Asynchronous planning method, device, equipment, medium and product for intelligent agent

Abstract

The embodiment of the present invention provides an asynchronous planning method for an intelligent agent, which can be applied to the field of artificial intelligence technology. The asynchronous planning method for an intelligent agent includes: executing an asynchronous planning process according to the current execution state of the intelligent agent; in response to the execution of the asynchronous planning process, obtaining a set of planned paths that match the expected execution state corresponding to the current execution state; and obtaining a valid planned path according to the set of planned paths. The embodiment of the present invention also provides an asynchronous planning device, equipment, storage medium and program product for an intelligent agent.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, specifically to the field of image processing technology, and more specifically to an asynchronous planning method, device, equipment, medium and product for an intelligent body. Background Art

[0002] Artificial Intelligence (AI) is an important driving force for the new round of scientific and technological revolution and industrial transformation. It is a new key technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. As an important part of intelligent science, artificial intelligence attempts to understand the essence of intelligence and produce a new intelligent machine (i.e., intelligent agent) that can respond in a similar way to human intelligence.

[0003] Existing intelligent agents usually use a parallel computing technology in the process of executing actions, which cannot reflect the advance planning based on the expected results of the current action. They just mechanically execute the current action and then plan the next action. They cannot well consider the real-time influences such as the current execution environment and the duration of the current action. As a result, there are significant differences between them and the real action execution reactions of humans, the execution of actions is not smooth enough, and the level of intelligence is low. Summary of the invention

[0004] In view of at least one of the above problems, the embodiments of the present invention are aimed at providing an asynchronous planning method, apparatus, device, medium and product for an intelligent agent that can perform asynchronous planning in advance according to the expected execution results, thereby providing a technical solution that can directly execute the optimal execution path planned in advance after the current action is completed, so as to save execution time and improve path execution efficiency, and enable smoother execution actions, which are more in line with the real action execution reactions of humans, thereby achieving a higher level of intelligence.

[0005] One aspect of an embodiment of the present invention provides an asynchronous planning method for an intelligent agent, which includes operations S201-S203:

[0006] In operation S201, an asynchronous planning process is executed according to the current execution state of the agent;

[0007] In operation S202, in response to the execution of the asynchronous planning process, a set of planned paths matching an expected execution state corresponding to the current execution state is obtained;

[0008] In operation S203, a valid planned path is acquired according to the planned path set.

[0009] According to an embodiment of the present invention, before executing the asynchronous planning process according to the current execution state of the agent, the method further includes: obtaining the current execution action of the agent; and obtaining the current execution state of the agent according to the current execution action.

[0010] According to one embodiment of the present invention, executing the asynchronous planning process according to the current execution state of the agent includes: executing the asynchronous planning process according to the current execution action of the agent and the estimated execution time of the current execution action.

[0011] According to one embodiment of the present invention, in executing an asynchronous planning process based on the current execution action of an intelligent agent and the estimated execution time of the current execution action, it includes: when the current execution action is in progress and the estimated execution time of the current execution action meets the preset time condition, executing the asynchronous planning process; wherein the preset time condition includes the model time condition of the current execution action defined by the preset time prediction model or the preset planning time condition for the main process of the intelligent agent to execute the current execution action.

[0012] According to one embodiment of the present invention, in response to the execution of the asynchronous planning process, before obtaining a set of planned paths that match the expected execution state corresponding to the current execution state, the method further includes: calling a completion function corresponding to the current execution state; and generating the expected execution state according to the completion function.

[0013] According to one embodiment of the present invention, in response to the execution of an asynchronous planning process, a set of planning paths that matches an expected execution state corresponding to a current execution state is obtained, including: generating at least one target planning path according to the expected execution state; performing a validity check on at least one target planning path to generate a set of planning paths.

[0014] According to an embodiment of the present invention, obtaining a valid planned path according to a planned path set includes: obtaining a real-time execution state of an intelligent agent; and obtaining a valid planned path matching the real-time execution state according to the planned path set.

[0015] According to one embodiment of the present invention, in obtaining a valid planned path that matches a real-time execution state based on a planned path set, it includes: performing verification on each planned path in the planned path set according to the real-time execution state to generate at least one verified planned path; and performing screening on each verified planned path of at least one verified planned path according to the real-time execution state to obtain a valid planned path.

[0016] According to an embodiment of the present invention, after acquiring the real-time execution status of the agent, the method further includes: interrupting the asynchronous planning process in response to the real-time execution status; and retrieving the asynchronous planning information of the asynchronous planning process after the interruption.

[0017] Another aspect of an embodiment of the present invention provides an asynchronous planning device for an intelligent agent, which includes an asynchronous execution module, a set acquisition module and a path acquisition module. The asynchronous execution module is used to execute an asynchronous planning process according to the current execution state of the intelligent agent; the set acquisition module is used to obtain a set of planned paths that match the expected execution state corresponding to the current execution state in response to the execution of the asynchronous planning process; and the path acquisition module is used to obtain a valid planned path according to the set of planned paths.

[0018] Another aspect of an embodiment of the present invention provides an electronic device, comprising one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute the above-mentioned asynchronous planning method of the intelligent agent.

[0019] Another aspect of an embodiment of the present invention provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to execute the above-mentioned asynchronous planning method for an intelligent agent.

[0020] Another aspect of an embodiment of the present invention provides a computer program product, including a computer program, which implements the above-mentioned asynchronous planning method of the intelligent agent when executed by a processor.

[0021] The asynchronous planning method of an intelligent agent provided in an embodiment of the present invention can at least partially solve the problem of low intelligence level in the intelligent agent action execution process in the related art, and thus can achieve at least one of the following technical effects:

[0022] The intelligent agent can make asynchronous plans in advance according to the expected execution results, so that it can directly execute the optimal execution path planned in advance after the current action is completed, in order to save execution time and improve path execution efficiency. It can have smoother execution actions and be more in line with real human action execution reactions, thus achieving a higher level of intelligence.

[0023] It should be understood that the above general description and the following detailed description are merely exemplary and illustrative and are not intended to limit the scope of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above contents and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0025] Figure 1 A schematic diagram showing an application scenario of an asynchronous planning method, apparatus, device, medium and program product for an intelligent agent according to an embodiment of the present invention;

[0026] Figure 2 Schematically shows a flow chart of an asynchronous planning method for an intelligent agent according to an embodiment of the present invention;

[0027] Figure 3 A flowchart of an application scenario of an asynchronous planning method for an intelligent agent according to an embodiment of the present invention is schematically shown;

[0028] Figure 4 A flowchart schematically shows another application scenario of the asynchronous planning method of an intelligent agent according to an embodiment of the present invention;

[0029] Figure 5 A schematic diagram showing a structural block diagram of an asynchronous planning device for an intelligent agent according to an embodiment of the present invention; and

[0030] Figure 6 A block diagram of an electronic device suitable for implementing an asynchronous planning method for an intelligent agent according to an embodiment of the present invention is schematically shown.

[0031] The above-mentioned drawings are part of the specification of the embodiments of the present invention, which illustrate exemplary embodiments of the present invention. The attached drawings and the description of the specification are used together to illustrate the principles of the embodiments of the present invention. It should be understood that the above general description of the drawings and the following specific implementations are only exemplary and illustrative, and they cannot limit the scope of the present invention. DETAILED DESCRIPTION

[0032] In order to make the objectives, technical solutions and advantages of the embodiments of the present invention more clearly understood, the spirit of the contents disclosed by the present invention will be clearly explained with the accompanying drawings and detailed descriptions below. After understanding the embodiments of the contents of the present invention, any technician in the relevant technical field can change and modify the techniques taught by the contents of the present invention without departing from the spirit and scope of the contents of the present invention.

[0033] The exemplary embodiments and descriptions of the present invention are used to explain the present invention, but are not intended to limit the present invention. In addition, elements / components with the same or similar reference numerals used in the drawings and embodiments are used to represent the same or similar parts.

[0034] The terms “first”, “second”, etc. used in the present invention do not particularly refer to an order or sequence, nor are they used to limit the present invention. They are only used to distinguish elements or operations described with the same technical terms.

[0035] The directional terms used in the present invention, such as up, down, left, right, front or back, etc., are only used to refer to the directions of the drawings. Therefore, the directional terms used are used to illustrate and not to limit the present invention.

[0036] The words “include,” “including,” “have,” “contain,” etc. used in the present invention are open-ended terms, meaning including but not limited to.

[0037] The term "and / or" used in the present invention includes any or all combinations of the items mentioned.

[0038] Regarding the present invention, "plurality" includes "two" and "more than two"; regarding the present invention, "plurality of groups" includes "two groups" and "more than two groups".

[0039] The terms "substantially" and "approximately" used in the present invention are used to modify any quantity or error that may vary slightly, but these slight changes or errors do not change their essence. Generally speaking, the range of slight changes or errors modified by such terms may be 20% in some embodiments, 10% in some embodiments, 5% in some embodiments, or other values. Those skilled in the art should understand that the aforementioned values ​​can be adjusted according to actual needs and are not limited thereto.

[0040] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0041] In the case of using expressions such as "at least one of A, B, and C, etc.", it should generally be interpreted in accordance with the meaning of the expression generally understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but not be limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or a system having A, B, C, etc.). In the case of using expressions such as "at least one of A, B, or C, etc.", it should generally be interpreted in accordance with the meaning of the expression generally understood by those skilled in the art (for example, "a system having at least one of A, B, or C" should include but not be limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or a system having A, B, C, etc.). Those skilled in the art should also understand that any transitional conjunctions and / or phrases that substantially represent two or more optional items, whether in the specification, claims, or drawings, should be understood to give the possibility of including one of these items, either of these items, or both of these items. For example, the phrase "A or B" should be understood to include the possibilities of "A" or "B", or "A and B".

[0042] Although the existing intelligent agent can process certain specific execution tasks, it usually executes the actions one by one in sequence according to the preset action trajectory of the execution task. Each action needs to be executed more coherently, otherwise it cannot achieve a good execution effect, and it is still impossible to "think" about the execution task and execute the action according to the expected result. To this end, another solution is proposed in the prior art, which enables the intelligent agent to autonomously identify its own actions according to the target execution task, and then control its own execution module to generate the current action to be executed, and then after executing the current action, continue to consider the next execution action until all execution actions are completed. Although this method seems to make the intelligent agent embody the function of "thinking", the intelligent agent usually adopts a parallel computing technology in the process of executing the action, which cannot reflect the advance planning according to the expected result of the current action. It just mechanically executes the current action and then plans the next action. It cannot well consider the real-time effects such as the current execution environment and the duration of the current action, which makes it significantly different from the real action execution reaction of humans, the execution action is not smooth enough, and the intelligence level is low.

[0043] In view of at least one of the technical problems existing in the above-mentioned prior art, the embodiments of the present invention aim to provide an asynchronous planning method, device, equipment, medium and product for an intelligent body that can perform asynchronous planning in advance according to the expected execution results, thereby providing a technical solution that can directly execute the optimal execution path planned in advance after the current action is completed, so as to save execution time and improve path execution efficiency, and enable smoother execution actions, which are more in line with the real action execution reactions of humans, thereby achieving a higher level of intelligence.

[0044] An aspect of an embodiment of the present invention provides an asynchronous planning method for an intelligent agent, which includes: executing an asynchronous planning process according to the current execution state of the intelligent agent; in response to the execution of the asynchronous planning process, obtaining a set of planned paths that match an expected execution state corresponding to the current execution state; and obtaining a valid planned path based on the planned path set.

[0045] Figure 1 An application scenario diagram of an asynchronous planning method, apparatus, device, medium, and program product for an intelligent agent according to an embodiment of the present invention is schematically shown.

[0046] like Figure 1 As shown, the application scenario 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0047] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only examples).

[0048] The terminal devices 101 , 102 , and 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0049] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the terminal devices 101, 102, and 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0050] It should be noted that the asynchronous planning method of the intelligent agent provided in the embodiment of the present invention can generally be executed by the server 105. Accordingly, the asynchronous planning device of the intelligent agent provided in the embodiment of the present invention can generally be set in the server 105. The asynchronous planning method of the intelligent agent provided in the embodiment of the present invention can also be executed by a server or a server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105. Correspondingly, the asynchronous planning device of the intelligent agent provided in the embodiment of the present invention can also be set in a server or a server cluster that is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and / or the server 105.

[0051] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0052] The following will be based on Figure 1 The scene described by Figure 2~Figure 4 The asynchronous planning method of the intelligent agent of the disclosed embodiment is described in detail.

[0053] like Figure 2 As shown, one aspect of an embodiment of the present invention provides an asynchronous planning method for an intelligent agent, which includes operations S201 to S203.

[0054] In operation S201, an asynchronous planning process is executed according to the current execution state of the agent;

[0055] In operation S202, in response to the execution of the asynchronous planning process, a set of planned paths matching an expected execution state corresponding to the current execution state is obtained; and

[0056] In operation S203, a valid planned path is acquired according to the planned path set.

[0057] Asynchronous planning: When the agent is performing an action, if the action takes a long time, it can perform asynchronous advance planning during this period. In this way, after the current action is completed, the pre-planned behavior path can be directly executed, saving time and improving the optimality of the path.

[0058] The intelligent agent can be the executor of the above-mentioned asynchronous planning method of the embodiment of the present invention, or it can be an executor controlled by the asynchronous planning method. Specifically, it can be a humanoid intelligent robot or other AI device, which usually has an actuator to complete specific action tasks. For example, a humanoid robot can use a mechanical manipulator to complete the action task of picking up an object, or even perform action tasks such as packaging products on an industrial production line. Among them, these action tasks can be used as the target execution tasks of the intelligent agent. In order to complete these target execution tasks, the intelligent agent needs to complete a series of execution actions in sequence.

[0059] The agent can detect its own environment and state information in real time, and confirm the progress of the current target execution task. Specifically, it can judge the spatial environment and its own state in each round of iteration. The current execution state can be at least one of the state information of the agent's own execution action and the state of its own environment detected at the current moment, wherein the state information of the agent's own execution action can include the name of the currently executed action and the execution information corresponding to the action name (such as the torque, angle of rotation, estimated execution time of the wrist, and the spatial position of each key point of the finger, etc.), and the state of the environment can include the information of the surrounding people and objects in the space where the agent is currently located, such as the size, position (distance) and even shape, color, etc. of obstacles on the route, and can also be the dynamic change information of the people in the surrounding environment, such as the speed of the people's movement, the emotional information on the face, etc.

[0060] In the embodiment of the present invention, the agent generally uses the main process to implement the target execution task, which is the core process for the agent to implement action execution. The asynchronous planning process is a key process in the embodiment of the present invention. It is mainly used to implement the planning of subsequent execution actions of the current execution action (i.e. asynchronous planning) when the agent executes the current execution action, and can be used as an auxiliary process of the main process.

[0061] When the asynchronous planning process is executed, it is equivalent to starting the control flow of asynchronous planning. Specifically, the subsequent execution actions of the current execution action can be planned according to the target execution result of the target execution task of the current execution state. Among them, the expected execution state is the state information of the self-execution action and the state of the environment that the intelligent agent matching the above target execution result needs to achieve at the next moment, which can be the execution state corresponding to the next execution action after the current execution action is completed. For example, the target execution task is to complete the "product packaging work", and the current execution action corresponding to the current execution state can be "taking the packaging carton board", and the corresponding next execution action can be "folding the packaging carton board into a packaging box", wherein the current execution action and the corresponding next execution action can have corresponding sub-actions, for example, "taking the packaging carton board" can include "turning around-walking to the picking position-bending-extending arms-opening palms-touching the designated position of the target packaging carton board-closing fingers-straightening waist-turning around-walking to the packaging position" and a series of sub-actions. Among them, the current execution action can also be any of the above sub-actions, and correspondingly, the next execution action can be the next action corresponding to the sub-action, which will not be described in detail.

[0062] The next execution action must be an action that matches the expected execution state. It can be a sub-action or an action set of sub-actions (i.e., a continuous action trajectory or action path). There can be different execution modes for the execution of the same execution action. For example, the sub-action "turn around" can be "clockwise turn" or "counterclockwise turn". For the execution action of the execution sequence of "bend over - extend arms - open palms", there can be different execution sequences, such as "extend arms - bend over - open palms", "open palms - extend arms - bend over" and other different execution sequence combinations. For the next execution state planned by the current execution state, the corresponding next execution action can correspond to different execution modes. The combination of actions or action sets of these different execution modes can be used as the planning path set. The planning path set is the set of all action execution mode combinations that exist for the next execution action (relative to the current execution action) corresponding to the expected execution state. That is, through the above asynchronous planning process, it can be realized that when the main process executes the current execution action, all sets of different execution modes of the next execution action that match the expected execution state after the current execution state ends can be obtained, thereby preliminarily completing the asynchronous planning.

[0063] Based on the above-mentioned set of planning paths, it is also necessary to further screen the set of different execution modes of the above-mentioned next execution action to obtain a valid planning path. The valid planning path is at least one execution mode action combination in the above-mentioned set of planning paths, and the execution actions included therein can constitute an action execution path. Among them, when the current execution action is completed, the main process will immediately execute the execution action specified by the valid planning path provided by the asynchronous planning process.

[0064] Therefore, the asynchronous planning method of the embodiment of the present invention aims to perform asynchronous planning in advance according to the expected execution results, thereby providing a technical solution that can directly execute the optimal execution path planned in advance after the current action is completed, in order to save execution time and improve path execution efficiency, so that the intelligent body can have more coherent, fluent and smooth execution actions, which are more in line with the real action execution reactions of humans, thereby achieving a higher level of intelligence.

[0065] In order to enable those skilled in the art to have a clearer understanding of the asynchronous planning method of the above-mentioned intelligent agent in the embodiment of the present invention, the following is further provided: Figure 3-Figure 4 Description.

[0066] in, Figure 3 A flowchart of an application scenario of an asynchronous planning method for an intelligent agent according to an embodiment of the present invention is schematically shown, referring to operations S301-309; Figure 4 A flowchart of another application scenario of the asynchronous planning method of an intelligent agent according to an embodiment of the present invention is schematically shown, referring to S301-S303, S307-309 and S401-S403.

[0067] like Figure 2-Figure 4 As shown, according to one embodiment of the present invention, before operation S201 executes the asynchronous planning process according to the current execution state of the agent, the method further includes:

[0068] Get the agent's current execution action;

[0069] Get the current execution state of the agent based on the current execution action.

[0070] like Figure 3 and Figure 4As shown in operation S301, in the main process A, each iteration of the agent will determine the current state of the agent. Specifically, the current action of the agent at the current moment can be detected first, and the state information of the agent's own action at the current moment can be determined based on the current action, which can specifically include the torque, angle of rotation, expected execution time of the wrist, and the spatial position of each key point of the finger. At the same time, when the current action is obtained, the agent's image sensor (such as a camera) can be used to detect the surrounding environment information that the agent can currently "see", such as whether there are obstacles, the location, size, shape, and movement speed of the obstacles.

[0071] Therefore, a comprehensive detection of the current execution state corresponding to the current execution action at the current moment can be achieved, so that a more accurate expected execution state can be generated in combination with the expected result of the target execution task. For example, after encountering an obstacle, the current execution action must be terminated or changed to an action to remove the obstacle. This can make the action execution of the intelligent agent more intelligent, and the actions are more closely aligned with the operating actions of real humans, making the action execution smoother and less prone to frustration.

[0072] like Figure 2-Figure 4 As shown, according to one embodiment of the present invention, in operation S201, executing the asynchronous planning process according to the current execution state of the agent includes:

[0073] Perform an asynchronous planning process based on the agent's current action and the estimated execution time of the current action.

[0074] like Figure 3 and Figure 4 As shown in operation S301, each execution action requires specific events from the start of execution to the completion of execution (excluding the case of execution termination), which can be as small as milliseconds (such as the blink of an eye in a sub-action) or as large as more than ten minutes (such as the action path "product packaging"). However, the execution time of the same execution action in different execution environments may also be different. Therefore, it is necessary to estimate the events required for the current execution action to finally complete the execution at the current moment based on the execution environment in which the current execution action is located and the execution time usually required for the execution action. The time estimated through actual application can be understood as the estimated execution time.

[0075] The asynchronous planning method of the embodiment of the present invention, in order to improve quality and efficiency, while avoiding the impact of the asynchronous planning process B on the execution process of the main process and reducing unnecessary resource occupation, needs to detect that the intelligent agent has a current execution action and at the same time meet the estimated execution time of the current execution action to be able to execute the asynchronous planning process, so as to prevent the situation where the asynchronous planning process has just been started and the current execution action has been completed or even has been completed long ago.

[0076] The asynchronous planning process B may have multiple processes relative to the main process A at the same time, so as to perform asynchronous planning on multiple current execution actions of the main process at the same time, such as Figure 3 and Figure 4 The single asynchronous planning process B shown is only used as an example. Usually, the asynchronous planning process B is started to perform asynchronous planning for the current execution action only when the current execution action is detected and the estimated execution time of the current execution action meets the asynchronous planning. Moreover, when the current execution action has a corresponding asynchronous planning process B, no new asynchronous planning process will be started for it.

[0077] The start of the asynchronous planning process B can be achieved through two Python function libraries, Asyncio+Joblib. Asyncio can achieve asynchronous effects, while Joblib can implement parallel computing processes, so that the asynchronous process can generate as many behavior paths as possible. Among them, Joblib can serialize more different types of parameter data when starting parallel computing functions; Asyncio can directly access class member variables without requiring such variables to be shared variables.

[0078] Therefore, when the agent is executing the current action, if the current action takes a long time, it can perform asynchronous advance planning during this period. In this way, after the current action is completed, the pre-planned behavior path provided by the asynchronous planning process can be directly executed, saving time and improving the optimality of the next action path.

[0079] like Figure 2-Figure 4 As shown, according to one embodiment of the present invention, in executing an asynchronous planning process according to a current execution action of an agent and an estimated execution time of the current execution action, the process includes:

[0080] When the current execution action is in progress and the estimated execution time of the current execution action meets the preset time conditions, the asynchronous planning process is executed; wherein the preset time conditions include the model time conditions of the current execution action defined by the preset time prediction model or the preset planning time conditions for the main process of the intelligent agent to execute the current execution action.

[0081] like Figure 3 and Figure 4 In operation S301, as mentioned above, the premise for starting the asynchronous planning process B is that the current execution action needs to be in the ongoing state, and must not be in the suspended state or even the terminated state. In addition, the estimated execution time needs to meet a specific preset time condition, that is, the estimated execution time of the current execution action according to the current execution environment.

[0082] The preset time condition may be a specific time threshold, which may be generated by a preset time prediction model trained by the execution action and its corresponding execution time data (model time condition), or may be a statistical calculation duration corresponding to the execution action when the agent executes the main process A, i.e., a preset planning time condition. When the estimated execution time is greater than or equal to the time threshold defined by the model time adjustment, the execution of the asynchronous planning process may be started.

[0083] The time threshold corresponding to the preset planning time condition can be determined by statistical data calculation. For example, the execution process of each execution action A is recorded, and the duration from the start to the end of the execution is formed to form a set of execution times for multiple executions of the execution action A. In this way, a sliding average time value T can be calculated:

[0084] T = (the last sliding average T' + the execution time of this action) / 2.

[0085] At the same time, the sliding average time S of the normal planning of the main process to execute action A is also counted:

[0086] S = (the last sliding average S' + this estimated execution time) / 2,

[0087] Among them, when the preset planning time condition can satisfy T>S*C (C is an empirical scalar), it is considered that the action duration is sufficient for advance asynchronous planning, otherwise it is considered that the action is completed quickly and the idle time is insufficient to start the asynchronous planning process B.

[0088] Therefore, considering that the current execution action takes some time to complete, if the current execution action is completed in a very short time, there is no need to start the asynchronous planning process. The main process only needs to continue planning and executing according to the execution status of the current execution action. For details, see Figure 3 and Figure 4 Operations S307-S309 are shown. Only when the execution cycle of the current execution action can meet the parallel asynchronous planning time of the asynchronous planning process, the asynchronous planning can be started to realize the advance planning of the next action corresponding to the current execution action, thereby achieving the effect of improving the execution efficiency and reducing the execution frustration.

[0089] like Figure 2-Figure 4 As shown, according to an embodiment of the present invention, before obtaining a set of planned paths matching an expected execution state corresponding to a current execution state in response to the execution of the asynchronous planning process in operation S202, the method further includes:

[0090] Call the completion function corresponding to the current execution state;

[0091] Generates the expected execution state based on the completion function.

[0092] like Figure 3 and Figure 4 In operation S302, in the asynchronous planning process B, first, the corresponding completion function is called according to the current execution state corresponding to all the currently executed actions in the main process A. Since the completion function can correspond to the expected execution state matching the current execution state, the changes brought about by the execution of the currently executed action to the agent and its matching environment can be effectively defined.

[0093] The expected execution state can define the execution goal or execution result that is expected to be achieved when the current execution action is executed, that is, it matches the expected target execution task. According to the expected execution state (which can be defined as state S_new), a more accurate planning basis can be provided for the planning of the next execution action after the current execution action is completed.

[0094] like Figure 2-Figure 4 As shown, according to an embodiment of the present invention, in operation S202, in response to the execution of the asynchronous planning process, a set of planned paths matching the expected execution state corresponding to the current execution state is obtained, including:

[0095] generating at least one target planning path according to the expected execution state;

[0096] A validity check is performed on at least one target planned path to generate a set of planned paths.

[0097] like Figure 3 and Figure 4As mentioned above, the operations S301-S302 shown in the figure can be used as the planning basis for the next execution action according to the expected execution state (S_new). For example, the target execution task can be "packing products", and the current execution action corresponding to the current moment in "packing products" can be "taking products" in the action set of "checking products (before packaging)", and the expected execution result corresponding to the expected execution state can be "completing the appearance inspection of the six sides of the product". When the agent is executing the current execution action "taking products", it is assumed that the "appearance inspection of the two sides of the product" has been completed at the same time. According to the above expected execution result, the asynchronous planning process B can plan the target execution action corresponding to the current execution action "completing the appearance inspection of the remaining 4 sides of the product" for the main process A in advance. Among them, the completion process of the target execution action can include the action trajectory composed of the combination of different execution actions such as "checking A side-B side-C side-D side in sequence", "checking B side-A side-C side-D side in sequence", and "checking D side-B side-C side-A side in sequence", that is, the target planning path. Each target planning path includes some corresponding sub-action combinations that can be executed continuously, such as "check surface D-surface B-surface C-surface A in sequence".

[0098] The asynchronous planning process B can perform validity checks on these target planning paths in combination with the current execution state and the expected execution state during each round of planning cycles, so as to extract the target planning paths that have passed the valid check into the corresponding data storage module (such as the results list) that can be accessed or called by the main process, and generate a set of planning paths based on the data storage module. The so-called validity check is to judge whether these target planning paths can be directly executed in the next step after the current execution action is completed according to the current action execution state and the expected action execution state of the main process A. For example, after the current execution action is completed, the intelligent agent cannot complete a certain rotation action because its own manipulator cannot complete the target planning path "checking D surface-B surface-C surface-A surface in sequence" in this case. At this time, the target planning path is an action path that cannot be executed. The target planning paths in the planning path set are all action paths that can be effectively executed.

[0099] Therefore, invalid planned paths can be effectively excluded, and it can be ensured that the planned paths in the planned path set can be effectively executed after the current execution action is completed.

[0100] like Figure 2-Figure 4 As shown, according to an embodiment of the present invention, in operation S203, obtaining a valid planned path according to a planned path set includes:

[0101] Get the real-time execution status of the agent;

[0102] Obtain a valid planned path that matches the real-time execution status based on the planned path set.

[0103] like Figure 3 Operations S303-S305 and S307 are shown and Figure 4 The real-time execution state of the agent is the latest state of the main process A obtained by re-detecting the execution status of the current execution action and its corresponding execution state after or during the asynchronous planning process B executes the asynchronous planning to generate the planning path set, such as whether the current execution action has been completed, the latest execution progress, the state change information of the execution process, etc., which can be referred to Figure 3 Operation S307 is shown.

[0104] Among them, the latest execution status can determine whether the intelligent agent can execute each target planning path in the planning path set at this time. For example, as mentioned above, in the process of the intelligent agent executing the current execution action "picking up the product", although the "appearance inspection of both sides of the product" can be completed at the same time, the wrist rotation angle is too large or too small due to an accident at a later moment, making it impossible for one or even all of the target planning paths in the planning path set to be executed smoothly as expected. Therefore, in this case, it is necessary to screen each target planning path in the planning path set according to the current real-time execution status to determine the effective planning path. Among them, the effective planning path can eventually be called by the main process, and can directly take over the execution of the action path after the current execution action is completed. In other words, if the current execution action has been completed, the main process A can directly call the effective planning path generated by the asynchronous planning process B for direct execution, such as Figure 3 If the effective planning path cannot be obtained, it is necessary to re-plan through the main process A, such as Figure 3 Operations S306-S308 are shown.

[0105] Therefore, for the planning path set composed of a sufficient number of target planning paths provided by asynchronous planning, the effect of further screening the optimal execution path can be achieved, so that the planning path set obtained by asynchronous planning can be further screened to obtain at least one better target execution path as a valid planning path.

[0106] like Figure 2-Figure 4 As shown, according to an embodiment of the present invention, obtaining a valid planned path matching the real-time execution state according to the planned path set includes:

[0107] Perform verification on each planned path of the planned path set according to the real-time execution state to generate at least one verified planned path;

[0108] Screening is performed on each verification plan path of at least one verification plan path according to the real-time execution state to obtain a valid plan path.

[0109] like Figure 3 Operations S304-S305 and Figure 4 As shown in operations S401-S402, in the asynchronous planning process B, each planned path in the planned path set can be verified according to the execution information and environmental information of the intelligent agent in the real-time execution state. Among them, the verification can be a simulated execution of the planned path according to the execution information and environmental information of the intelligent agent in the above-mentioned real-time state. When it can be simulated and executed in the real-time execution state, the verification passes, otherwise the verification fails. The planned path that passes the verification is the verified planned path. At the same time, the verification process can also evaluate the simulated execution process of each planned path in the planned path set according to the simulated execution. For example, some verified planned paths can be executed faster, more accurately and more easily, and their evaluation values ​​can be higher, so that these verified planned paths that pass the verification can be screened according to the evaluation values. The verified planned path with the highest evaluation value is used as the valid planned path.

[0110] Among them, if the verification evaluation value of the valid planning path is high enough, it can be directly called by the main process A, and there is no need to continue planning. After the current execution action is completed, the action execution is directly implemented, such as Figure 3 S306-S309 and Figure 4 If the verification evaluation value of the effective planning path is not high enough, it is necessary to continue planning, and the main process A will continue the main process planning to execute the next execution action after the current execution action is completed. Figure 3 S306-S308 and Figure 4 S403-S308 shown.

[0111] Therefore, through the above-mentioned path verification method, the planned path set can be further simulated and verified according to the real real-time execution status of the current execution action, and finally a specific effective planning path with a sufficiently high evaluation value can be refined. Therefore, the effective planning path can more accurately match the next execution action of the current execution action, thereby effectively ensuring the overall action execution fluency of the intelligent body, further improving the fit between the execution process and the real human action, making it more realistic and natural, and more intelligent.

[0112] like Figure 2-Figure 4 As shown, according to one embodiment of the present invention, after obtaining the real-time execution state of the agent, the method further includes:

[0113] Responsive to the real-time execution status, interrupting the asynchronous planning process;

[0114] Retrieve the asynchronous planning information of the asynchronous planning process after the interruption.

[0115] like Figure 3 and Figure 4 As shown in operation S307, when the real-time execution status is that the current execution action has been completed, it is necessary to force the asynchronous planning process to be interrupted, and at the same time, all planning information (i.e., asynchronous planning information) including the planning path set generated by the asynchronous planning process B before the interruption is retrieved, usually called to the main process A, or it can be called by the main process A.

[0116] Among them, the asynchronous planning process B can start a loop, and each round will generate a candidate behavior path. At the beginning of each round, a state indicator variable will be judged. The variable has three states: -1 (asynchronous planning has exited), 0 (asynchronous planning has not been started), and 1 (asynchronous planning is in progress). If the variable value is found to be 0 in the loop judgment, the loop will be broken directly (i.e. interrupted); after the loop ends, the variable will be set to -1 at the end of the asynchronous planning process, indicating that the process is about to exit. When the main process A needs to end the asynchronous planning process B, it can set the variable from 1 to 0, and then wait for the asynchronous planning process B to set the variable to -1. When the variable value is detected to be -1, the main process A knows that the asynchronous planning process B has ended, and it will collect all the planning paths obtained by the asynchronous planning, and reset the corresponding variable value to 0.

[0117] In one embodiment of the present invention, when the asynchronous planning process B is executed normally and is not interrupted, Figure 3 As shown, the operations of path verification (corresponding to operation S304), obtaining a valid path (corresponding to operation S305), and whether to re-plan (corresponding to operation S306) can be performed in the asynchronous planning process B. Correspondingly, in another embodiment of the present invention, Figure 4 As shown, when the current execution action is actually completed and the asynchronous planning process B has not yet been completed, the asynchronous planning process B will be forced to end, such as Figure 4 As shown, the operations of path verification (corresponding to operation S401), obtaining valid paths (corresponding to operation S402) and whether to re-plan (corresponding to operation S403) can be executed in the main process A. In other words, when the asynchronous planning process B is suddenly interrupted, the main process A can simulate and verify the planning paths in the obtained real-time execution status planning path set (such as the results list) one by one.

[0118] Therefore, even if the actual completion time of the currently executed action cannot be predicted, the forced interruption method can be used as a judgment bar. In the real-time execution state, as long as the currently executed action is found to be completed, the asynchronous planning process can be forced to stop, and the results of the existing asynchronous planning can be transferred to the main process, thereby avoiding the situation where a long time is needed to wait for the asynchronous planning process to end naturally.

[0119] In combination with the asynchronous planning method of the intelligent agent in the above-mentioned embodiment of the present invention, it can be realized that when the intelligent agent performs a current action (such as walking over to pick up something), the intelligent agent can start asynchronous planning, and plan the subsequent action path (such as putting the thing somewhere) in advance according to its own state after getting the thing (such as the thing in hand). At the same time, when the current action is completed, if the path planned asynchronously in advance is found to be feasible after verification in the real-time state, the path can be executed directly without the need for planning in the main process; if it is found that one of the action paths planned in advance is not feasible after verification (for example, the thing is not actually successfully taken, so the putting action cannot be performed), then another path planned in advance (such as continuing to complete the taking of things) is selected for verification and execution. When the execution results, continue to perform asynchronous planning in advance (such as performing the next step of advance planning based on the expectation of being able to put things down), and so on.

[0120] In summary, the asynchronous planning method of an intelligent agent provided in an embodiment of the present invention can at least partially solve the problem of low intelligence level in the intelligent agent action execution process in the related art, and thus can achieve at least one of the following technical effects:

[0121] The intelligent agent can make asynchronous plans in advance according to the expected execution results, so that it can directly execute the optimal execution path planned in advance after the current action is completed, in order to save execution time and improve path execution efficiency. It can have smoother execution actions and be more in line with real human action execution reactions, thus achieving a higher level of intelligence.

[0122] Based on the above-mentioned asynchronous planning method of an intelligent agent, the present invention also provides an asynchronous planning device for an intelligent agent. Figure 5 The device is described in detail.

[0123] Figure 5 The structure block diagram of the asynchronous planning device of an intelligent agent according to an embodiment of the present invention is schematically shown.

[0124] like Figure 5 As shown, the asynchronous planning device 500 of the intelligent agent in this embodiment includes an asynchronous execution module 510, a set acquisition module 520 and a path acquisition module 530.

[0125] The asynchronous execution module 510 is used to execute the asynchronous planning process according to the current execution state of the agent. In one embodiment, the asynchronous execution module 510 can be used to execute the operation S201 described above, which will not be repeated here.

[0126] The set acquisition module 520 is used to obtain a set of planned paths that match the expected execution state corresponding to the current execution state in response to the execution of the asynchronous planning process. In one embodiment, the set acquisition module 520 can be used to perform the operation S202 described above, which will not be repeated here.

[0127] The path acquisition module 530 is used to acquire a valid planned path according to the planned path set. In one embodiment, the path acquisition module 530 can be used to perform the operation S203 described above, which will not be described in detail here.

[0128] According to an embodiment of the present invention, any multiple modules among the asynchronous execution module 510, the set acquisition module 520 and the path acquisition module 530 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present invention, at least one of the asynchronous execution module 510, the set acquisition module 520 and the path acquisition module 530 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware and firmware or in any appropriate combination of any of them. Alternatively, at least one of the asynchronous execution module 510, the set acquisition module 520 and the path acquisition module 530 can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be executed.

[0129] Figure 6 A block diagram of an electronic device suitable for implementing an asynchronous planning method for an intelligent agent according to an embodiment of the present invention is schematically shown.

[0130] The electronic device provided by an embodiment of the present invention includes one or more processors and a memory, the memory being used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute the asynchronous planning method of the above-mentioned intelligent agent.

[0131] like Figure 6As shown, the electronic device 600 according to an embodiment of the present invention includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 to a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include an onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0132] In RAM 603, various programs and data required for the operation of electronic device 600 are stored. Processor 601, ROM 602 and RAM 603 are connected to each other via bus 604. Processor 601 performs various operations of the method flow according to the embodiment of the present invention by executing the program in ROM 602 and / or RAM 603. It should be noted that the program can also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 can also perform various operations of the method flow according to the embodiment of the present invention by executing the program stored in the one or more memories.

[0133] According to an embodiment of the present invention, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the I / O interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0134] The present invention also provides a computer-readable storage medium on which executable instructions are stored. When the instructions are executed by a processor, the processor executes the asynchronous planning method of the above-mentioned intelligent agent.

[0135] The computer-readable storage medium may be included in the device / apparatus / system described in the above embodiment; or it may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present invention is implemented.

[0136] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present invention, the computer-readable storage medium may include the ROM 602 and / or RAM 603 described above and / or one or more memories other than ROM 602 and RAM 603.

[0137] An embodiment of the present invention also includes a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the asynchronous planning method of the above-mentioned intelligent agent.

[0138] The computer program includes program codes for executing the method shown in the flow chart. When the computer program product is run in a computer system, the program codes are used to enable the computer system to implement the method provided by the embodiment of the present invention.

[0139] The computer program executes the above functions defined in the system / device of the embodiment of the present invention when it is executed by the processor 601. According to the embodiment of the present invention, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0140] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 609, and / or installed from a removable medium 611. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0141] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, the above functions defined in the system of the embodiment of the present invention are performed. According to the embodiment of the present invention, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0142] According to an embodiment of the present invention, the program code for executing the computer program provided by the embodiment of the present invention can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0143] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0144] It will be appreciated by those skilled in the art that the features described in the various embodiments and / or claims of the present invention may be combined and / or combined in various ways, even if such combinations and / or combinations are not explicitly described in the present invention. In particular, the features described in the various embodiments and / or claims of the present invention may be combined and / or combined in various ways without departing from the spirit and teachings of the present invention. All of these combinations and / or combinations fall within the scope of the present invention.

[0145] The embodiments of the present invention are described above. However, these embodiments are only for the purpose of illustration, and are not intended to limit the scope of the present invention. Although each embodiment is described above, it does not mean that the measures in each embodiment cannot be used in combination. The scope of the present invention is defined by the attached claims and their equivalents. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. An asynchronous planning method for an intelligent agent, characterized in that: The agent comprises a robot having an actuator, and the method comprises: executing an asynchronous planning process according to the current execution state of the agent; In response to the execution of the asynchronous planning process, obtaining a set of planned paths that match an expected execution state corresponding to a current execution state; and Acquire a valid planned path according to the planned path set; Wherein, before executing the asynchronous planning process according to the current execution state of the agent, the method further includes: Get the current execution action of the agent; Acquire the current execution state of the agent according to the current execution action; Wherein, in the process of executing the asynchronous planning according to the current execution state of the agent, it includes: executing an asynchronous planning process according to a current execution action of the agent and an estimated execution time of the current execution action; Wherein, in the asynchronous planning process according to the current execution action of the agent and the estimated execution time of the current execution action, it includes: When the currently executing action is in progress and the estimated execution time of the currently executing action meets the preset time condition, the asynchronous planning process is executed; wherein the preset time condition includes the model time condition of the currently executing action defined by the preset time prediction model or the preset planning time condition for the main process of the intelligent agent to execute the currently executing action.

2. The method according to claim 1, characterized in that: Before acquiring a set of planned paths matching an expected execution state corresponding to a current execution state in response to the execution of the asynchronous planning process, the method further includes: Calling the completion function corresponding to the current execution state; The expected execution state is generated according to the completion function.

3. The method according to claim 1, characterized in that In the step of obtaining, in response to the execution of the asynchronous planning process, a set of planned paths matching an expected execution state corresponding to a current execution state, comprising: generating at least one target planning path according to the expected execution state; A validity check is performed on at least one target planned path to generate the planned path set.

4. The method according to claim 1, characterized in that In the acquiring a valid planned path according to the planned path set, comprising: Obtaining the real-time execution status of the agent; The effective planned path matching the real-time execution state is acquired according to the planned path set.

5. The method according to claim 4, characterized in that In acquiring the effective planned path matching the real-time execution state according to the planned path set, comprising: Performing verification on each planned path of the planned path set according to the real-time execution state to generate at least one verified planned path; Screening is performed on each verification plan path of the at least one verification plan path according to the real-time execution state to obtain the valid plan path.

6. The method according to claim 4, characterized in that After obtaining the real-time execution state of the agent, the method further includes: In response to the real-time execution status, interrupting the asynchronous planning process; Retrieve the asynchronous planning information of the asynchronous planning process after the interruption.

7. An asynchronous planning device for an intelligent agent for implementing the method according to any one of claims 1 to 6, characterized in that: include: an asynchronous execution module, for executing an asynchronous planning process according to a current execution state of the agent; A set acquisition module, configured to acquire, in response to the execution of the asynchronous planning process, a set of planned paths matching an expected execution state corresponding to a current execution state; as well as The path acquisition module is used to acquire a valid planned path according to the planned path set.

8. An electronic device comprising: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to execute the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images