding instruction object is the employee who needs to attend the meeting. For another example, in a warehouse sorting scenario, when the instruction task is to move goods, the corresponding instruction object is the goods that need to be moved.

[0039] In some embodiments, one instruction task may point to multiple instruction objects. In some embodiments, multiple instruction tasks may correspond to the same instruction object.

[0040] An execution robot is a robot that transmits a command task. The execution robot can transmit the command task to the command object corresponding to the command task.

[0041] In some embodiments, the execution robot corresponding to the instruction task can be determined in a variety of ways. For example, the intelligent management center can determine the robot that has transmitted the instruction task to the instruction object the most times as the execution robot based on historical data.

[0042] The following uses a work scenario as an example to illustrate how to determine the execution robot.

[0043] In some embodiments, the intelligent management center can determine the effective communication range of the instruction object based on the personal data and activity range of the instruction object; in response to the presence of at least one robot in the effective communication range, one of the at least one robots is determined as the execution robot corresponding to the instruction task, and the instruction task is added to the task list of the execution robot.

[0044] Personal data refers to relevant information of the instruction object. For example, personal data may include the position and workstation of the instruction object (e.g., employee). The position and workstation of the instruction object can be determined by querying the employee information table.

[0045] The activity range refers to the area where the instruction object often moves. For example, the activity range of an employee can be determined based on the employee's historical activity trajectory.

[0046] The effective communication range refers to the range within which the instruction object can receive the instruction task. In some embodiments, the effective communication range can be determined by querying an information table containing the corresponding relationship between the personal data of the instruction object, the activity range and the effective communication range. Among them, the information table containing the corresponding relationship between the personal data of the instruction object, the activity range and the effective communication range can be determined by prior knowledge or historical data.

[0047] In some embodiments, in response to the existence of at least one robot in the effective communication range, the intelligent management center may determine the robot with the least amount of tasks in the effective communication range as the execution robot. In some embodiments, in response to the existence of at least one robot in the effective communication range, the intelligent management center may also determine the robot that transmits the most instruction tasks to the instruction object within the effective communication range as the execution robot.

[0048] In some embodiments, in response to the absence of a robot in the effective communication range, the intelligent management center may determine the robot closest to the instruction object as the execution robot.

[0049] In some embodiments of the present specification, in response to the absence of a robot in the effective communication range, the robot closest to the instruction object is determined as the execution robot, which is conducive to the rapid communication of the instruction task.

[0050] The task list includes a set of command tasks that the robot needs to complete. Different command tasks in the task list can be arranged in the order of execution.

[0051] In some embodiments, the execution order can be determined based on the distance between the executing robot and the instruction object. For example, the closer the distance, the earlier the execution order. In some embodiments, the execution order can be determined based on the importance of the task. For example, the higher the importance of the task, the earlier the execution order. For more information about the importance of tasks, see Figure 2and its related description.

[0052] In some embodiments, the task list can be stored in the internal memory of the robot. In some embodiments, the task list can be stored in the intelligent management center. This specification does not limit this.

[0053] In some embodiments, the intelligent management center can add the instruction task to the task list of the execution robot through communication control.

[0054] In some embodiments, the newly added instruction task can be at any position in the task list. For example, the newly added instruction task can be at the top of the task list, indicating that the task is executed first; or at the bottom of the task list, indicating that the task is executed last.

[0055] In some embodiments of the present specification, determining the corresponding execution robot according to the instruction task and the instruction object can ensure that the instruction task is transmitted to the instruction object in a timely and efficient manner, thereby improving the work efficiency of the execution robot and the completion speed of the instruction task.

[0056] Step 130, based on the instruction task and its instruction object, plan the instruction delivery strategy of the execution robot.

[0057] The command delivery strategy refers to the plan followed by the execution robot when delivering multiple command tasks in its task list.

[0058] In some embodiments, the instruction delivery strategy includes at least a delivery route and a delivery time sequence.

[0059] The transmission route can be the route that the robot takes to execute multiple command tasks in the transmission task list. For example, the transmission route can be to first reach the position corresponding to command task A, move to the position corresponding to command task B after command task A is completed, and move to the position corresponding to command task C after task B is completed.

[0060] The communication time sequence may be a time sequence for executing each instruction task in the robot communication task list.

[0061] In some embodiments, the communication time sequence may include executing a preset stop time of the robot at at least one preset point in the communication route.

[0062] The preset point may be a preset stop point where the executing robot will stop after reaching the preset point. For example, in a work scenario, the preset point may be the location of office equipment (e.g., a printer, a fax machine, etc.). For another example, the preset point may be a task execution location close to the command task (e.g., the workstation location of the command object).

[0063] The task execution position refers to the position where the robot is located when executing the command task. For example, in a work scene, the task execution position corresponding to the command task can be the workstation where the command object is located.

[0064] The preset dwell time may be the time for the execution robot to stay at a preset point.

[0065] In some embodiments, the transmission time sequence may also include the arrival time of the execution robot arriving at the location of at least one instruction object. The arrival time is the time for the execution robot to move from the initial position (i.e., the position of the execution robot before starting the transmission instruction task) or the task execution position of the previous instruction task to the corresponding position of the current instruction task.

[0066] In some embodiments, the command transmission strategy for executing the robot can be determined in a variety of ways. For example, the intelligent management center can plan multiple routes based on the initial position of the robot and the position of the command object corresponding to each command task, and determine the shortest route as the transmission route. For another example, the intelligent management center can determine the corresponding preset residence time based on the type of each command task. Among them, the correspondence between different task types and different residence times can be determined based on prior knowledge or historical data.

[0067] In some embodiments, the intelligent management center can determine the transmission route through a preset algorithm based on the task importance of the target instruction task, the object position data of the target instruction object, and the robot position data of the executing robot; and determine the transmission time sequence based on the transmission route and the task importance of the target instruction task. For more information about this embodiment, see Figure 3 and its related description.

[0068] In some embodiments, when the execution robot includes at least one, the intelligent management center can update the instruction transmission strategy of at least one execution robot through the collision determination model. For more information about this embodiment, see Figure 3 and its related description.

[0069] Step 140 , in response to receiving a communication completion notification from the execution robot, marking the corresponding instruction task as completed.

[0070] The communication completion notification may be a notification message of the completion of the instruction task sent by the execution robot to the intelligent management center when the execution robot completes the communication of the instruction task.

[0071] When the execution robot completes one or more instruction tasks in its task list, the intelligent management center can mark the one or more instruction tasks as completed.

[0072] Marking a command task as completed indicates that the command task has been completed. The intelligent management center can delete the command task from the task list of the execution robot or place it at the end of the task list.

[0073] In some embodiments of the present specification, by determining the execution robot corresponding to the instruction task based on the instruction object of the instruction task, an execution robot that is more suitable for executing the instruction task can be determined; by planning the instruction communication strategy of the execution robot based on the instruction task and its instruction object, the most efficient instruction communication strategy can be obtained, thereby improving the work efficiency of the robot and ensuring the successful communication of the instruction task.

[0074] It should be noted that the above description of the relevant process is only for example and explanation, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to the process under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

[0075] Figure 2 It is an exemplary schematic diagram of a planning instruction delivery strategy according to some embodiments of the present specification.

[0076] In some embodiments, the intelligent management center can determine the task importance 222 of the target instruction task based on the task type 211, instruction-related object 212, and associated task 213 of the target instruction task in the task list of the executing robot; determine the communication route 270 through a preset algorithm based on the task importance 222 of the target instruction task, the object position data 221 of the instruction-related object 212, and the robot position data 223 of the executing robot; and determine the communication time series 280 based on the communication route 270 and the task importance 222 of the target instruction task.

[0077] The target instruction task refers to the instruction task that the execution robot needs to convey. In some embodiments, the target instruction task may include all or part of the instruction tasks in the task list of the execution robot. The instruction tasks in the task list may include the part with the highest execution order, the remaining part except the instruction tasks marked as completed, etc.

[0078] The task type 211 of the target command task is the same as the task type of the command task. For more information, see Figure 1 And the related description will not be repeated here.

[0079] The instruction-related object 212 refers to the personnel related to the instruction task. The instruction-related object 212 includes the target instruction object and the management personnel. The target instruction object refers to the instruction object corresponding to the target instruction task. For more information about instruction objects and management personnel, see Figure 1 and its related description.

[0080] The associated task 213 refers to an instruction task related to the target instruction task. For example, if the transmission of a certain instruction task needs to be carried out after the target instruction task is completed, then the instruction task is an associated task of the target instruction task. For another example, if the instruction object corresponding to a certain instruction task is the same as or close to the target instruction object corresponding to the target instruction task, then the instruction task is an associated task of the target instruction task.

[0081] In some embodiments, the intelligent management center can analyze all instruction tasks in the task list of the execution robot to determine the associated tasks of the target instruction task.

[0082] The task importance level 222 of the target instruction task is an indicator used to indicate the importance of the target instruction task.

[0083] In some embodiments, the intelligent management center can construct a vector to be matched based on the task type 211, the instruction-related object 212, and the associated task 213 of the target instruction task of the robot, and determine the task importance 222 of the target instruction task based on the search result of the vector to be matched in the vector database. For example, the vector to be matched can be (x1, y1, z1). Among them, x1 represents the task type 211 of the target instruction task, y1 represents the instruction-related object 212, and z1 represents the associated task 213.

[0084] In some embodiments, the vector database may include multiple groups of reference vectors and their corresponding task importances. The reference vectors may be constructed based on the task type of the historical target instruction task, the instruction associated object corresponding to the historical target instruction task, and the associated task corresponding to the historical target instruction task. The task importance corresponding to the reference vector may be determined by the system or manually after evaluation based on historical data.

[0085] In some embodiments, the intelligent management center may determine a vector that meets a preset condition in a vector database based on the vector to be matched, and determine the vector that meets the preset condition as an associated vector. The preset condition may refer to a judgment condition for determining an associated vector. In some embodiments, the preset condition may include that the vector distance is less than a distance threshold, the vector distance is minimum, etc. In some embodiments, the intelligent management center may determine the task importance of the target instruction task based on the task importance corresponding to the associated vector.

[0086] The object location data refers to data related to the location of the instruction-related object. In some embodiments, the object location data includes the location data of the target instruction object and the location data of the management personnel.

[0087] In some embodiments, the intelligent management center can obtain the object location data based on the positioning device of the instruction-related object. The positioning device of the instruction-related object can be a smart phone, smart watch, etc. of the target instruction object.

[0088] The robot position data refers to data related to the position of the executing robot. In some embodiments, the intelligent management center can obtain the robot position data based on the positioning device of the executing robot.

[0089] The preset algorithm refers to a pre-set algorithm for determining the transmission route.

[0090] In some embodiments, the preset algorithm may be in various forms. For example, the preset algorithm may be: determining the route with the least total movement distance of the robot as the transmission route.

[0091] In some embodiments, the preset algorithm may include: determining the virtual distance 250 for the executing robot to perform a target instruction task based on the task importance 222 of the target instruction task, the task type 211 of the target instruction task and the moving distance 240 for the executing robot to perform a target instruction task; determining the transmission route 270 of the executing robot through the shortest route algorithm based on the virtual distance 250 for the executing robot to perform each target instruction task and the task execution position 260 corresponding to each target instruction task.

[0092] The moving distance 240 refers to the actual physical distance that the robot needs to travel to perform a target instruction task. In some embodiments, the intelligent management center can calculate the moving distance 240 that the robot needs to travel to perform a target instruction task based on the object position data 221 and the robot position data 223, taking into account the terrain and / or house type of the actual application scenario.

[0093] The virtual distance 250 refers to the weighted distance for the robot to execute a certain target instruction task.

[0094] In some embodiments, the intelligent management center can determine the virtual distance 250 for the robot to perform a certain target instruction task by weighting the task importance 222 of the target instruction task, the task type 211 of the target instruction task, and the moving distance 240 of the robot to perform a certain target instruction task. In some embodiments, the calculation formula of the virtual distance can be: the virtual distance for the robot to perform a certain target instruction task = A*task importance of the target instruction task + B*task type of the target instruction task + C*moving distance for the robot to perform a certain target instruction task. Among them, A, B, and C are weights. In some embodiments, the weights can be determined based on historical experience and / or actual conditions. For example, when the task importance of a certain target instruction task is high, A can be set to a negative number to make the virtual distance as small as possible to ignore the influence of the actual physical distance, so that the robot can give priority to the target instruction task during the communication process.

[0095] For more information about the task execution location, see Figure 1 and its related description.

[0096] The shortest path algorithm may include Dijkstra's algorithm, Bellman-Ford's algorithm, Floyd's algorithm, and the like.

[0097] In some embodiments, the intelligent management center can determine the route with the shortest total virtual distance for the execution robot to complete all target instruction tasks based on the virtual distance 250 of the execution robot to execute each target instruction task and the task execution position 260 corresponding to each target instruction task through the shortest path algorithm, and determine the route as the communication route 270.

[0098] In some embodiments, each time the execution robot completes a target instruction task, the intelligent management center can re-determine the transmission route of the remaining target instruction tasks.

[0099] In some embodiments, each time a new target instruction task appears, the intelligent management center can re-determine the delivery route.

[0100] In some embodiments, the communication route is related to scene data where the target instruction object is located.

[0101] The scene data refers to information related to the scene that the target instruction object is in. For example, the scene data may include the number of people around, whether the target instruction object is at the task execution location, and whether the task execution location corresponds to the target instruction object.

[0102] The following will take the work scene as an example to explain the scene data related to the target instruction object in the transmission route. For example, in the work scene, the target instruction object is an employee, and the corresponding scene data may include the number of people around the workstation, whether the workstation corresponds to the employee, whether the employee is at his or her workstation, etc.

[0103] In some embodiments, the intelligent management center can determine the number of people around the workstation of the target instruction object based on the positioning devices of each employee in the work area.

[0104] In some embodiments, the intelligent management center can control the execution robot to verify the identity of the person when arriving at the task execution location to determine whether the workstation corresponds to the employee. When there are other employees at the workstation of the target instruction object, the transmission of the target instruction task is suspended.

[0105] In some embodiments, the intelligent management center can determine whether an employee is at his or her workstation by obtaining object location data and the location of the employee's corresponding workstation.

[0106] In some embodiments, the communication of the target instruction task needs to ensure that the employee is at his / her workstation before it can be carried out. For example, the instruction task may include dispatching a document, and the target instruction task of dispatching a document needs to ensure that the employee is at his / her workstation and signs for it before it can be carried out.

[0107] In some embodiments, the intelligent management center can calculate the physical distance between the employee and the workstation. When it is identified that the physical distance between the employee and the workstation exceeds a distance threshold, it is determined that the employee is not at his or her workstation and the target instruction task is delayed.

[0108] In some embodiments, the intelligent management center can determine the physical distance between the employee and the workstation based on the employee's positioning device and the employee's workstation location.

[0109] In some embodiments, the distance threshold may be preset based on historical experience or historical data.

[0110] In some embodiments, the distance threshold is related to the task importance of the target instruction task. The higher the task importance of the target instruction task, the larger the distance threshold. The higher the task importance of the target instruction task, the longer the execution robot can be allowed to wait at the task execution position, so the distance threshold can be larger.

[0111] In some embodiments, the distance threshold is related to regional environmental characteristics.

[0112] Regional environmental characteristics refer to environmental characteristics within the work area. In some embodiments, regional environmental characteristics may include personnel distribution in each sub-area within the region. Personnel distribution refers to the distribution of the number of personnel in each sub-area. The intelligent management center can divide the work area into multiple sub-areas based on various reasonable methods, such as dividing the work area into multiple sub-areas.

[0113] In some embodiments, the more dispersed the personnel distribution is, the larger the distance threshold can be. When the personnel distribution in each sub-area of ​​the region is relatively dispersed, it can be inferred that the probability that the target instruction object corresponding to other target instruction tasks other than the currently transmitted target instruction task is not at their own workstation is relatively high. At this time, determining a larger distance threshold can enable the executing robot to transmit the current target instruction task, avoiding the waste of computing resources caused by delayed transmission.

[0114] In some embodiments, the distance threshold may also be determined by a distance threshold determination model.

[0115] The distance threshold determination model is a machine learning model, such as a neural network model (NN) and the like.

[0116] In some embodiments, the input of the distance threshold determination model includes regional environmental characteristics, the task type of the target instruction task, and the task importance of the target instruction task; and the output includes the distance threshold.

[0117] In some embodiments, the distance threshold determination model can be based on the training acquisition of the first training sample with the first label. In some embodiments, each group of training samples of the first training sample can include sample area environmental characteristics, task types of sample target instruction tasks, and task importance of sample target instruction tasks. The first label of each group of first training samples is the distance threshold corresponding to the group of first training samples.

[0118] In some embodiments, the intelligent management center may determine the first training sample based on historical data.

[0119] In some embodiments, the intelligent management center can test different candidate distances for the execution robot during the actual communication process based on the first training sample, and determine the candidate distance with good results as the distance threshold corresponding to the first training sample. In some embodiments, the intelligent management center can also use the average of multiple candidate distances with good results as the corresponding distance threshold. Among them, the candidate distance with good results means that under the setting of the distance threshold, the execution robot does not suffer any work delays during the communication process, and no employees report that they did not receive the information in time.

[0120] In some embodiments of the present specification, the distance threshold can be determined efficiently and accurately through a distance threshold determination model, and a distance threshold with better effect than manual determination can be obtained. It can more intelligently determine whether an employee is at his or her workstation, which is beneficial to the subsequent determination of the communication route.

[0121] In some embodiments, the intelligent management center can update the communication route according to the scene data, wherein the communication route is determined based on the aforementioned preset algorithm.

[0122] In some embodiments, the intelligent management center can update the delivery route according to the number of people around the workstation of the target instruction object. For example, when the number of people around the workstation exceeds the number threshold, the delivery route can be updated to achieve the delayed delivery of the target instruction task currently being delivered.

[0123] In some embodiments, when the intelligent management center detects that an employee is not at his / her workstation or the workstation does not correspond to the employee, the intelligent management center can update the communication route to delay the communication of the target instruction task currently being communicated.

[0124] Exemplarily, the original communication route sequence is target instruction task 1 → target instruction task 2 → target instruction task 3. When target instruction task 2 meets the postponement condition, the intelligent management center can postpone the communication of target instruction task 2 in a variety of ways, so that the actual communication route sequence is target instruction task 1 → target instruction task 3 → target instruction task 2. For example, the way to achieve delayed communication includes: the intelligent management center can set the weight corresponding to the task type of target instruction task 2 to infinity, so that the communication sequence of target instruction task 2 in the communication route is the last. For another example, the intelligent management center can temporarily delete target instruction task 2 from the task list or store it in the waiting list of the execution robot. Among them, the postponement conditions can include that the number of people around the workstation exceeds the number threshold, the employee is not at the workstation, or the workstation does not correspond to the employee, etc.

[0125] Some embodiments of this specification take into account the impact of scenario data on the communication route, and can make more flexible adjustments to the communication route in combination with scenario data. For example, for a target instruction task with negative criticism, when there are many people around the employee, the communication of the instruction can be postponed to avoid other employees knowing the negative criticism instruction to the target instruction object. Recalculating the communication route based on changes in scenario data helps to update the communication route that best suits the current scenario based on the scenario data.

[0126] In some embodiments of this specification, based on the virtual distance and the task execution position corresponding to each target instruction task, the communication route is determined by the shortest route algorithm, which helps to consider the task type and task importance of the target instruction task during the communication process and determine a communication route that is more suitable for actual communication. For example, when the task importance of a certain target instruction task is high, the influence of the moving distance can be ignored and the target instruction task can be communicated first.

[0127] In some embodiments, the intelligent management center can calculate the ratio of the distance between each preset point in the transmission route and the speed of the execution robot to determine the arrival time of the execution robot at each preset point.

[0128] In some embodiments, the intelligent management center can determine the preset stay time by querying a preset table based on the task type of the target instruction task and the importance of the target instruction task. The preset table can include a correspondence between the task type of different target instruction tasks, the task importance and different preset stay times. For example, the correspondence can be: when the task type of the target instruction task is more complex and the task importance is higher, the preset stay time is longer.

[0129] Preset tables can be preset based on historical data. For more information on conveying time series, see Figure 1 and its related description.

[0130] In some embodiments, when the preset stay time corresponding to a preset point is longer, the intelligent management center can repeat the target instruction task communicated at the preset point to ensure that the employee clearly receives the target instruction task.

[0131] In some embodiments of the present specification, a communication route is determined by a preset algorithm, and a communication time sequence is determined based on the communication route and the task importance of the target instruction task. This helps to plan a command communication strategy that is more suitable for the current target instruction task based on the task type and task importance of the target instruction task while ensuring that the robot moves the shortest distance.

[0132] Figure 3 is an exemplary schematic diagram of applying a collision determination model according to some embodiments of the present specification.

[0133] In some embodiments, when the execution robot includes at least one, the intelligent management center can update the instruction communication strategy 340 of at least one execution robot through the collision determination model 320.

[0134] The collision determination model 320 may be a machine learning model, such as a graph neural network model (GNN), etc.

[0135] In some embodiments, the input of the collision determination model 320 includes the task graph 310, and the output includes the collision data 330 between robot nodes. Among them, is the edge output in the GNN corresponding to the collision data 330 between robot nodes.

[0136] The task map 310 refers to a map used to represent the relationship between various information between the execution robot and the employee corresponding to the instruction task. In some embodiments, the intelligent management center can construct a task map based on relevant information. The relevant information includes robot location data, employee location data, instruction communication strategy, etc.

[0137] In some embodiments, the nodes of the task graph 310 may include robot nodes and employee nodes.

[0138] The robot node corresponds to the execution robot. The node characteristics of the robot node can reflect the relevant characteristics of the corresponding execution robot. For example, the node characteristics of the robot node can include the task data of the instruction to be transmitted, the robot position data, and the instruction transmission strategy. The task data of the instruction to be transmitted can include the task type, instruction object, and associated tasks of the instruction task to be transmitted. The instruction task to be transmitted includes the target instruction task in the task list that has not been transmitted yet. For more information about the task list, target instruction task, task type, and associated tasks, please refer to Figure 2 Related description.

[0139] An employee node may correspond to an employee. The node features of an employee node may reflect the relevant features of the corresponding employee. For example, the employee node features may include the employee's historical location data, which refers to the employee's historical location data when moving, such as moving speed, moving range, etc. The employee's location data may affect the execution robot's delivery route, so that the actual delivery route may be different from the planned delivery route.

[0140] In some embodiments, the node features of the task map 310 also include route data. In some embodiments, the node features of the robot nodes in the task map 310 also include route data.

[0141] In some embodiments, the route data includes the wireless signal strength of each location point on the transmission route of the execution robot, whether it is a corner, and the degree of the turning angle.

[0142] In some embodiments of this specification, by introducing route data into node features, it is helpful to analyze the task map more comprehensively in combination with route data to reduce collisions between executing robots. For example, two executing robots may not detect each other at a corner, or there may be a delay at a location with low wireless signal strength. In this case, the intelligent management center can make timely adjustments to avoid collisions.

[0143] In some embodiments, the edges of the task graph 310 may include first-type edges and second-type edges. The first-type edges may be edges formed by connecting robot nodes to each other. The second-type edges may be edges formed by connecting robot nodes to employee nodes. The edge feature may reflect the distance between the corresponding two nodes.

[0144] The collision data includes the collision probability of the execution robot with other execution robots at each position point on the transmission route.

[0145] The collision determination model 320 can be obtained by training based on a second training sample with a second label. The second training sample can be a historical task map determined based on historical data, and the nodes and node features, edges and edge features of the historical task map are similar to the task map. The second label can be the collision probability between robot nodes actually corresponding to the second training sample.

[0146] In some embodiments, the second label can be obtained based on experiments. The experimental process includes: conducting experiments under the condition corresponding to the second training sample, and measuring the actual collision probability between each executing robot based on the distance between the executing robots. For example, the intelligent management center can conduct actual experiments under the condition of the historical task map, record the distance s between any two executing robots at each location point, and when s is greater than the distance threshold x, the collision probability is set to 0; when s is 0, the collision probability is 1; when s is between x and 0, It can be understood that the smaller the distance s between the executing robots is, the greater the collision probability is.

[0147] In some embodiments, the intelligent management center may update the instruction communication strategy 340 of at least one execution robot based on the output of the collision determination model 320 .

[0148] In some embodiments, when the collision data 330 between robot nodes is greater than a preset collision threshold, the intelligent management center can adjust the instruction communication strategy of the corresponding execution robot, adjust the task map based on the adjusted instruction communication strategy, and input the adjusted task map into the collision determination model 320 again. If the collision data 330 between robot nodes output this time are all less than the preset collision threshold, the adjusted instruction communication strategy is used as the final updated instruction communication strategy 340. The preset collision threshold can be determined based on prior knowledge or historical data.

[0149] In some embodiments, the manner of adjusting the instruction communication strategy may include modifying the communication route and modifying the communication time sequence.

[0150] In some embodiments, the modification of the communication route by the intelligent management center includes: controlling the execution robot to avoid the position points in the communication route where the collision probability is greater than the preset collision threshold during the communication process. In some embodiments, the modification of the communication route by the intelligent management center includes: modifying the communication order of each target instruction task in the communication route and regenerating the communication route. For example, the intelligent management center can exchange any two target instruction tasks with similar importance and regenerate the communication route. For more information about the importance of target instruction tasks, see Figure 2 and its related description.

[0151] In some embodiments, the modification of the communication time sequence by the intelligent management center includes: increasing or decreasing the speed of the execution robot, or increasing or decreasing the preset stay time of the execution robot at a certain task execution location, so that the execution robot arrives at the collision location point earlier or later.

[0152] In some embodiments, the modification of the communication route and the modification of the communication time sequence can be performed independently or simultaneously.

[0153] In some embodiments of the present specification, when determining the collision data 330 between robot nodes, the impact of employee movement and other execution robot positions on the robot communication route is taken into account, which can make the collision data more consistent with the actual situation and improve the accuracy of determining the collision data between robot nodes.

[0154] Some embodiments of the present specification provide an artificial intelligence-based service device, comprising at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute at least part of the computer instructions to implement the artificial intelligence-based service method as described in any of the above embodiments.

[0155] Some embodiments of the present specification provide a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed by a processor, the artificial intelligence-based service method described in any of the above embodiments is implemented.

[0156] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only for example and does not constitute a limitation of this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements and corrections to this specification. Such modifications, improvements and corrections are suggested in this specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of this specification.

[0157] At the same time, this specification uses specific words to describe the embodiments of this specification. For example, "some embodiments" means a certain feature, structure or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that "some embodiments" mentioned twice or more in different places in this specification do not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of this specification can be appropriately combined.

[0158] In addition, unless explicitly stated in the claims, the order of the processing elements and sequences described in this specification, the use of alphanumeric characters, or the use of other names are not intended to limit the order of the processes and laminar flow hoods in this specification. Although the above disclosure discusses some of the invention embodiments currently considered useful through various examples, it should be understood that such details are only for illustrative purposes, and the attached claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the essence and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0159] Similarly, it should be noted that in order to simplify the description disclosed in this specification and thus help understand one or more embodiments of the invention, in the above description of the embodiments of this specification, multiple features are sometimes combined into one embodiment, figure or description thereof. However, this disclosure method does not mean that the features required by the subject matter of this specification are more than the features mentioned in the claims. In fact, the features of the embodiments are less than all the features of the single embodiment disclosed above.

[0160] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A service method based on artificial intelligence, characterized in that: The method is executed by an intelligent management center, and the method includes: Receive instructions and tasks from management personnel; Based on the instruction object of the instruction task, determining the execution robot corresponding to the instruction task; Based on the instruction task and its instruction object, planning the instruction communication strategy of the execution robot, the instruction communication strategy at least includes a communication route and a communication time sequence, the communication time sequence includes a preset stay time of the execution robot at at least one preset point in the communication route; In response to receiving the communication completion notification from the execution robot, the corresponding instruction task is marked as completed.

2. The method according to claim 1, characterized in that The determining, based on the instruction object of the instruction task, the execution robot corresponding to the instruction task comprises: Determine the effective communication scope of the instruction object based on the personal data and activity scope of the instruction object; In response to at least one robot being within the effective communication range, one of the at least one robot is determined as the execution robot corresponding to the instruction task, and the instruction task is added to a task list of the execution robot.

3. The method according to claim 1, characterized in that The planning of the command transmission strategy of the execution robot based on the command task and its command object includes: Determine the task importance of the target instruction task based on the task type, instruction-related object, and related tasks of the target instruction task in the task list of the execution robot; Determining the transmission route through a preset algorithm based on the task importance of the target instruction task, the object position data of the instruction-associated object, and the robot position data of the executing robot; The communication time sequence is determined based on the communication route and the task importance of the target instruction task.

4. The method according to claim 3, characterized in that The execution robot includes at least one, and the planning of the instruction transmission strategy of the execution robot based on the instruction task and its instruction object also includes: The instruction communication strategy of the at least one execution robot is updated through a collision determination model, and the collision determination model is a machine learning model.

5. A service system based on artificial intelligence, characterized in that: The system includes a receiving module, a determining module, a planning module and a marking module: The receiving module is used to receive the instructions and tasks issued by the management personnel; A determination module, used to determine an execution robot corresponding to the instruction task based on the instruction object of the instruction task; A planning module, configured to plan a command communication strategy of the execution robot based on the command task and its command object, wherein the command communication strategy at least includes a communication route and a communication time sequence, wherein the communication time sequence includes a preset stay time of the execution robot at at least one preset point in the communication route; The marking module is used to mark the corresponding instruction task as completed in response to receiving a communication completion notification from the execution robot.

6. The artificial intelligence-based service system according to claim 5, wherein the determination module is further configured to: Determine the effective communication scope of the instruction object based on the personal data and activity scope of the instruction object; In response to at least one robot being within the effective communication range, one of the at least one robot is determined as the execution robot corresponding to the instruction task, and the instruction task is added to a task list of the execution robot.

7. The artificial intelligence-based service system according to claim 5, wherein the planning module is further used for: Determine the task importance of the target instruction task based on the task type, instruction-related object, and related tasks of the target instruction task in the task list of the execution robot; Determining the transmission route through a preset algorithm based on the task importance of the target instruction task, the object position data of the instruction-associated object, and the robot position data of the executing robot; The communication time sequence is determined based on the communication route and the task importance of the target instruction task.

8. The service system based on artificial intelligence according to claim 7, wherein the execution robot comprises at least one, and the planning module is further used for: The instruction communication strategy of the at least one execution robot is updated through a collision determination model, and the collision determination model is a machine learning model.

9. A service device based on artificial intelligence, the device comprising at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is used to execute at least part of the computer instructions to implement the artificial intelligence-based service method as described in any one of claims 1 to 4.

10. A computer-readable storage medium storing computer instructions, characterized in that: When the computer instructions are executed by a processor, an artificial intelligence-based service method as described in any one of claims 1 to 4 is implemented.