A method for robot operation task instantiation and related apparatus

Abstract

This invention relates to the field of robotics, specifically to a method and related apparatus for instantiating robot operation tasks. The invention first creates an ontology using preset knowledge of all possible operation tasks, all types of entities required to complete these tasks, and robots of those types. This ontology is a static, immutable database that retains entity and robot information required to complete any possible operation task. Then, based on a specific operation task, the invention filters corresponding preset knowledge from the ontology, adding corresponding attribute values, relationships, and temporal information to the filtered preset knowledge to form an instance that the robot can recognize. The instances in this invention originate from the ontology, and the preset knowledge within the ontology is combined according to the requirements of the operation task to obtain instances for any operation task, thereby enhancing the flexibility of the instances.

Description

Technical Field

[0001] This invention relates to the field of robotics, specifically to a method and related apparatus for instantiating robot operation tasks. Background Technology

[0002] When robots work in unfamiliar environments, it is necessary to pre-construct instances to guide their operation. The robot then performs its tasks according to the semantic information contained within those instances. Current technologies only set specific instances based on the specific scenario, thus reducing the robot's execution efficiency when working based on those instances. In other words, instances set only for specific scenarios lack flexibility, thereby reducing the robot's execution efficiency.

[0003] In summary, existing technologies reduce the flexibility of building instances.

[0004] Therefore, existing technologies still need to be improved and enhanced. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method and related apparatus for instantiating robot operation tasks, which solves the problem of reduced flexibility in constructing instances in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, the present invention provides a method for instantiating robot operation tasks, comprising:

[0008] Create an ontology to store all preset knowledge, which is used to characterize the robot and the entities operated by the robot, as well as the operational tasks performed by the robot through the operation of the entities.

[0009] The preset knowledge corresponding to the operation task is selected from the ontology and recorded as target knowledge;

[0010] Based on the operation task, the hierarchical relationship between each target knowledge and the attribute value of each target knowledge are determined. The hierarchical relationship is used to characterize the correlation and temporal information between each target knowledge.

[0011] Based on the hierarchical relationships between the target knowledge items and the attribute values ​​of each target knowledge item, an instance is constructed to express the semantics of the operation task.

[0012] In one implementation, the creation of an ontology for storing all preset knowledge, the preset knowledge being used to characterize the robot, entities operated by the robot, and operational tasks performed by the robot through operating the entities, includes:

[0013] Determine the scene information where the entity is located, the attribute information of the entity, and the component information of the robot in the preset knowledge;

[0014] An ontology is created based on the scene information, the attribute information of the entity, and the component information of the robot. The ontology does not include the hierarchical relationship between the scene information, the attribute information of the entity, and the component information of the robot.

[0015] In one implementation, the creation of an ontology for storing all preset knowledge, the preset knowledge being used to characterize the robot, entities operated by the robot, and operational tasks performed by the robot through operating the entities, includes:

[0016] Determine the scene information where the entity is located, the attribute information of the entity, and the component information of the robot in the preset knowledge;

[0017] Build a pre-defined task;

[0018] The preset task is decomposed into various preset sub-tasks;

[0019] Create preset actions corresponding to each of the preset subtasks, wherein the preset actions are actions corresponding to the components of the robot;

[0020] Based on the scene information, the entity's attribute information, the robot's component information, each preset subtask, and the preset actions corresponding to each preset subtask, an ontology is created.

[0021] In one implementation, the step of filtering the preset knowledge corresponding to the operation task from the ontology, denoted as target knowledge, includes:

[0022] Select tasks that match the operation task from the preset tasks, and denot them as matching tasks;

[0023] From the preset actions, select the preset actions from the target knowledge that correspond to each subtask of the matching task;

[0024] Based on the selected preset actions, the component information in the target knowledge corresponding to the matching task is selected from each component information;

[0025] Identify the robot in the target knowledge corresponding to the filtered component information;

[0026] Filter entities from the target knowledge that correspond to the matching task from the entities in the ontology;

[0027] Based on the selected entities, the scene information in which the entities are located is selected from each of the scene information.

[0028] In one implementation, the step of determining the hierarchical relationship between the target knowledge items and the attribute values ​​of each target knowledge item based on the operation task, wherein the hierarchical relationship is used to characterize the correlation and temporal information between the target knowledge items, includes:

[0029] Determine the timing information corresponding to each of the selected preset actions;

[0030] Determine the timing information corresponding to each subtask of the selected matching task;

[0031] Determine the association between the filtered entities and the filtered component information;

[0032] Determine the correlation between the selected preset actions, the selected entities, and the selected scene information.

[0033] In one implementation, constructing an instance to express the semantics of the operation task based on the hierarchical relationship between the target knowledge items and the attribute values ​​of each target knowledge item includes:

[0034] The arrows indicate the timing information corresponding to each preset action, the timing information corresponding to each subtask of the matched task, the correlation between the filtered entity and the filtered component information, and the correlation between the preset action, the filtered entity, and the filtered scene information, thus obtaining the template of the operation task.

[0035] Determine the scene region number, the entity model, and the execution parameter value of the preset action in the attribute values;

[0036] Based on the template of the operation task, the number of the scene area, the model of the entity, and the execution parameter value of the preset action, an instance is constructed to express the semantics of the operation task.

[0037] In one implementation, constructing an instance to express the semantics of the operation task based on the template of the operation task, the number of the scene area, the model of the entity, and the execution parameter values ​​of the preset action includes:

[0038] The scene area number is placed at the location of the scene area on the template, the entity model is placed at the location of the entity on the template, and the execution parameter value is placed at the location of the preset action on the template to obtain an instance that expresses the semantics of the operation task.

[0039] Secondly, embodiments of the present invention also provide a robot operation task instantiation apparatus, wherein the apparatus comprises the following components:

[0040] The ontology creation module is used to create an ontology for storing all preset knowledge, which is used to characterize the robot and the entities operated by the robot, as well as the operation tasks performed by the robot through the operation entities.

[0041] The knowledge filtering module is used to filter out the preset knowledge corresponding to the operation task from the ontology, and denoted as target knowledge;

[0042] The hierarchical attribute construction module is used to determine the hierarchical relationship between each target knowledge and the attribute value of each target knowledge according to the operation task. The hierarchical relationship is used to characterize the correlation and temporal information between each target knowledge.

[0043] The instance construction module is used to construct instances that express the semantics of the operation task based on the hierarchical relationship between the target knowledge and the attribute values ​​of each target knowledge.

[0044] Thirdly, embodiments of the present invention also provide a terminal device, wherein the terminal device includes a memory, a processor, and a robot operation task instantiation program stored in the memory and executable on the processor, wherein when the processor executes the robot operation task instantiation program, it implements the steps of the robot operation task instantiation method described above.

[0045] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a robot operation task instantiation program, wherein when the robot operation task instantiation program is executed by a processor, the steps of the robot operation task instantiation method described above are implemented.

[0046] Beneficial Effects: This invention first creates an ontology using pre-defined knowledge of all possible operational tasks, all types of entities required to complete these tasks, and robots of those types. In other words, the ontology is a static, immutable database that retains entity and robot information necessary to complete any possible operational task. Then, based on a specific operational task, this invention filters corresponding pre-defined knowledge from the ontology, adding corresponding attribute values, relationships, and temporal information to the filtered pre-defined knowledge to form instances that can be recognized by the robot. The instances of this invention originate from the ontology, and the pre-defined knowledge within the ontology is combined according to the requirements of the operational task to obtain instances for any operational task, thereby enhancing the flexibility of the instances. Attached Figure Description

[0047] Figure 1 This is an overall flowchart of the present invention;

[0048] Figure 2 This is a structural diagram of the main body in an embodiment of the present invention;

[0049] Figure 3 This is a template structure diagram in an embodiment of the present invention;

[0050] Figure 4 This is an example structural diagram from an embodiment of the present invention;

[0051] Figure 5 This is a block diagram illustrating the internal structure of a terminal device provided in an embodiment of the present invention. Detailed Implementation

[0052] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments and accompanying drawings. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0053] Research has found that when robots work in unfamiliar environments, it is necessary to pre-construct instances to guide their operation. The robot then performs its tasks according to the semantic information contained within those instances. Current technologies only set specific instances based on specific scenarios, thus reducing the efficiency of robots performing tasks based on those instances. In other words, instances set only according to specific scenarios lack flexibility, thereby reducing the robot's execution efficiency.

[0054] To address the aforementioned technical problems, this invention provides a method and related apparatus for instantiating robot operation tasks, solving the problem of reduced flexibility in constructing instances in existing technologies. Specifically, firstly, an ontology is created to store all preset knowledge (preset knowledge represents the robot, entities operated by the robot, and the operation tasks performed by the robot through these entities); then, preset knowledge corresponding to the operation task (denoted as target knowledge) is selected from the ontology; next, based on the operation task, the hierarchical relationships between target knowledge items and the attribute values ​​of each target knowledge item are determined; finally, based on the hierarchical relationships, temporal information, and attribute values ​​of each target knowledge item, an instance expressing the semantics of the operation task is constructed.

[0055] For example, the ontology (which is a database) pre-stores all types of operation tasks, such as operation task A, operation task B, and operation task C. The ontology also pre-stores entity 'a' and robot component 'aa' required to perform operation task A (the robot operation entity completes the operation task, and the robot component constitutes the various parts of the robot; operation tasks, entities, and robot components constitute pre-defined knowledge), entity 'b' and robot component 'bb' required to perform operation task B, and entity 'c' and robot component 'cc' required to perform operation task C. The ontology, as the first level, is used to express a set of abstract concepts and to record the correspondence between entities, operation tasks, and robot components.

[0056] When a robot needs to perform a certain operation task, the operation task is compared with all types of operation tasks in the ontology. If the operation task belongs to operation task A, then the pre-defined knowledge such as entity a and robot component aa is filtered out from the ontology, and an association is established between robot component aa and entity a (the association means that entity a is controlled by component aa, that is, there is a certain association between component aa and entity a), thus obtaining the template as the second level.

[0057] Adding specific attribute information to entity 'a' on the template creates an instance (third level) for a specific operation task. The robot can then perform specific operations on the entity based on the semantic information contained in that instance.

[0058] Exemplary methods

[0059] The robot operation task instantiation method of this embodiment can be applied to a terminal device, which can be a terminal product with data processing capabilities, such as a robot controller. In this embodiment, as... Figure 1 As shown, the robot operation task instantiation method specifically includes the following steps S100 to S400:

[0060] S100, Create an ontology for storing all preset knowledge, which is used to characterize the robot and the entities operated by the robot, as well as the operational tasks performed by the robot through the operating entities.

[0061] S200, the preset knowledge corresponding to the operation task is filtered out from the ontology and recorded as target knowledge.

[0062] S300, based on the operation task, determine the hierarchical relationship between each of the target knowledge and the attribute value of each target knowledge. The hierarchical relationship is used to characterize the correlation and temporal information between each of the target knowledge.

[0063] S400, based on the hierarchical relationship between the target knowledge and the attribute values ​​of each target knowledge, an instance for expressing the semantics of the operation task is constructed.

[0064] In one embodiment, step S100 is performed as follows: determining the scene information of the entity, the attribute information of the entity, and the component information of the robot in the preset knowledge; creating an ontology based on the scene information, the attribute information of the entity, and the component information of the robot, wherein the ontology does not contain the hierarchical relationship between the scene information, the attribute information of the entity, and the component information of the robot.

[0065] like Figure 2 As shown, an entity is essentially a static database that stores all six levels: scenarios, entities, agents, tasks, skills, and actions. It also stores the semantic knowledge of these six levels, the classification of action attributes, and the classification of entity attributes. For example... Figure 2 As shown, entity attributes are categorized into appearance, physical properties, semantics, and entity structure. The ontology level stores only six levels of knowledge and does not store associations or temporal relationships between layers.

[0066] In another embodiment, the specific process of step S100 is as follows: determine the scene information where the entity is located, the attribute information of the entity, and the component information of the robot in the preset knowledge; construct a preset task; decompose the preset task to obtain each preset sub-task; create preset actions corresponding to each preset sub-task, wherein the preset actions are actions corresponding to the components of the robot; and create an ontology based on the scene information, the attribute information of the entity, the component information of the robot, each preset sub-task, and the preset actions corresponding to each preset sub-task.

[0067] For example, there is a preset task A, which includes preset subtasks a1 and a2. Preset subtask a1 requires robot component b1 to perform action d1 on an entity in scene c1 to complete, and preset subtask a2 requires robot component b2 to perform action d2 on an entity in scene c2 to complete. Then, preset subtasks a1 and a2, scene c1, scene c2, component b1, component b2, action d1, and action d2 constitute the main body.

[0068] In one embodiment, step 200 includes the following steps S201 to S206:

[0069] S201, Select the task that matches the operation task from the preset tasks, and denot it as the matching task.

[0070] S202, from each of the preset actions, select the preset actions from the target knowledge corresponding to each subtask of the matching task.

[0071] S203, based on the preset actions selected, filter out the component information in the target knowledge corresponding to the matching task from each component information.

[0072] S204, determine the robot in the target knowledge corresponding to the filtered component information.

[0073] S205, Filter the entities in the target knowledge that correspond to the matching task from the various entities in the ontology.

[0074] S206, Based on the selected entities, the scene information where the entities are located is selected from each of the scene information.

[0075] Step S200 involves selecting the corresponding components, preset actions of the components, entities that the components operate on, and the scene in which the components and entities are located for the template of a certain task to be created later.

[0076] For example, if we need to create a task template for a robot to grasp an object (entity) on a conveyor belt (scene information) using its gripper (component), before creating the template, we first need to select the corresponding gripper component, the scene of the conveyor belt, and the object entity from the main body. Then, we need to select the actions that the gripper needs to perform to complete the sub-tasks corresponding to this task from the main body. (The sub-tasks can be, in order, locating the object, gripping the object, and moving the object to a specified position on the conveyor belt. The corresponding gripper actions are, in order, moving the gripper to the specified object, tightening the gripper, and moving the gripper.)

[0077] In one embodiment, step S400 includes the following steps S401, S402, and S403:

[0078] S401, the arrows are used to indicate the timing information corresponding to each preset action, the timing information corresponding to each subtask of the matched task, the correlation between the filtered entity and the filtered component information, and the correlation between the preset action, the filtered entity, and the filtered scene information, to obtain the template of the operation task.

[0079] like Figure 3 As shown, the matched task is a flexible flat cable assembly task. The robot uses a camera's RealSense camera to locate the material area where the flexible flat cable needs to be installed, and then diagonally moves the cable to that area. This positioning followed by diagonal movement constitutes the timing information for the robot's preset actions. Similarly, locating the cable and then diagonally moving it constitutes the timing information for the sub-task. There is a correlation between the cable positioning and the material area. Establishing the timing and correlation from the preset knowledge filtered from the robot body yields the following... Figure 3 The template shown is a template that includes templated knowledge at six levels: scene, entity, agent, task, skill, and action at the ontology level. It also adds associations or temporal relationships between different levels. Each type of operation task can be considered a template, such as the assembly of a flexible flat cable. The template level, as the second level, is dynamic and allows for operations such as adding and deleting templates.

[0080] S402, determine the scene area number, the entity model, and the execution parameter value of the preset action in the attribute value.

[0081] S403, the scene area number is placed at the location of the scene area on the template, the entity model is placed at the location of the entity on the template, and the execution parameter value is placed at the location of the preset action on the template, to obtain an instance for expressing the semantics of the operation task.

[0082] like Figure 4 As shown, number the scene areas according to the specific task. For example, add number -x to the material area, and add number such as -x to the flexible flat cable entity. Figure 4 The model shown adds execution parameter values ​​to the preset actions, such as adding displacement to the preset action of diagonal movement.

[0083] In summary, this invention first uses pre-defined knowledge—including all possible operational tasks, all types of entities required to complete these tasks, and robots of those types—to create an ontology. This ontology is a static, immutable database that retains entity and robot information necessary to complete any possible operational task. Then, based on a specific operational task, this invention filters corresponding pre-defined knowledge from the ontology, adding corresponding attribute values, relationships, and temporal information to the filtered knowledge to form instances that can be recognized by the robot. The instances of this invention originate from the ontology, and the pre-defined knowledge within the ontology is combined according to the requirements of the operational task to obtain instances for any operational task, thereby enhancing the flexibility of the instances.

[0084] Furthermore, this invention employs a multi-level robot skill operation knowledge representation method, which represents robot operation skills in a knowledge-based manner, decomposing them from top to bottom into three levels: Level 1, Level 2, and Level 3. Level 1 is the ontology level, used to represent a set of abstract concepts; Level 2 is the template level, used to represent a certain type of operation task; and Level 3 is the instance level, used to represent a specific operation task. This invention can be applied to complex robot skill application scenarios. By constructing knowledge representation from a three-level structure of ontology, template, and instance, it not only ensures the stability of the knowledge base architecture (ontology) but also guarantees the flexibility of the knowledge base (template).

[0085] Exemplary device

[0086] This embodiment also provides a robot operation task instantiation device, which includes the following components:

[0087] The ontology creation module is used to create an ontology for storing all preset knowledge, which is used to characterize the robot and the entities operated by the robot, as well as the operation tasks performed by the robot through the operation entities.

[0088] The knowledge filtering module is used to filter out the preset knowledge corresponding to the operation task from the ontology, and denoted as target knowledge;

[0089] The hierarchical attribute construction module is used to determine the hierarchical relationship between each target knowledge and the attribute value of each target knowledge according to the operation task. The hierarchical relationship is used to characterize the correlation and temporal information between each target knowledge.

[0090] The instance construction module is used to construct instances that express the semantics of the operation task based on the hierarchical relationship between the target knowledge and the attribute values ​​of each target knowledge.

[0091] Based on the above embodiments, the present invention also provides a terminal device, the principle block diagram of which can be as follows: Figure 5As shown, the terminal device includes a processor, memory, network interface, display screen, and temperature sensor connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for instantiating a robot operation task. The display screen can be an LCD screen or an e-ink screen. The temperature sensor is pre-installed inside the terminal device to detect the operating temperature of the internal components.

[0092] Those skilled in the art will understand that Figure 5 The block diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the terminal device to which the present invention is applied. The specific terminal device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0093] In one embodiment, a terminal device is provided, comprising a memory, a processor, and a robot operation task instantiation program stored in the memory and executable on the processor. When the processor executes the robot operation task instantiation program, it implements the following operation instructions:

[0094] Create an ontology to store all preset knowledge, which is used to characterize the robot and the entities operated by the robot, as well as the operational tasks performed by the robot through the operation of the entities.

[0095] The preset knowledge corresponding to the operation task is selected from the ontology and recorded as target knowledge;

[0096] Based on the operation task, the hierarchical relationship between each target knowledge and the attribute value of each target knowledge are determined. The hierarchical relationship is used to characterize the correlation and temporal information between each target knowledge.

[0097] Based on the hierarchical relationships between the target knowledge items and the attribute values ​​of each target knowledge item, an instance is constructed to express the semantics of the operation task.

[0098] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for instantiating robot operation tasks, characterized in that, include: Create an ontology to store all preset knowledge, which is used to characterize the robot and the entities operated by the robot, as well as the operational tasks performed by the robot through the operation of the entities. The preset knowledge corresponding to the operation task is selected from the ontology and recorded as target knowledge; Based on the operation task, the hierarchical relationship between each target knowledge and the attribute value of each target knowledge are determined. The hierarchical relationship is used to characterize the correlation and temporal information between each target knowledge. Based on the hierarchical relationship between the target knowledge and the attribute values ​​of each target knowledge, an instance is constructed to express the semantics of the operation task; The creation of an ontology for storing all preset knowledge, which characterizes the robot, the entities manipulated by the robot, and the operational tasks performed by the robot through these entities, includes: Determine the scene information where the entity is located, the attribute information of the entity, and the component information of the robot in the preset knowledge; Build a pre-defined task; The preset task is decomposed into various preset sub-tasks; Create preset actions corresponding to each of the preset subtasks, wherein the preset actions are actions corresponding to the components of the robot; Based on the scene information, the attribute information of the entity, the component information of the robot, each preset sub-task and the preset action corresponding to each preset sub-task, an ontology is created; The step of filtering the preset knowledge corresponding to the operation task from the ontology, denoted as target knowledge, includes: Select tasks that match the operation task from the preset tasks, and denot them as matching tasks; From the preset actions, select the preset actions from the target knowledge that correspond to each subtask of the matching task; Based on the selected preset actions, the component information in the target knowledge corresponding to the matching task is selected from each component information; Identify the robot in the target knowledge corresponding to the filtered component information; Filter entities from the target knowledge that correspond to the matching task from the entities in the ontology; Based on the selected entities, the scene information where the entities are located is selected from each of the scene information; Based on the operational task, the hierarchical relationship between the target knowledge items and the attribute values ​​of each target knowledge item are determined. The hierarchical relationship is used to characterize the correlation and temporal information between the target knowledge items, including: Determine the timing information corresponding to each of the selected preset actions; Determine the timing information corresponding to each subtask of the selected matching task; Determine the association between the filtered entities and the filtered component information; Determine the correlation between the selected preset actions, the selected entities, and the selected scene information.

2. The robot operation task instantiation method as described in claim 1, characterized in that, The creation of an ontology for storing all preset knowledge, which characterizes the robot, the entities manipulated by the robot, and the operational tasks performed by the robot through these entities, includes: Determine the scene information where the entity is located, the attribute information of the entity, and the component information of the robot in the preset knowledge; An ontology is created based on the scene information, the attribute information of the entity, and the component information of the robot. The ontology does not include the hierarchical relationship between the scene information, the attribute information of the entity, and the component information of the robot.

3. The robot operation task instantiation method as described in claim 1, characterized in that, The step of constructing an instance to express the semantics of the operation task based on the hierarchical relationship between the target knowledge items and the attribute values ​​of each target knowledge item includes: The arrows indicate the timing information corresponding to each preset action, the timing information corresponding to each subtask of the matched task, the correlation between the filtered entity and the filtered component information, and the correlation between the preset action, the filtered entity, and the filtered scene information, thus obtaining the template of the operation task. Determine the scene region number, the entity model, and the execution parameter value of the preset action in the attribute values; Based on the template of the operation task, the number of the scene area, the model of the entity, and the execution parameter value of the preset action, an instance is constructed to express the semantics of the operation task.

4. The robot operation task instantiation method as described in claim 3, characterized in that, The step of constructing an instance to express the semantics of the operation task based on the template of the operation task, the number of the scene area, the model of the entity, and the execution parameter value of the preset action includes: The scene area number is placed at the location of the scene area on the template, the entity model is placed at the location of the entity on the template, and the execution parameter value is placed at the location of the preset action on the template to obtain an instance that expresses the semantics of the operation task.

5. A robot operation task instantiation device, characterized in that, The device comprises the following components: The ontology creation module is used to create an ontology for storing all preset knowledge, which is used to characterize the robot and the entities operated by the robot, as well as the operation tasks performed by the robot through operating the entities. The knowledge filtering module is used to filter out the preset knowledge corresponding to the operation task from the ontology, and denoted as target knowledge; The hierarchical attribute construction module is used to determine the hierarchical relationship between each target knowledge and the attribute value of each target knowledge according to the operation task. The hierarchical relationship is used to characterize the correlation and temporal information between each target knowledge. An instance construction module is used to construct instances that express the semantics of the operation task based on the hierarchical relationship between the target knowledge and the attribute values ​​of each target knowledge. The creation of an ontology for storing all preset knowledge, which characterizes the robot, the entities manipulated by the robot, and the operational tasks performed by the robot through these entities, includes: Determine the scene information where the entity is located, the attribute information of the entity, and the component information of the robot in the preset knowledge; Build a pre-defined task; The preset task is decomposed into various preset sub-tasks; Create preset actions corresponding to each of the preset subtasks, wherein the preset actions are actions corresponding to the components of the robot; Based on the scene information, the attribute information of the entity, the component information of the robot, each preset sub-task and the preset action corresponding to each preset sub-task, an ontology is created; The step of filtering the preset knowledge corresponding to the operation task from the ontology, denoted as target knowledge, includes: Select tasks that match the operation task from the preset tasks, and denot them as matching tasks; From the preset actions, select the preset actions from the target knowledge that correspond to each subtask of the matching task; Based on the selected preset actions, the component information in the target knowledge corresponding to the matching task is selected from each component information; Identify the robot in the target knowledge corresponding to the filtered component information; Filter entities from the target knowledge that correspond to the matching task from the entities in the ontology; Based on the selected entities, the scene information where the entities are located is selected from each of the scene information; Based on the operational task, the hierarchical relationship between the target knowledge items and the attribute values ​​of each target knowledge item are determined. The hierarchical relationship is used to characterize the correlation and temporal information between the target knowledge items, including: Determine the timing information corresponding to each of the selected preset actions; Determine the timing information corresponding to each subtask of the selected matching task; Determine the association between the filtered entities and the filtered component information; Determine the correlation between the selected preset actions, the selected entities, and the selected scene information.

6. A terminal device, characterized in that, The terminal device includes a memory, a processor, and a robot operation task instantiation program stored in the memory and executable on the processor. When the processor executes the robot operation task instantiation program, it implements the steps of the robot operation task instantiation method as described in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a robot operation task instantiation program, which, when executed by a processor, implements the steps of the robot operation task instantiation method as described in any one of claims 1-4.

Images