Artificial intelligence interaction method and system based on uniform object identification and artificial intelligence body

Abstract

The invention relates to an artificial intelligence interaction method and system based on uniform object identification and an artificial intelligence body, and is applied to the field of artificial intelligence. After an interaction request is received, a server determines whether the interaction request meets a preset interaction condition, and the interaction condition comprises that a to-be-verified identifier in the service request is a unique identifier of an artificial intelligence. And in response to determining that the interaction request meets the interaction condition, the server acquires a verification public key according to the to-be-verified identifier in the service request, and verifies the to-be-verified signature by using the verification public key. And in response to the fact that the signature to be verified passes verification, the server assists the initiating object of the interaction request to establish communication connection with at least one target interaction object. Therefore, a safe and effective artificial intelligence and real world interaction system is constructed.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to an artificial intelligence interaction method, system and artificial intelligence entity based on unified object identification. Background Art

[0002] In recent years, pre-trained large language models have continued to evolve in structure and scale. With the emergence of multimodal functions and emergent phenomena, there is an optimistic view that the era of artificial general intelligence (AGI) is approaching.

[0003] The functions of the pre-trained large language model can be summarized into two levels. The first is to build a conversational knowledge base through fine-tuning training and retrieval-augmented generation (RAG); the second is to access and organize external data and functional interfaces through tool calls (tools) and task agents (agents), thereby building the ability to interact with external information and even the real world.

[0004] At present, how to build a safe and effective interaction system between artificial intelligence and the real world is an urgent problem that needs to be solved. Summary of the invention

[0005] In order to build a safe and effective interaction system between artificial intelligence and the real world, the present application provides an artificial intelligence interaction method, system and artificial intelligence entity based on unified object identification.

[0006] In the first aspect, the embodiment of the present application provides an artificial intelligence interaction method based on a unified object identification. The method is executed by a server, and includes: receiving an interaction request, the interaction request includes an identification to be verified, a signature to be verified, and a unique identification of at least one target interaction object; determining whether the interaction request meets a preset interaction condition, the interaction condition includes that the identification to be verified is a unique identification of an artificial intelligence entity; in response to determining that the interaction request meets the interaction condition, obtaining a verification public key according to the identification to be verified, and using the verification public key to verify the signature to be verified; in response to the signature to be verified passing the verification, assisting the initiator of the interaction request to establish a communication connection with the at least one target interaction object.

[0007] In some embodiments, the interaction request also includes a public key to be verified, and obtaining the verification public key according to the identifier to be verified includes: determining whether the public key to be verified matches the identifier to be verified according to a preset public key generation algorithm; in response to determining that the public key to be verified matches the identifier to be verified, determining the public key to be verified as the verification public key.

[0008] In some embodiments, the unique identifier of an artificial intelligence entity includes a version part, an instance part, and an embodiment condition part; when the target interactive object is an entity that is not an artificial intelligence entity, the unique identifier of the target interactive object includes an entity organization, an entity category, and a custom code.

[0009] In some embodiments, assisting the initiating object of the interaction request to establish a communication connection with the at least one target interaction object includes: determining a compatible communication method based on the version part in the identifier to be verified and the entity category in the unique identifier of the at least one target interaction object; assisting the initiating object of the interaction request to establish a communication connection with the at least one target interaction object according to the compatible communication method.

[0010] In some embodiments, assisting the initiating object of the interaction request to establish a communication connection with the at least one target interaction object includes: returning an access interface of the at least one target interaction object to the initiating object of the interaction request.

[0011] In some embodiments, after assisting the initiating object of the interaction request to establish a communication connection with the at least one target interaction object, the method further includes:

[0012] Acquire knowledge graphs from the initiator of the interaction request and the at least one target interaction object respectively, to obtain at least two sub-graphs;

[0013] Fusing the at least two sub-graphs to obtain a fused knowledge graph;

[0014] The fused knowledge graph is sent to the initiator of the interaction request and the at least one target interaction object respectively.

[0015] In some embodiments, the at least one target interaction object includes a graph calculation party and at least one sub-graph provider, and the graph calculation party is used to: fuse the knowledge graph provided by the initiating object of the interaction request and the knowledge graph provided by the at least one sub-graph provider, and send the fused knowledge graph to the initiating object of the interaction request and the at least one sub-graph provider respectively.

[0016] In some embodiments, the unique identifier is generated based on the MA identification system of ISO / IEC15459 specification.

[0017] In the second aspect, the embodiment of the present application provides an artificial intelligence interaction system based on unified object identification. The artificial intelligence interaction system is implemented on the server side, and includes: a receiving module for receiving an interaction request, the interaction request includes an identification to be verified, a signature to be verified, and a unique identification of at least one target interaction object; a verification module for: determining whether the interaction request meets a preset interaction condition, the preset interaction condition includes that the identification to be verified is a unique identification of an artificial intelligence entity; in response to determining that the interaction request meets the interaction condition, obtaining a verification public key according to the identification to be verified, and using the verification public key to verify the signature to be verified; an assistance module for: in response to the signature to be verified passing the verification, assisting the initiator of the interaction request to establish a communication connection with the at least one target interaction object.

[0018] In some embodiments, the artificial intelligence interaction system also includes a sub-graph fusion module, which is used to: obtain knowledge graphs from the initiator of the interaction request and the at least one target interaction object respectively to obtain at least two sub-graphs; fuse the at least two sub-graphs to obtain a fused knowledge graph; and send the fused knowledge graph to the initiator of the interaction request and the at least one target interaction object respectively.

[0019] In a third aspect, an embodiment of the present application provides an artificial intelligence entity, including a processor and a memory. The memory stores a computer program, and when the processor executes the computer program, the artificial intelligence interaction method based on the unified identification of objects as described in the first aspect is implemented.

[0020] Compared with the prior art, the technical solution provided by the embodiments of the present application has the following advantages:

[0021] (1) Relying on the unified object identification system and unified identification resolution system, a safe and effective interaction system between artificial intelligence and the real world is built.

[0022] (2) With the interaction system between artificial intelligence and the real world, artificial intelligence entities can expand the cognitive system together with other objects. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0024] Figure 1 Schematic diagram of the scenario of interaction between artificial intelligence and the real world.

[0025] Figure 2 An exemplary architecture diagram of the MA identity resolution system provided in an embodiment of the present application.

[0026] Figure 3 An exemplary flowchart of an artificial intelligence interaction method based on unified object identification provided in an embodiment of the present application.

[0027] Figure 4 An exemplary flowchart of the process of expanding the knowledge graph of an artificial intelligence entity provided in an embodiment of the present application.

[0028] Figure 5A A schematic diagram of the first scenario of an artificial intelligence entity expanding a knowledge graph provided in an embodiment of the present application.

[0029] Figure 5B Schematic diagram of the second scenario of the artificial intelligence entity expanding the knowledge graph provided in the embodiment of the present application.

[0030] Figure 6 An exemplary block diagram of an artificial intelligence interaction system based on unified object identification provided in an embodiment of the present application.

[0031] Figure 7 A schematic diagram of the structure of an artificial intelligence entity provided in an embodiment of the present application. DETAILED DESCRIPTION

[0032] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the embodiments of the present application will be further described below. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0033] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application can also be implemented in other ways different from those described herein. Obviously, the embodiments in the specification are only part of the embodiments of the present application, not all of the embodiments.

[0034] Figure 1 Schematic diagram of an application scenario of artificial intelligence interacting with the real world provided in an embodiment of the present application. Scenario 100 includes multiple artificial intelligence entities 110, multiple external objects 120 and a server 130.

[0035] Strictly speaking, artificial intelligence is part of the real world. For the convenience of expression, the real world outside artificial intelligence is referred to as the external world. It can be understood that the external object 120 is an object belonging to the external world.

[0036] The interaction system between artificial intelligence and the real world includes three parts: a unified object identification system, a cognitive system of artificial intelligence entities, and a unified identification resolution system. Among them, the unified identification resolution system is implemented on the server 130.

[0037] The object unified identification system includes two aspects: unified identification of the artificial intelligence entity 110 and unified identification of the external object 120. In some embodiments, the unique identification of the artificial intelligence entity 110 includes identification of its version, instance, embodiment conditions and other elements, which correspond to the species, individual and physical sign information of the artificial intelligence entity 110 respectively (in the terminology of the intelligent manufacturing standard, the above content is called "technical object"). Establishing a unified identification is not only to issue an "identity card" to each artificial intelligence entity 110, but also to support the self-cognition of the artificial intelligence entity 110. In some embodiments, the unique identification of the external object 120 includes entity organization, entity category and custom coding to realize the identification and coding of various physical objects, so that the artificial intelligence entity 110 can accurately identify each physical object instance.

[0038] The cognitive system of artificial intelligence about the world is the "worldview" of artificial intelligence 110. Compared with "near-field" cognition such as embodied cognition and environmental cognition, artificial intelligence 110 transforms the knowledge about the categories, hierarchies, relationships and attributes of external objects learned from semantics into a structured knowledge system (for example, a knowledge graph), so that it can accurately understand, access and manipulate various objects, and build sufficiently complex task capabilities to interact with the real world.

[0039] The unified identification resolution system for artificial intelligence will build a connection hub between artificial intelligence entities 110 and between artificial intelligence entities 110 and external objects 120, mainly including two types of functions: entity identity resolution and external object resolution service. The entity identity resolution mainly plays the role of a "public security bureau", issuing an "identity card" for each artificial intelligence instance (i.e., artificial intelligence entity 110), and providing resolution of the real identity of the artificial intelligence entity 110. It is the basis for multiple artificial intelligence entities 110 to interact with each other and build a trusted intelligent group. The external object resolution service is the "safety gate" for artificial intelligence entities 110 to access and manipulate the real world (including digital and physical), and the precise interface, permission management, and security control of artificial intelligence entities 110 to access various objects in the real world are realized.

[0040] The following uses the MA identity resolution system as an example to explain. Figure 2 ,To ensure the uniqueness of the identifier, the overall architecture of the MA ,identification resolution system is a tree hierarchical structure, which from top to bottom are the MA root node, the MA first-level node, the MA second-level node and the MA third-level node.

[0041] The MA root node is at the top level of the tree hierarchy, and the identification prefix assigned to the root node is "MA". There are multiple root nodes, and all root nodes form the MA root node alliance chain. In addition, each root node can also have multiple mirror root nodes. Different root nodes are responsible for managing their subordinate first-level nodes, including the addition, deletion, and update of first-level nodes. At the same time, the MA root node maintains the information of its subordinate first-level nodes, including node prefixes, IP addresses, and identification registration volume, resolution volume and other data. MA root nodes synchronize the prefixes and IP addresses of the MA first-level nodes they maintain with each other, and all root nodes have a full amount of first-level node prefixes and IP addresses.

[0042] The MA first-level node is a subordinate node of the root node, generally a national region node or an industry field node. The identification prefix assigned to the first-level node is "MA. Country region code / industry field code". Each first-level node can only be managed by one root node. The MA first-level node is responsible for managing its subordinate MA second-level nodes, including the addition, deletion and update of second-level nodes. At the same time, the MA first-level node maintains the information of its subordinate second-level nodes, including node prefixes, IP addresses, and identification registration volume, resolution volume and other data. The MA first-level node must synchronize the identification registration volume, resolution volume and other data to the root node to which it belongs.

[0043] MA secondary nodes are subordinate nodes of primary nodes, generally regional nodes or industry nodes. The identification prefix assigned to secondary nodes is "MA. Country code / Industry code. Regional code / Industry code". Each secondary node can only be managed by one primary node. MA secondary nodes are responsible for managing their subordinate MA third-level nodes, including the addition, deletion and update of third-level nodes. At the same time, MA secondary nodes maintain information on their subordinate third-level nodes, including node prefixes, IP addresses, and identification registration and resolution data. MA secondary nodes must synchronize identification registration and resolution data to their subordinate primary nodes.

[0044] MA level 3 nodes are subordinate nodes of level 2 nodes, usually city nodes or enterprise nodes. The identifier prefix assigned to level 3 nodes is "MA. Country code / industry code. Region code / industry code. City code / enterprise code". Each MA level 3 node can only be managed by one level 2 node. MA level 3 nodes are responsible for managing identifiers, including identifier registration, resolution, deletion and update, as well as identifier access rights management. MA level 3 nodes must synchronize identifier registration and resolution data to the level 2 nodes to which they belong. In the MA identifier resolution system, level 3 nodes are nodes responsible for storing identifier data, and identifier resolution requests will be ultimately responded to by level 3 nodes.

[0045] The artificial intelligence agent 110 may send an interaction request to the server 130 to establish a communication connection with at least one target interaction object. The at least one target interaction object may include at least one artificial intelligence agent 110 and / or at least one external object. In some embodiments, the artificial intelligence agent 110 may establish a communication connection with at least one other artificial intelligence agent 110 through the server 130, thereby building a trusted intelligent group.

[0046] In some embodiments, the artificial intelligence agent 110 specifically refers to an embodied intelligent agent, which involves a robot form and AI capabilities. The embodied intelligent agent is not limited to a humanoid robot, but also includes physical robots of other forms, such as a robot dog, an autonomous vehicle, a robotic arm, etc.

[0047] The external object 120 may be any type of device with interaction (communication) capabilities, such as a smart phone, a personal computer, a server, an Internet of Things device, etc.

[0048] The server 130 is used to receive and process interaction requests from the artificial intelligence agent 110 to assist the artificial intelligence agent 110 in establishing a communication connection with at least one target interaction object.

[0049] Figure 3 This is an exemplary flow chart of an artificial intelligence interaction method based on object unified identification provided in an embodiment of the present application. Process 200 is executed by the server 130. Figure 3 As shown, process 300 includes the following steps.

[0050] Step 210: receiving an interaction request, wherein the interaction request includes an identifier to be verified, a signature to be verified, and a unique identifier of at least one target interaction object.

[0051] In the case that the content of the interaction request has not been tampered with, the identifier to be verified is the unique identifier of the initiator of the interaction request, and the signature to be verified is the signature of the initiator of the interaction request.

[0052] In some embodiments, the interaction request also includes a public key to be verified. The details of the use of the public key to be verified can be found in the relevant description of step 230, which will not be elaborated here.

[0053] In some embodiments, the interaction request further includes a target task type to indicate the type of task that needs to be performed together after the communication connection is established, for example, a subgraph fusion task.

[0054] Step 220, determining whether the interaction request satisfies a preset interaction condition, wherein the interaction condition includes that the identifier to be verified is a unique identifier of an artificial intelligence entity.

[0055] The server 130 can determine whether the identifier to be verified is a unique identifier of an artificial intelligence entity by checking the format of the identifier to be verified (such as field composition) or a specific field of the identifier to be verified (used to distinguish whether the entity is an artificial intelligence entity).

[0056] In some embodiments, the unique identifier of an artificial intelligence entity includes a version part, an instance part, and an embodiment condition part. When the target interactive object is an entity that is not an artificial intelligence entity (i.e., an external object), the unique identifier of the target interactive object includes an entity organization, an entity category, and a custom code.

[0057] The version part may include one or more version information such as the identification code issuing agency, the intelligent entity issuing entity, the algorithm name, and the version number. The identification code issuing agency may be an identification code issuing agency recognized by standard organizations such as ISO and IEC. For example, the Zhongguancun Industry and Information Technology QR Code Technology Research Institute is an issuing agency of the "MA" logo authorized by the three major international standard organizations ISO, CEN and AIM. The intelligent entity issuing entity refers to the unit or individual that issues the artificial intelligence entity. The specific code may include the country, unit name, individual name, and ID number. The algorithm name refers to the name of the algorithm used by the artificial intelligence entity, which needs to be unique within the scope of the intelligent entity issuing entity. The version number refers to the version of the algorithm used by the artificial intelligence entity, which needs to be unique within the algorithm scope.

[0058] The instance part contains the instance number. The same version can have multiple instances. In practical applications, sequential numbering or segmented numbering can be used.

[0059] The embodiment condition part includes the number of the embodiment condition. In practical applications, sequential numbering or segmented numbering can be used. One segmented numbering method is: form. application scenario. sequence number. For example, the form can include human form, robot dog, etc., and the application scenario can include home, office, convenience store, etc.

[0060] Entity organization refers to the organization to which the entity belongs, including the issuing agency of the identification code and the organization code. The issuing agency and the organization code can be separated by a specific symbol such as ".". Among them, the issuing agency of the identification code can be an issuing agency recognized by standard organizations such as ISO and CEN, such as the "MA" logo corresponding to Zhongguancun Gongxin QR Code Technology Research Institute. The organization code is assigned by the issuing agency of the identification code, and the format is not limited. It can be a combination of letters and numbers.

[0061] Entity category refers to the category of entity within the organization. This structure can contain multiple layers of subdivision structure, and the multiple layers can be separated by specific symbols such as ".".

[0062] Custom encoding can be in the form of serial numbers, etc.

[0063] In some embodiments, the unique identifiers of the artificial intelligence entity and the external object are generated based on the MA identifier system specified by ISO / IEC 15459. In combination with the above embodiments, a field indicating whether the identifier is the unique identifier of the artificial intelligence entity can be set in the MA identifier (also called the MA code).

[0064] Step 230: In response to determining that the interaction request satisfies the interaction condition, a verification public key is obtained according to the identifier to be verified, and the verification public key is used to verify the signature to be verified.

[0065] It can be understood that in response to determining that the interaction request does not meet the interaction condition, the server 130 can return a message indicating that the interaction failed.

[0066] In some embodiments, stronger interaction conditions may be set. For example, when the interaction request also includes the target task type, the interaction condition may also include that the target task type is in the set of legal task types specified by each target interaction object. Specifically, assuming that the target task type is a subgraph fusion task, then satisfying the interaction condition requires that the set of legal task types specified by each target interaction object includes a subgraph fusion task. For another example, when the identifiers of at least one object are both identifiers of an artificial intelligence entity, the interaction condition may also include that the application scenarios in the identifier to be verified and the identifier of at least one object are the same (for example, both are offices).

[0067] In some embodiments, the interaction request also includes a public key to be verified. Based on this, the server 130 can obtain the verification public key in the following manner: determine whether the public key to be verified matches the identifier to be verified according to a preset public key generation algorithm; in response to determining that the public key to be verified matches the identifier to be verified, determine the public key to be verified as the verification public key. The public key generation algorithm is used to generate the public key of the public key holder based on the identifier of the public key holder. The identifier as the input of the algorithm and the public key as the output of the algorithm form a one-to-one correspondence. Therefore, when the public key generated based on the identifier to be verified is the same as the public key to be verified, it means that the public key to be verified matches the identifier to be verified. It can be understood that in response to determining that the public key to be verified does not match the identifier to be verified (that is, the public key generated based on the identifier to be verified is different from the public key to be verified), the server 130 can return a message of interaction failure.

[0068] As an example only, the signature verification process is as follows: use the verification public key to decrypt the signature to be verified to obtain a first summary value (such as a hash value); use a preset algorithm (such as a hash algorithm) to perform a summary calculation on the content to be verified of the interaction request to obtain a second summary value, wherein the content to be verified includes the identifier to be verified and the unique identifier of at least one target interaction object (it may also include more content in addition to the signature to be verified, such as the target task type); compare the first summary value and the second summary value; if the first summary value and the second summary value are the same, the signature to be verified passes the verification; if the first summary value and the second summary value are different, the signature to be verified fails the verification.

[0069] Step 240: In response to the signature to be verified passing verification, assisting the initiator of the interaction request to establish a communication connection with the at least one target interaction object.

[0070] It can be understood that, in response to the signature to be verified failing to pass the verification, the server 130 may return a message indicating that the interaction has failed.

[0071] In some embodiments, the server 130 may maintain a communication connection with both the initiator of the interaction request and the at least one target interaction object. In this way, the initiator of the interaction request and the at least one target interaction object establish an indirect communication connection through the server 130.

[0072] Of course, the server 130 can also assist the initiator of the interaction request to establish a direct communication connection with the at least one target interaction object. In this way, after the initiator of the interaction request establishes a communication connection with the at least one target interaction object, the initiator of the interaction request can disconnect from the server 130.

[0073] In some embodiments, the server 130 can determine the compatible communication mode according to the version part in the identification to be verified and the entity category in the unique identification of at least one target interactive object. Both the version part and the entity category can reflect the communication mode supported by the device. For example, both contain operating system information, thereby determining the compatible communication mode of different devices. Furthermore, the server 130 can assist the initiator of the interaction request to establish a communication connection with the at least one target interactive object according to the compatible communication mode. In this way, support is provided for the interaction between the artificial intelligence entity and various devices.

[0074] In some embodiments, the server 130 may return the access interface of the at least one target interactive object to the initiating object of the interactive request to assist the initiating object of the interactive request in establishing a communication connection with the at least one target interactive object. The server 130 may pre-save an object access interface list, which includes access interfaces of multiple objects (artificial intelligence entities / external objects).

[0075] Figure 4 An exemplary flow chart of the process of expanding the knowledge graph of an artificial intelligence entity provided in an embodiment of the present application. In some embodiments, after step 240, process 200 also includes Figure 4 Steps 310 to 330 are shown. That is, the server 130 is also used to provide sub-graph fusion service.

[0076] Step 310: Obtain knowledge graphs from the initiator object of the interaction request and the at least one target interaction object respectively to obtain at least two sub-graphs.

[0077] A knowledge graph is a graph-based data structure used to represent and store entities, concepts, attributes, and their relationships in the real world. It organizes information into nodes (entities or concepts) and edges (relationships).

[0078] refer to Figure 5A , the initiator of the interaction request (denoted as P_0) and n target interaction objects (denoted as P_1, ..., P_n) each provide a knowledge graph to the server 130, and the knowledge graph from each object can be regarded as a subgraph (denoted as X). Then, the server 130 obtains n+1 subgraphs, which are denoted as X_0, X_1, ..., X_n.

[0079] Step 320, fusing the at least two sub-graphs to obtain a fused knowledge graph.

[0080] In some embodiments, subgraph fusion may be performed according to a predetermined rule. In some embodiments, subgraph fusion may be performed by a machine learning method. In some embodiments, subgraphs may be converted into embedding vectors and then matched and fused in the embedding space.

[0081] Step 330: Send the fused knowledge graph to the initiator of the interaction request and the at least one target interaction object respectively.

[0082] The fused knowledge graph (denoted as Y) contains more knowledge, that is, it has more nodes and edges. After the artificial intelligence entity establishes a communication connection with at least one target interactive object, the cognitive system can be improved through subgraph fusion. As a participant in the subgraph fusion, the at least one target interactive object can also obtain the fused knowledge graph.

[0083] In some embodiments, reference Figure 5B , the at least one target interaction object includes a graph calculation party 140 and at least one subgraph provider (denoted as P_1, ..., P_n), and the graph calculation party 140 is used to: fuse the knowledge graph (denoted as X_0) provided by the initiator of the interaction request (denoted as P_0) and the knowledge graph (denoted as X_1, ..., X_n) provided by the at least one subgraph provider, and send the fused knowledge graph (denoted as Y) to the initiator of the interaction request and the at least one subgraph provider respectively. The graph calculation party 140 can be a computing device with sufficient computing power, which is specially used for subgraph fusion. In this way, it can avoid the server 130 being overloaded, causing it to respond too slowly or even crash.

[0084] Figure 6 This is an exemplary block diagram of an artificial intelligence interaction system based on unified object identification provided in an embodiment of the present application. The system 400 can be implemented at the server 130. Figure 6 As shown, the system 400 includes a receiving module 410 , a verification module 420 , and an assistance module 430 .

[0085] The receiving module 410 is used to receive an interaction request, where the interaction request includes an identifier to be verified, a signature to be verified, and a unique identifier of at least one target interaction object.

[0086] The verification module 420 is used to: determine whether the interaction request satisfies a preset interaction condition, wherein the preset interaction condition includes that the identifier to be verified is a unique identifier of an artificial intelligence entity; in response to determining that the interaction request satisfies the interaction condition, obtain a verification public key according to the identifier to be verified, and use the verification public key to verify the signature to be verified.

[0087] The assisting module 430 is used for assisting the initiating object of the interaction request to establish a communication connection with the at least one target interaction object in response to the signature to be verified passing the verification.

[0088] In some embodiments, the system 400 also includes a sub-graph fusion module 440, which is used to: obtain knowledge graphs from the initiator object of the interaction request and the at least one target interaction object respectively to obtain at least two sub-graphs; fuse the at least two sub-graphs to obtain a fused knowledge graph; and send the fused knowledge graph to the initiator object of the interaction request and the at least one target interaction object respectively.

[0089] Figure 7 This is a schematic diagram of the structure of the artificial intelligence entity provided in the embodiment of the present application. Figure 7As shown, the artificial intelligence entity 500 includes a processor 510 and a memory 520. The memory 510 stores a computer program, and when the processor 520 executes the computer program, the artificial intelligence interaction method based on the unified identification of objects provided in the embodiment of the present application is implemented.

[0090] In some embodiments, the artificial intelligence agent 500 may be an embodied agent, i.e., an AI robot.

[0091] More details about the AI ​​interaction method can be found in Figure 3 and its related descriptions, which will not be repeated here.

[0092] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusions, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to the process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "including a..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0093] The above is only an embodiment of the present application, and the embodiment enables those skilled in the art to understand and implement the present application. Various modifications to the embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and features disclosed herein.

Claims

1. An artificial intelligence interaction method based on unified object identification, characterized in that: The method is executed by the server, and includes: receiving an interaction request, wherein the interaction request includes an identifier to be verified, a signature to be verified, and a unique identifier of at least one target interaction object; Determining whether the interaction request satisfies a preset interaction condition, wherein the interaction condition includes that the identifier to be verified is a unique identifier of an artificial intelligence entity; In response to determining that the interaction request satisfies the interaction condition, obtaining a verification public key according to the identifier to be verified, and verifying the signature to be verified using the verification public key; In response to the signature to be verified passing verification, assisting the initiating object of the interaction request to establish a communication connection with the at least one target interaction object.

2. The method according to claim 1, characterized in that The interaction request also includes a public key to be verified, and obtaining the verification public key according to the identifier to be verified includes: Determining whether the public key to be verified matches the identifier to be verified according to a preset public key generation algorithm; In response to determining that the public key to be verified matches the identifier to be verified, the public key to be verified is determined as the verification public key.

3. The interactive method according to claim 1, characterized in that: The unique identifier of the artificial intelligence entity includes a version part, an instance part and an embodiment condition part; when the target interaction object is an entity that is not an artificial intelligence entity, the unique identifier of the target interaction object includes an entity organization, an entity category and a custom code.

4. The method according to claim 3, characterized in that Assisting the initiator of the interaction request to establish a communication connection with the at least one target interaction object, including: Determining a compatible communication mode according to a version portion in the identification to be verified and an entity category in the unique identification of the at least one target interactive object; Assisting the initiator of the interaction request to establish a communication connection with the at least one target interaction object according to the compatible communication mode.

5. The method according to claim 1, characterized in that Assisting the initiator of the interaction request to establish a communication connection with the at least one target interaction object, including: An access interface of the at least one target interaction object is returned to the initiator object of the interaction request.

6. The method according to claim 1, characterized in that After assisting the initiating object of the interaction request to establish a communication connection with the at least one target interaction object, the method further includes: Acquire knowledge graphs from the initiator of the interaction request and the at least one target interaction object respectively, to obtain at least two sub-graphs; Fusing the at least two sub-graphs to obtain a fused knowledge graph; The fused knowledge graph is sent to the initiator of the interaction request and the at least one target interaction object respectively.

7. The method according to claim 1, characterized in that The at least one target interaction object includes a graph calculation party and at least one sub-graph provider, and the graph calculation party is used to: fuse the knowledge graph provided by the initiator object of the interaction request and the knowledge graph provided by the at least one sub-graph provider, and send the fused knowledge graph to the initiator object of the interaction request and the at least one sub-graph provider respectively.

8. The method according to claim 1, characterized in that The unique identifier is generated based on the MA identification system of ISO / IEC15459 specification.

9. An artificial intelligence interaction system based on unified object identification, characterized in that: The artificial intelligence interaction system is implemented on the server side, and includes: A receiving module, configured to receive an interaction request, wherein the interaction request includes an identifier to be verified, a signature to be verified, and a unique identifier of at least one target interaction object; A verification module, configured to: determine whether the interaction request satisfies a preset interaction condition, wherein the preset interaction condition includes that the identifier to be verified is a unique identifier of an artificial intelligence entity; in response to determining that the interaction request satisfies the interaction condition, obtain a verification public key according to the identifier to be verified, and use the verification public key to verify the signature to be verified; The assisting module is used for assisting the initiating object of the interaction request to establish a communication connection with the at least one target interaction object in response to the signature to be verified passing the verification.

10. An artificial intelligence entity, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and when the processor executes the computer program, the artificial intelligence interaction method based on unified object identification as described in any one of claims 1 to 8 is implemented.

Images