Robot skill operation method based on parameter expression knowledge base, terminal and medium

Abstract

The application discloses a robot skill operation method based on a parameter expression knowledge base, a terminal and a medium, wherein the method comprises the following steps: acquiring a first knowledge base of a robot, and determining a target entity and associated entity information thereof according to the first knowledge base; acquiring a second knowledge base of the robot, and determining the relationship between the target entity and other entities in a current scene according to the second knowledge base and the entity information; acquiring a third knowledge base of the robot, and searching for corresponding skills and action primitives according to the third knowledge base and the relationship; and controlling corresponding execution mechanisms and elements to execute a target task in the current scene according to the searched skills and action primitives. The application constructs a robot knowledge base, so that the robot can effectively access and search for required skill operation knowledge from the knowledge base in the actual use process, thereby improving the efficiency of the robot in acquiring skills and executing tasks in a strange scene.

Description

Technical Field

[0001] The present invention relates to the field of robots, and in particular to a robot skill operation method, terminal and medium based on a parameter expression knowledge base. Background Art

[0002] Currently, robots are widely used in industries such as industry, agriculture, healthcare, and the military. Key research areas in robotics include robot manipulation and motion planning. As society continues to evolve, the environments and scenarios robots face become increasingly complex. Technological advancements are placing ever-stricter demands on robot efficiency.

[0003] To accomplish goal-oriented object manipulation tasks in complex environments and scenarios, robots need to memorize a large amount of target object attributes, such as their shape, weight, and characteristics (e.g., fragile, hot, soft, or wet). To perfectly execute tasks and achieve goals according to appropriate motion requirements, robots must master various motion parameters and target attributes.

[0004] In existing robot platforms, facing complex environments and scenarios, robots need to continuously learn in order to obtain the motion parameters and attribute information of the target tasks corresponding to the robot; moreover, after the robot learns, the skills and action expression mechanisms obtained can only adapt to the robot and cannot help other robots to achieve mutual learning and knowledge sharing; this requires other robots to learn the motion parameters and attribute information of the target tasks corresponding to them, which is not conducive to human-machine collaboration and the robot's rapid execution of target tasks.

[0005] Therefore, the existing technology needs to be improved. Summary of the Invention

[0006] The technical problem to be solved by the present invention is that, in response to the defects of the existing technology, the present invention provides a robot skill operation method, terminal and medium based on a parameter expression knowledge base, aiming to solve the technical problem in the existing technology of low efficiency of robots in acquiring skills and performing tasks in unfamiliar scenarios.

[0007] The technical solutions adopted by the present invention to solve the technical problems are as follows:

[0008] In a first aspect, the present invention provides a robot skill operation method based on a parameter expression knowledge base, the robot skill operation method based on the parameter expression knowledge base comprising the following steps:

[0009] Obtaining a first knowledge base of the robot, and determining a target entity and its associated entity information based on the first knowledge base;

[0010] Obtaining a second knowledge base of the robot, and determining, based on the second knowledge base and the entity information, a relationship between the target entity and other entities in a current scenario;

[0011] Obtaining a third knowledge base of the robot, and searching for corresponding skills and action primitives based on the third knowledge base and the relationship;

[0012] Based on the skills and action primitives found, the corresponding actuators and components are controlled to perform the target tasks in the current scenario.

[0013] In one implementation, the acquiring of the first knowledge base of the robot and determining the target entity and its associated entity information based on the first knowledge base may include:

[0014] Pre-setting the first knowledge base, the second knowledge base, and the third knowledge base according to the node information;

[0015] The node information includes node name, node ID, node attributes, node labels and node relationships.

[0016] In one implementation, obtaining a first knowledge base of the robot and determining a target entity and its associated entity information based on the first knowledge base includes:

[0017] Obtaining task information of the robot;

[0018] Acquire a first knowledge base of the robot according to the task information;

[0019] The target entity and its associated entity information are determined according to the first knowledge base; wherein the entity information includes: entity type, entity attributes and entity associated file information.

[0020] In one implementation, obtaining a second knowledge base of the robot and determining, based on the second knowledge base and the entity information, a relationship between the target entity and other entities in the current scenario includes:

[0021] Acquire a second knowledge base of the robot according to the target entity;

[0022] Determine the scenario in which the target entity is located, and obtain, based on the scenario, a set of relationships between entities corresponding to the scenario in the second knowledge base;

[0023] Searching the relationship between the target entity and other entities in the relationship set according to entity classification;

[0024] The relationships include: positional relationships, logical relationships, and active and passive relationships.

[0025] In one implementation, searching the relationship between the target entity and other entities in the relationship set according to the entity classification includes:

[0026] The relationship between the target component and other components, operating platforms and actuators is searched in the relationship set according to the entity classification.

[0027] In one implementation, obtaining a third knowledge base of the robot and searching for corresponding skills and action primitives based on the third knowledge base and the relationship includes:

[0028] Acquire a third knowledge base of the robot according to the relationship;

[0029] Find the skill tree corresponding to the relationship in the current task and scenario;

[0030] The skills and action primitives required for the current task are determined according to the skill tree.

[0031] In one implementation, searching for a skill tree corresponding to the relationship in the current task and scenario includes:

[0032] Build skill trees corresponding to multiple tasks and scenarios through the arrangement and combination of different skills and different action primitives.

[0033] In one implementation, controlling corresponding actuators and components based on the found skills and action primitives includes:

[0034] Determining corresponding execution parameters based on the found skills and action primitives;

[0035] The corresponding actuators and components are controlled according to the execution parameters.

[0036] In a second aspect, the present invention provides a terminal comprising: a processor and a memory, wherein the memory stores a robot skill operation program based on a parameter expression knowledge base, and the robot skill operation program based on the parameter expression knowledge base is used to implement the robot skill operation method based on the parameter expression knowledge base as described in the first aspect when executed by the processor.

[0037] In a third aspect, the present invention provides a medium, which is a computer-readable storage medium, and which stores a robot skill operation program based on a parameter expression knowledge base. When the robot skill operation program based on the parameter expression knowledge base is executed by a processor, it is used to implement the robot skill operation method based on the parameter expression knowledge base as described in the first aspect.

[0038] The present invention adopts the above technical solution to achieve the following effects:

[0039] The present invention constructs a robot knowledge base based on node parameters, so that the robot can effectively access and search for required skill operation knowledge from the knowledge base during actual use, thereby improving the efficiency of the robot in acquiring skills and performing tasks in unfamiliar scenarios; moreover, the knowledge base constructed in the present invention can facilitate skill editing and skill modification, as well as facilitate the verification of the consistency and completeness of the skill knowledge in the knowledge base, thereby effectively and accurately defining the robot's operating skills in the expression of node parameters, making the target entities and their logical relationships required for the robot to perform target tasks in unfamiliar scenarios more rigorous, thereby ensuring the robot's task execution efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0041] Figure 1 It is a flow chart of a robot skill operation method based on a parameter expression knowledge base in one implementation of the present invention.

[0042] Figure 2 It is a flow chart of robot knowledge base expression in one implementation of the present invention.

[0043] Figure 3 It is a schematic diagram of a robot target library in one implementation of the present invention.

[0044] Figure 4 It is a schematic diagram of a robot scene library in one implementation of the present invention.

[0045] Figure 5 It is a schematic diagram of constructing a robot skill library in one implementation of the present invention.

[0046] Figure 6 It is a schematic diagram of node information at each level in a screw assembly scenario in one implementation of the present invention.

[0047] Figure 7 It is a schematic diagram of the node relationship at each level in a screw assembly scenario in one implementation of the present invention.

[0048] Figure 8 It is an overall schematic diagram of the knowledge base after being exported in a screw assembly scenario in one implementation of the present invention.

[0049] Figure 9 It is a functional principle diagram of a terminal in one implementation of the present invention.

[0050] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of the present invention more clear and distinct, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0052] Exemplary Methods

[0053] like Figure 1 As shown, an embodiment of the present invention provides a robot skill operation method based on a parameter expression knowledge base, and the robot skill operation method based on the parameter expression knowledge base includes the following steps:

[0054] Step S100: Acquire a first knowledge base of the robot, and determine a target entity and its associated entity information based on the first knowledge base.

[0055] In this embodiment, the robot skill operation method based on the parameter expression knowledge base is applied to a terminal, and the terminal includes but is not limited to: a robot, or a terminal device for controlling a robot, such as a smart TV, a mobile phone, and a tablet computer.

[0056] In this embodiment, in the robot skill operation method based on the parameter expression knowledge base, the parameter expression knowledge base is based on the method to facilitate the robot to quickly acquire the required skill operation knowledge according to the target task in an unfamiliar scene, thereby improving the efficiency of the robot in acquiring skills and performing tasks in unfamiliar scenes; this embodiment determines the target entity and the relationship between the target entity and other entities from the robot skill knowledge base according to the target scene and target task that the robot needs to complete, and then obtains the skill and action primitives based on the determined relationship, which can realize the robot's motion planning for complex target scenes, thereby enhancing the robot's adaptability to complex scenes, so that the robot can quickly perform target tasks in complex scenes.

[0057] Specifically, if Figure 2 As shown, when the robot operation skills are expressed in the form of knowledge, they are divided from the upper layer to the lower layer into: a first knowledge base 100, a second knowledge base 200 and a third knowledge base 300.

[0058] Among them, the first knowledge base 100 is a target library, which can be used to express the target entity of the robot during the skill operation process, such as: the operating platform, actuators and related components required by the robot during the skill operation process; various entities and their associated entity information can be stored through the first knowledge base 100, so that the robot can select the corresponding target entity and its associated entity information according to the target task in the application scenario.

[0059] The second knowledge base 200 is a scene library under the first knowledge base 100, which can be used to express the relationship between entities of the robot in different scenes, such as: the relationship between components and operating platforms, the relationship between actuators and components, etc.; different scene information can be stored through the second knowledge base 200, so that the robot can select the corresponding entity relationship information according to the determined target entity in the current scene.

[0060] The third knowledge base 300 is a skill base under the second knowledge base 200, which can be used to express different skill information under the current scene and entity relationship, for example, skill 301: shape recognition, skill 302: tool selection, etc.; different skills and action primitive information can be stored through the third knowledge base 300, so that the robot can select the required skills and action primitives according to the determined target entities and relationships to perform the target tasks.

[0061] In an application scenario, in order to ensure that the robot can obtain the required knowledge base during the skill operation process, before implementing the robot skill operation method based on the parameter expression knowledge base of this embodiment, it is necessary to set the first knowledge base 100, the second knowledge base 200 and the third knowledge base 300 according to the node information, that is, the basic units of the first knowledge base 100, the second knowledge base 200 and the third knowledge base 300 are all nodes.

[0062] Specifically, in this embodiment, in each node of the knowledge base, the node information includes the node name, node ID, node attributes, node labels, and node relationships; wherein the node attributes can describe the characteristics of the node and the relationship, including but not limited to the following forms:

[0063] Concept attributes, such as: Material: stainless steel; Color: black;

[0064] Numeric attributes, for example: diameter: 10mm; specification: 10mm*15mm*15mm;

[0065] Execution parameters, for example: screwing depth: 8mm; maximum force: 50N; screwing position: [x, y, z];

[0066] Environmental attributes, for example: production cycle: 5Days.

[0067] That is, in one implementation of this embodiment, the following steps are included before step S100:

[0068] Step S001: pre-set the first knowledge base, the second knowledge base, and the third knowledge base according to node information.

[0069] In this embodiment, after setting up the required knowledge base, when implementing the robot skill operation method based on the parameter expression knowledge base, it is necessary to first clarify the target task of the robot in the current scenario, so as to obtain the target entity and its associated entity information according to the determined target task.

[0070] Specifically, when determining the target task of the robot, the scene information and task information input by the staff can be obtained, and the target task in the current scene can be determined based on the obtained scene information and task information; alternatively, the current scene information of the robot can be obtained by shooting with a camera, for example: the current scene information of the robot is a mobile phone parts assembly site; then, the target task information of the robot is obtained by analysis based on the obtained current scene information.

[0071] Furthermore, after obtaining the target task information of the robot in the current scenario, the first knowledge base of the robot can be obtained based on the task information, and then the target entity and its associated entity information can be determined based on the first knowledge base; wherein, the entity information includes: entity type, entity attributes and entity associated file information.

[0072] like Figure 3 As shown, in this embodiment, the acquired first knowledge base includes multiple entities, such as: operating platform, actuator, component 1, component 2...component N; assuming that the target entity is determined to be component 1 based on the acquired task information, the entities associated with component 1 are operating platform, actuator, component N, etc.

[0073] In the above-mentioned target entities, the operating platform represents the working platform that the robot needs to use when performing the target task, the actuator represents the actuator that the robot needs to use when performing the target task; the component represents the component that the robot needs to use when performing the target task, including: component 1, component 2...component N; for example: in the nut assembly scenario, the components can be nuts, screws, bolts and other components; by obtaining the target entity and related entities from the first knowledge base, these entities can be used to implement the skill operation process required for the task when the robot performs the target task.

[0074] That is, in one implementation of this embodiment, step S100 specifically includes the following steps:

[0075] Step S101, obtaining the task information of the robot;

[0076] Step S102, obtaining a first knowledge base of the robot according to the task information;

[0077] Step S103: determining the target entity and its associated entity information according to the first knowledge base.

[0078] This embodiment can obtain a first knowledge base of the parameter expression knowledge base according to the robot's task information, so that the robot can find the corresponding target entity in the first knowledge base, and thus implement the corresponding skill operation through the target entity to perform the target task.

[0079] like Figure 1 As shown, in one implementation of the embodiment of the present invention, the robot skill operation method based on the parameter expression knowledge base further includes the following steps:

[0080] Step S200: Acquire a second knowledge base of the robot, and determine the relationship between the target entity and other entities in the current scenario based on the second knowledge base and the entity information.

[0081] In this embodiment, after obtaining the target entity and its associated entity information from the first knowledge base, it is also necessary to obtain the robot's second knowledge base based on the first knowledge base, so as to find the relationship of the target entity in the current scenario in the second knowledge base.

[0082] like Figure 4 As shown, the second knowledge base is a knowledge base established based on the first knowledge base, that is, the robot's scenario library is a knowledge base established based on the target library; multiple application scenarios are set in the second knowledge base, and a scenario is composed of related relationships in the first knowledge base, for example: the relationship between related components, actuators and operating platforms; the scenario library is used to record and store the relationship between entities in different scenarios, which is a collection of entity-relationship-entity, and provides a classification search function for entities.

[0083] Furthermore, after obtaining the second knowledge base, first use the previously input scene information or the scene information obtained by shooting analysis to determine the scene where the target entity is located; then, in this scene, obtain the corresponding relationship set between entities; finally, use the classification search function provided in the second knowledge base to search for the relationship between the target entity and other entities in the relationship set, so as to subsequently determine the required skills and action primitives based on the searched relationships.

[0084] It is worth mentioning that in the above relationships, the relationship between the target entity and other entities can be a positional relationship, a logical relationship, an active and passive relationship, etc.; it represents a certain connection between each entity and is not limited to the special relationship of a single entity.

[0085] That is, in one implementation of this embodiment, step S200 specifically includes the following steps:

[0086] Step S201, obtaining a second knowledge base of the robot according to the target entity;

[0087] Step S202, determining the scenario in which the target entity is located, and obtaining a set of relationships between entities corresponding to the scenario in the second knowledge base according to the scenario;

[0088] Step S203 : searching the relationship between the target entity and other entities in the relationship set according to entity classification.

[0089] In one implementation of this embodiment, taking into account the differences in entity types, when searching for entity relationships, a diffusion search is performed with the target entity as the center; specifically, taking the target element as an example, the classification of the target entity can be determined first, and then the relationship between the target element and other elements, operating platforms, and actuators can be searched in the relationship set based on the entity classification; thereby, the intrinsic connection between the target element and other entities is determined based on the searched relationship.

[0090] That is, in one implementation of this embodiment, step S203 specifically includes the following steps:

[0091] Step S203a: searching the relationship between the target component and other components, operating platforms, and actuators in the relationship set according to the entity classification.

[0092] This embodiment can obtain a scenario library based on the target library, and then search the scenario library for the relationship between the target component and other components, operating platforms and actuators in a certain scenario, so as to determine the intrinsic connection between the target component and other entities based on the searched relationship.

[0093] like Figure 1 As shown, in one implementation of the embodiment of the present invention, the robot skill operation method based on the parameter expression knowledge base further includes the following steps:

[0094] Step S300: Acquire the third knowledge base of the robot, and search for corresponding skills and action primitives based on the third knowledge base and the relationship.

[0095] In this embodiment, after obtaining the relationship between the target entity and other related entities, the third knowledge base is obtained based on the second knowledge base, that is, the skill base is obtained based on the scenario base; then, in the skill base, the skill tree corresponding to the above relationship under the current task and scenario can be searched, and the skills and action primitives required for the current task can be determined based on the skill tree; wherein, the skill tree includes skill 1, skill 2...skill N, and each skill includes action primitive 1, action primitive 2...action primitive N; the skill tree also includes a set of relationships between skills and each action primitive.

[0096] Specifically, the third knowledge base records the skills, action sequences, and the correspondence between skills and action sequences under different tasks; and in the third knowledge base, a task-customized skill tree can be implemented, and a skill tree that meets the needs can be built by adding and deleting skills, and by adding and deleting action primitives.

[0097] In addition, the third knowledge base provides a reasoning query function; for example, a search query can be implemented in a structured database; a task-customized skill tree can be implemented by combining different task skills with action primitives; or a task-customized skill tree can be implemented by adjusting the order of task skills and action primitives.

[0098] That is, in one implementation of this embodiment, step S300 specifically includes the following steps:

[0099] Step S301, obtaining a third knowledge base of the robot according to the relationship;

[0100] Step S302, searching for a skill tree corresponding to the relationship in the current task and scenario;

[0101] Step S303: determining the skills and action primitives required for the current task according to the skill tree.

[0102] In one implementation of this embodiment, before searching for the skill tree, the robot can construct skill trees corresponding to multiple tasks and scenarios through the permutations and combinations of different skills and different action primitives; that is, the existing skills and / or newly added skills are combined in a permutation and combination manner to obtain different skill trees; and the existing skills and / or newly added action primitives are combined in a permutation and combination manner to obtain different skill trees; the specific permutation and combination methods are not described in detail in this embodiment.

[0103] like Figure 5 As shown, Figure 5The skill library construction flow chart shows the implementation of a skill library for storing skills. The skill library primarily consists of task-skill-action primitives. The combination and order of skills and action primitives vary for different tasks. The skill library interface allows for customizable skill trees for tasks, which can be expanded and deleted. The skill library is first persistently stored in a structured database. The hierarchical relationship between task skills and action primitives is then constructed in a graph database to facilitate reasoning queries.

[0104] That is, in one implementation of this embodiment, step S302 specifically includes the following steps:

[0105] Step S302a: Construct skill trees corresponding to multiple tasks and scenarios by permuting and combining different skills and different action primitives.

[0106] This embodiment searches for a skill tree corresponding to the target entity relationship based on the skill library, thereby obtaining the skills and action primitives required to perform the target task in the skill tree, so that the robot can implement the skill operation process of the target task based on the acquired skills and action primitives.

[0107] like Figure 1 As shown, in one implementation of the embodiment of the present invention, the robot skill operation method based on the parameter expression knowledge base further includes the following steps:

[0108] Step S400: Control corresponding actuators and components based on the skills and action primitives found to execute the target task in the current scenario.

[0109] In this embodiment, after obtaining the required skills and action primitives, the corresponding execution parameters are determined based on the found skills and action primitives; since the action primitives are driven by the action primitive module, and the action primitive module is implemented by the relevant parameters of the robot itself, these relevant parameters are closely related to the functional configuration of the robot itself. Therefore, according to the node information of the skills and action primitives, the required execution parameters can be determined to control the corresponding actuators and components according to the execution parameters; among them, the node information and execution parameters can be obtained by table lookup.

[0110] Step S401, determining corresponding execution parameters according to the found skills and action primitives;

[0111] Step S402: controlling corresponding actuators and components according to the execution parameters.

[0112] In one implementation of this embodiment, when setting the nodes of each knowledge base, the nodes in each robot knowledge base contain relationships such as start, next, end, has task, etc. At the same time, optimization and supplementation can be performed based on specific nodes such as objects and tools to enrich the content of the knowledge base; among them, the supplementation method can be a manual import method or a robot self-learning import method.

[0113] To facilitate understanding of the present embodiment, Figure 6 、 Figure 7 as well as Figure 8 The following example illustrates the target and scenario of "screw assembly" under "mobile phone parts assembly":

[0114] like Figure 6 As shown, through the node information of the knowledge base, the robot scenario obtained from node 100 and node 200 is: "mobile phone parts assembly", the target task is: "screw assembly", and the target entity obtained through the first knowledge base is the components, actuators and operating platform corresponding to the screw assembly.

[0115] like Figure 7 and Figure 8 As shown, the relationship between the entity nodes is obtained through the second knowledge base, and the required skills and action primitives are obtained through the third knowledge base.

[0116] For example, skill 300: size recognition, skill 301: shape recognition, skill 302: tool selection; action 400: recognition (deep learning size), action 401: recognition (deep learning shape), action 402: recognition (deep learning tool).

[0117] Finally, by controlling the corresponding components to execute action primitives in sequence, the corresponding skills can be realized; and by implementing different skill operation processes in sequence, the target tasks can be completed based on the target entity in specific relationships and scenarios.

[0118] This embodiment constructs a robot knowledge base based on node parameters, so that the robot can effectively access and search for required skill operation knowledge from the knowledge base during actual use, thereby improving the efficiency of the robot in acquiring skills and performing tasks in unfamiliar scenarios; moreover, the knowledge base constructed in this embodiment can facilitate skill editing and skill modification, as well as facilitate the verification of the consistency and completeness of the skill knowledge in the knowledge base, thereby effectively and accurately defining the robot's operating skills in the expression of node parameters, making the target entities and their logical relationships required for the robot to perform target tasks in unfamiliar scenarios more rigorous, thereby ensuring the robot's task execution efficiency.

[0119] Exemplary devices

[0120] Based on the above embodiment, the present invention further provides a terminal, whose principle block diagram can be shown as follows: Figure 9 shown.

[0121] The terminal includes: a processor, memory, interface, display screen and communication module connected through a system bus; wherein the processor of the terminal is used to provide computing and control capabilities; the memory of the terminal includes a medium and an internal memory; the medium is a computer-readable storage medium, which stores an operating system and a computer program; the internal memory provides an environment for the operation of the operating system and computer program in the medium; the interface is used to connect to external terminal devices, such as mobile terminals and computers; the display screen is used to display corresponding robot skill operation information based on a parameter expression knowledge base; the communication module is used to communicate with a cloud server or a mobile terminal.

[0122] When the computer program is executed by a processor, it is used to implement a robot skill operation method based on a parameter expression knowledge base.

[0123] It will be understood by those skilled in the art that Figure 9 The principle block diagram shown in the figure is only a block diagram of a partial structure related to the solution of the present invention, and does not constitute a limitation on the terminal to which the solution of the present invention is applied. The specific terminal may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0124] In one embodiment, a terminal is provided, which includes: a processor and a memory, wherein the memory stores a robot skill operation program based on a parameter expression knowledge base, and the robot skill operation program based on the parameter expression knowledge base is used to implement the above-mentioned robot skill operation method based on the parameter expression knowledge base when executed by the processor.

[0125] In one embodiment, a medium is provided, wherein the medium stores a robot skill operation program based on a parameter expression knowledge base, and the robot skill operation program based on the parameter expression knowledge base is used to implement the above-mentioned robot skill operation method based on the parameter expression knowledge base when executed by a processor.

[0126] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The computer program can be stored in a non-volatile medium. When executed, the computer program can include the processes in the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory.

[0127] In summary, the present invention provides a robot skill operation method, terminal and medium based on a parameter expression knowledge base, wherein the method includes: obtaining a first knowledge base of the robot, and determining a target entity and its associated entity information based on the first knowledge base; obtaining a second knowledge base of the robot, and determining the relationship between the target entity and other entities in the current scenario based on the second knowledge base and the entity information; obtaining a third knowledge base of the robot, and searching for corresponding skills and action primitives based on the third knowledge base and the relationship; and controlling corresponding actuators and components based on the found skills and action primitives to perform the target task in the current scenario. The present invention constructs a robot knowledge base so that the robot can effectively access and search for required skill operation knowledge from the knowledge base during actual use, thereby improving the efficiency of the robot in acquiring skills and performing tasks in unfamiliar scenarios.

[0128] It should be understood that the application of the present invention is not limited to the above examples. For those skilled in the art, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A robot skill operation method based on a parameter expression knowledge base, characterized in that: The robot skill operation method based on the parameter expression knowledge base includes: Obtaining a first knowledge base of the robot, and determining a target entity and its associated entity information based on the first knowledge base; Obtaining a second knowledge base of the robot, and determining, based on the second knowledge base and the entity information, a relationship between the target entity and other entities in a current scenario; the second knowledge base is a knowledge base established based on the first knowledge base, and a plurality of application scenarios are provided in the second knowledge base, each of the application scenarios being composed of corresponding relationships in the first knowledge base; Obtaining a third knowledge base of the robot, and searching for corresponding skills and action primitives based on the third knowledge base and the relationship; Based on the skills and action primitives found, the corresponding actuators and components are controlled to perform the target tasks in the current scenario; The obtaining of the third knowledge base of the robot and searching for corresponding skills and action primitives according to the third knowledge base and the relationship include: Acquire a third knowledge base of the robot according to the relationship; Find the skill tree corresponding to the relationship in the current task and scenario; Determining the skills and action primitives required for the current task according to the skill tree; The step of searching for a skill tree corresponding to the relationship in the current task and scenario previously includes: Build skill trees corresponding to multiple tasks and scenarios through the arrangement and combination of different skills and different action primitives; The controlling of corresponding actuators and components according to the skills and action primitives found includes: Determining corresponding execution parameters based on the found skills and action primitives; The corresponding actuators and components are controlled according to the execution parameters.

2. The robot skill operation method based on parameter expression knowledge base according to claim 1 is characterized in that: The step of obtaining a first knowledge base of the robot and determining a target entity and its associated entity information based on the first knowledge base may include: Pre-setting the first knowledge base, the second knowledge base, and the third knowledge base according to the node information; The node information includes node name, node ID, node attributes, node labels and node relationships.

3. The robot skill operation method based on parameter expression knowledge base according to claim 2 is characterized in that: The acquiring of a first knowledge base of the robot and determining a target entity and its associated entity information based on the first knowledge base includes: Obtaining task information of the robot; Acquire a first knowledge base of the robot according to the task information; The target entity and its associated entity information are determined according to the first knowledge base; wherein the entity information includes: entity type, entity attributes and entity associated file information.

4. The robot skill operation method based on parameter expression knowledge base according to claim 1 is characterized in that: The acquiring of a second knowledge base of the robot and determining, based on the second knowledge base and the entity information, a relationship between the target entity and other entities in the current scenario includes: Acquire a second knowledge base of the robot according to the target entity; Determine the scenario in which the target entity is located, and obtain, based on the scenario, a set of relationships between entities corresponding to the scenario in the second knowledge base; Searching the relationship between the target entity and other entities in the relationship set according to entity classification; The relationships include: positional relationships, logical relationships, and active and passive relationships.

5. The robot skill operation method based on parameter expression knowledge base according to claim 4 is characterized in that: The step of searching the relationship between the target entity and other entities in the relationship set according to entity classification includes: The relationship between the target component and other components, operating platforms and actuators is searched in the relationship set according to the entity classification.

6. A terminal, characterized in that: include: A processor and a memory, wherein the memory stores a robot skill operation program based on a parameter expression knowledge base, and when the robot skill operation program based on the parameter expression knowledge base is executed by the processor, it is used to implement the robot skill operation method based on the parameter expression knowledge base as described in any one of claims 1 to 5.

7. A medium, characterized in that The medium is a computer-readable storage medium, which stores a robot skill operation program based on a parameter expression knowledge base. When the robot skill operation program based on the parameter expression knowledge base is executed by a processor, it is used to implement the robot skill operation method based on the parameter expression knowledge base as described in any one of claims 1-5.

Images