Space-time AI data processing method in block chain fusion meta universe scene
Through the blockchain fusion of spatiotemporal AI data processing methods of metacosmic scenes, dynamically update NPC attributes and interaction behaviors, the problem of restricted interaction mode between NPC and user in the existing technology is solved, and the dynamic evolution of NPC and user interaction in the metacosmic is realized and the real-world interaction simulation is realized.
Patent Information
- Application Number
- CN202510223424.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-27
- Publication Date
- 2025-06-27
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The failure of the prior art to effectively utilize spatiotemporal data in the metaverse has resulted in the NPC interaction mode still being limited by traditional modes, lacking simulations of dynamic evolution and real-world interaction.
Through blockchain fusion of spatiotemporal AI data processing methods in metacosmic scenarios, users' historical interaction data, NPC attribute information and game environment context information are collected, two-dimensional structural data are generated and input into the metacosmic interaction model. The deep learning model is used to dynamically update NPC attributes and interaction behaviors, and the dynamic evolution of NPC interactions with users is realized.
It realizes the dynamic evolution of NPC and user interaction process, changes NPC attributes and interaction modes, simulates the interaction effect of the real world, and transcends the limitations of traditional dynamic interaction.
Smart Images
Figure CN120204728A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of the metaverse, and more specifically, it relates to a method for processing spatio-temporal AI data in a blockchain-integrated metaverse scenario. Background Art
[0002] The metaverse is a highly dynamic and interactive virtual environment, in which various entities and activities generate a large amount of spatio-temporal data. Efficiently processing and utilizing this spatio-temporal data is crucial for achieving the immersion, interactivity, and intelligence of the metaverse. First, the blockchain can be used for privacy protection of spatio-temporal data and good-structured time-series data can be obtained, providing good support for the dynamic operation of the metaverse; in large-scale multiplayer online metaverse games, the intelligent interaction of NPCs is the key to enhancing immersion, which requires not only the recognition of time-series data on the timeline but also the recognition of the spatial structure among users, NPCs, and interaction knowledge; although such recognition methods are adopted in the prior art, they still do not deviate from the traditional mode of interaction between NPCs and users. Summary of the Invention
[0003] The present invention provides a method for processing spatio-temporal AI data in a blockchain-integrated metaverse scenario to solve the technical problems in the related art.
[0004] The present invention provides a method for processing spatio-temporal AI data in a blockchain-integrated metaverse scenario, including the following steps:
[0005] Step 100, collecting metaverse scenario information from the blockchain, where the metaverse scenario information includes historical interaction data of users, attribute information of NPCs, and context information of the game environment;
[0006] Step 200, generating first two-dimensional structure data, where the first two-dimensional structure data includes a first data matrix and a first relationship matrix. A unit of the first data matrix represents the first one-dimensional structure data of an independent object, and the independent objects include users, NPCs, scenes, tasks, and interaction events. A unit of the first data matrix representing an independent object only contains the metaverse scenario information of this independent object;
[0007] The element in the i-th row and j-th column of the first relationship matrix represents the association between the independent objects represented by the i-th unit and the j-th unit of the first data matrix. If there is an association, the value of this element in the first relationship matrix is 1, otherwise it is 0;
[0008] There is an association between an interaction event and the users, NPCs, scenes, and tasks associated with this interaction event;
[0009] There is an association between a user and the NPC with which the user is interacting;
[0010] The NPC is associated with all tasks;
[0011] A one-dimensional structure data of the first type includes n data items sorted by time. The t-th data item represents the metaverse scene information of the corresponding independent object collected at the t-th moment;
[0012] Step 300: Input the two-dimensional structure data of the first type into the metaverse interaction model. The metaverse interaction model includes a first intermediate layer, a second intermediate layer, a first output layer, and a second output layer. The first intermediate layer inputs the one-dimensional structure data of the first type and outputs the first intermediate representation data to the second intermediate layer. The second intermediate layer also inputs the first relationship matrix. The second intermediate layer outputs the second intermediate representation data to the first output layer and the second output layer. The first output layer outputs the result representing the NPC interaction action, and the second output layer outputs the result representing the NPC attribute change.
[0013] Furthermore, the user's historical interaction data includes conversation content, behavior decisions, and emotional expressions.
[0014] Furthermore, the attribute information of the NPC includes identity, personality, ability, etc.
[0015] Furthermore, the context information of the game environment includes scenes, tasks, and events.
[0016] Furthermore, the calculation formula of the first intermediate layer of the metaverse interaction model is as follows:
[0017] h t =tanh(W hh h t-1 +W xh x t +b h )
[0018] Where h t represents the t-th first intermediate representation data output by the first intermediate layer, h t-1 represents the (t - 1)-th first intermediate representation data, x t represents the t-th data item of the one-dimensional structure data of the first type, W hh and W xh are the first and second weight parameters, b h is the first bias parameter, and tanh is the hyperbolic tangent function.
[0019] Furthermore, the calculation formula of the second intermediate layer is as follows:
[0020]
[0021] Where k v represents the second intermediate representation data of the v-th unit of the first data matrix, M (v)is the set of cells associated with the v-th cell of the first data matrix, represents the n-th first intermediate representation data output when the first one-dimensional structure data of the v-th cell of the first data matrix is input to the first intermediate layer, c v is a normalization constant (default is 13), σ is the sigmoid function, W k represents the third weight parameter, b k is the second bias parameter.
[0022] Furthermore, the calculation formula of the first output layer is as follows:
[0023] OUT v = σ(W out *k v +b out ), v ∈ M (NPC)
[0024] where OUT v represents the first output vector of the NPC represented by the v-th cell of the first data matrix, and the c-th component value represents the probability value of the c-th interaction action. The interaction action with the largest probability value is selected as the output. The interaction actions include the interaction behaviors between the NPC and the user and the tasks that the NPC can assign to the user, k v represents the second intermediate representation data of the v-th cell of the first data matrix, M (NPC) represents the set of all cells representing the NPC in the first data matrix, W out is the first output weight parameter, b out is the first output bias parameter, and σ represents the sigmoid function.
[0025] Furthermore, the calculation formula of the second output layer is as follows:
[0026] y v = σ(W y k v +b y ), v ∈ M (NPC)
[0027] where y v represents the second output vector of the NPC represented by the v-th cell of the first data matrix, and the c-th component value represents the probability value of the c-th attribute strategy. The attribute strategy with the largest probability value is selected as the output. The attribute strategies include the various attribute values of the NPC, k v represents the second intermediate representation data of the v-th cell of the first data matrix, W y is the second output weight parameter, b y is the second output bias parameter, and σ represents the sigmoid function.
[0028] Further, the steps for training the metaverse interaction model include:
[0029] Step 101, initialize the parameters of the metaverse interaction model;
[0030] Step 102, observe the metaverse scene information S at time t t , the policy (including interaction actions and attribute policies) A executed at time t t , the metaverse scene information S at time t+1 t+1 , execute the policy A t to obtain the reward R t ;
[0031] Step 103, then calculate the policy error:
[0032]
[0033] where δ t represents the policy error at time t, γ represents the discount factor, γ ∈ [0, 1], represents the sum of the maximum probability values in the first output vector and the second output vector output by the metaverse interaction model when inputting S t+1 , represents the sum of the probability values corresponding to the policy A in the first output vector and the second output vector output by the metaverse interaction model when inputting S t ; t
[0034]
[0035] r q represents the average satisfaction of all users with the qth NPC, b q represents the attribute change ratio of the qth NPC, F t represents the per-unit value of the data processing volume when executing the policy A t , and γ1, γ2, and γ3 are the expected target weights, with default values of 0.6, 0.2, and 0.3 respectively.
[0036] Step 104, update the metaverse interaction model, and the update formula is as follows:
[0037]
[0038] β ∈ [0, 1], β represents the step size of deep learning, and ← represents passing the update;
[0039] Step 105, iterate steps 102-104 until the metaverse interaction model converges or the number of iterations reaches the set value.
[0040] The present invention provides a computer storage medium for storing computer-readable instructions that, when read, can execute the foregoing spatio-temporal AI data processing method in a blockchain-integrated metaverse scenario.
[0041] The beneficial effects of the present invention are as follows:
[0042] The present invention can achieve dynamic evolution in the interaction process between NPCs and users through the fusion processing of spatio-temporal data in combination with an AI model, change the attributes of NPCs and the interaction mode between NPCs and users, achieve the interaction effect of simulating the real world, and the change of NPC attributes can exceed the dynamic interaction of the real world. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] Figure 1 is a flowchart of a spatio-temporal AI data processing method in a blockchain-integrated metaverse scenario of the present invention. DETAILED DESCRIPTION OF THE INVENTION
[0044] Now, the subject matter described herein will be discussed with reference to exemplary embodiments. It should be understood that discussing these embodiments is only to enable those skilled in the art to better understand and thus implement the subject matter described herein, and changes can be made to the functions and arrangements of the elements discussed without departing from the scope of protection of the content of this specification. Each example can omit, substitute, or add various processes or components as needed. Additionally, the features described in some examples can also be combined in other examples.
[0045] In at least one embodiment of the present invention, a spatio-temporal AI data processing method in a blockchain-integrated metaverse scenario is disclosed, as Figure 1 shown, including the following steps:
[0046] Step 100, collect metaverse scenario information from the blockchain, where the metaverse scenario information includes the historical interaction data of users, the attribute information of NPCs, and the context information of the game environment;
[0047] Among them, the historical interaction data of users includes conversation content, behavior decisions, emotional expressions, etc.
[0048] The attribute information of NPCs, such as identity, personality, ability, etc.
[0049] The context information of the game environment, such as scenes, tasks, events, etc.
[0050] Step 200, generate the first two-dimensional structure data. The first two-dimensional structure data includes the first data matrix and the first relationship matrix. A cell of the first data matrix represents the first one-dimensional structure data of an independent object. The independent objects include users, NPCs, scenes, tasks, and interaction events. A cell of the first data matrix representing an independent object only contains the metaverse scene information of that independent object;
[0051] The element in the i-th row and j-th column of the first relationship matrix represents the association between the independent objects represented by the i-th cell and the j-th cell of the first data matrix. If there is an association, the value of this element in the first relationship matrix is 1, otherwise it is 0;
[0052] There is an association between an interaction event and the users, NPCs, scenes, and tasks associated with that interaction event;
[0053] There is an association between a user and the NPC with whom the user is interacting;
[0054] An NPC is associated with all tasks;
[0055] A first one-dimensional structure data includes n data items sorted by time. The t-th data item represents the metaverse scene information of the corresponding independent object collected at the t-th moment;
[0056] Step 300, input the first two-dimensional structure data into the metaverse interaction model. The metaverse interaction model includes a first intermediate layer, a second intermediate layer, a first output layer, and a second output layer. The first intermediate layer inputs the first one-dimensional structure data and outputs the first intermediate representation data to the second intermediate layer. The second intermediate layer also inputs the first relationship matrix. The second intermediate layer outputs the second intermediate representation data to the first output layer and the second output layer. The first output layer outputs the result representing the NPC interaction action, and the second output layer outputs the result representing the change in NPC attributes.
[0057] The calculation formula of the first intermediate layer of the metaverse interaction model is as follows:
[0058] h t =tanh(W hh h t-1 +W xh x t +b h )
[0059] Where h t represents the t-th first intermediate representation data output by the first intermediate layer, h t-1 represents the (t - 1)-th first intermediate representation data, x t represents the t-th data item of the first one-dimensional structure data, W hh and W xh are the first and second weight parameters, and b his the first bias parameter, and tanh is the hyperbolic tangent function.
[0060] In one embodiment of the present invention, the calculation formula of the second intermediate layer is as follows:
[0061]
[0062] where k v represents the second intermediate representation data of the v-th cell of the first data matrix, and M (v) is the set of cells associated with the v-th cell of the first data matrix, represents the n-th first intermediate representation data output when the first one-dimensional structure data of the v-th cell of the first data matrix is input to the first intermediate layer, c v is a normalization constant (default is 13), σ is the sigmoid function, and W k represents the third weight parameter, and b k is the second bias parameter.
[0063] In one embodiment of the present invention, the calculation formula of the first output layer is as follows:
[0064] OUT v = σ(W out * k v + b out ), v ∈ M (NPC)
[0065] where OUT v represents the first output vector of the NPC represented by the v-th cell of the first data matrix, and the c-th component value thereof represents the probability value of the c-th interaction action. The interaction action with the largest probability value is selected as the output. The interaction actions include the interaction behaviors between the NPC and the user and the tasks that the NPC can assign to the user. k v represents the second intermediate representation data of the v-th cell of the first data matrix, and M (NPC) represents the set of all cells representing the NPC in the first data matrix, W out is the first output weight parameter, and b out is the first output bias parameter, and σ represents the sigmoid function.
[0066] The calculation formula of the second output layer is as follows:
[0067] y v = σ(W y k v + b y ), v ∈ M (NPC)
[0068] where y vThe second output vector of the NPC represented by the v-th cell of the first data matrix, where the c-th component value represents the probability value of the c-th attribute strategy. The attribute strategy with the largest probability value is selected as the output. The attribute strategy includes various attribute values of the NPC, k v The second intermediate representation data of the v-th cell of the first data matrix, W y Is the second output weight parameter, b y Is the second output bias parameter, and σ represents the sigmoid function.
[0069] The steps of training the metaverse interaction model include:
[0070] Step 101, initialize the parameters of the metaverse interaction model;
[0071] Step 102, observe the metaverse scene information S at time t t , the strategy (including interaction actions and attribute strategies) A executed at time t t , the metaverse scene information S at time t + 1 t+1 , execute the strategy A t and obtain the reward R t ;
[0072] Step 103, then calculate the policy error:
[0073]
[0074] where δ t represents the policy error at time t, γ represents the discount factor, γ ∈ [0, 1], represents the sum of the maximum probability values in the first output vector and the second output vector output by the metaverse interaction model when inputting S t+1 , represents the sum of the probability values corresponding to the strategy A in the first output vector and the second output vector output by the metaverse interaction model when inputting S t ; t
[0075]
[0076] r q represents the average satisfaction degree of all users with the q-th NPC, b q represents the attribute change ratio of the q-th NPC, F t represents the per-unit value of the data processing volume when executing the strategy A t , and γ1, γ2, and γ3 are expected target weights, with default values of 0.6, 0.2, and 0.3 respectively.
[0077] Step 104, update the metaverse interaction model, and the update formula is as follows:
[0078]
[0079] β ∈ [0, 1], where β represents the step size of deep learning, and ← represents passing the update;
[0080] Step 105, iterate steps 102 - 104 until the metaverse interaction model converges or the number of iterations reaches the set value. The default value of this value is 30.
[0081] The present invention provides a computer storage medium for storing computer-readable instructions, which can execute the aforementioned spatio-temporal AI data processing method in a blockchain-integrated metaverse scenario when the computer-readable instructions are read.
[0082] The embodiments of the present invention have been described above, but these embodiments are not limited to the above specific implementation manners. The above specific implementation manners are merely illustrative rather than restrictive. Under the inspiration of this embodiment, those of ordinary skill in the art can also make more equivalent embodiments in various forms, all of which fall within the protection scope of this embodiment.
Claims
1. A method for processing spatiotemporal AI data in a blockchain-integrated metaverse scenario, characterized in that: The following steps are involved: Step 100, collecting metaverse scene information from the blockchain, the metaverse scene information includes the user's historical interaction data, NPC attribute information, and game environment context information; Step 200, generating No. 1 two-dimensional structure data, the No. 1 two-dimensional structure data includes No. 1 data matrix and No. 1 relationship matrix, one unit of the No. 1 data matrix represents No. 1 one-dimensional structure data of an independent object, the independent object includes a user, an NPC, a scene, a task and an interactive event, and one unit of the No. 1 data matrix representing an independent object only contains the metaverse scene information of the independent object; The element in the i-th row and j-th column of the No. 1 relationship matrix represents the association between the i-th unit of the No. 1 data matrix and the independent object represented by the j-th unit. If there is an association, the value of this element of the No. 1 relationship matrix is 1, otherwise it is 0; There is a correlation between the interaction event and the user, NPC, scene and task associated with the interaction event; There is a relationship between the user and the NPC they are interacting with; NPCs are associated with all quests; A number one one-dimensional structure data includes n data items sorted by time, and the tth data item represents the metaverse scene information of the corresponding independent object collected at the tth moment; Step 300, input the two-dimensional structure data No. 1 into the metaverse interaction model, the metaverse interaction model includes a first intermediate layer, a second intermediate layer, a first output layer and a second output layer, wherein the first intermediate layer inputs the one-dimensional structure data No. 1, outputs the first intermediate representation data to the second intermediate layer, the second intermediate layer also inputs the relationship matrix No. 1, the second intermediate layer outputs the second intermediate representation data to the first output layer and the second output layer, the first output layer outputs the result representing the NPC interaction action, and the second output layer outputs the result representing the NPC attribute change.
2. According to claim 1, a method for processing spatiotemporal AI data in a blockchain-integrated metaverse scenario is characterized in that: The user's historical interaction data, including conversation content, behavioral decisions, and emotional expressions.
3. According to claim 1, a method for processing spatiotemporal AI data in a blockchain-integrated metaverse scenario is characterized in that: NPC's attribute information includes identity, personality, ability, etc.
4. According to claim 1, a method for processing spatiotemporal AI data in a blockchain-integrated metaverse scenario is characterized in that: The context information of the game environment includes scenes, tasks, and events.
5. According to claim 1, a method for processing spatiotemporal AI data in a blockchain-integrated metaverse scenario is characterized in that: The calculation formula of the first intermediate layer of the Metaverse interaction model is as follows: h t =tanh(W hh h t-1 +W xh x t +b h ) where h t represents the tth first intermediate representation data output by the first intermediate layer, h t-1 represents the t-1th first intermediate representation data, x t represents the tth data item of the one-dimensional structure data, W hh and W xh are the first and second weight parameters, b h is the first bias parameter and tanh is the hyperbolic tangent function.
6. A method for processing spatiotemporal AI data in a blockchain-integrated metaverse scenario according to claim 5, characterized in that: The calculation formula for the second intermediate layer is as follows: where k v The second intermediate representation data of the vth unit of the data matrix No. 1, M (v) is the set of cells associated with the vth cell of data matrix 1, The nth first intermediate representation data output when the one-dimensional structure data of the vth unit of the data matrix is input into the first intermediate layer, c v is a normalization constant (default is 13), σ is the sigmoid function, W k represents the third weight parameter, b k is the second bias parameter.
7. A method for processing spatiotemporal AI data in a blockchain-integrated metaverse scenario according to claim 6, characterized in that: The calculation formula of the first output layer is as follows: OUT v =σ(W out *k v +b out ),v∈M (NPC) Where OUT v The first output vector of the NPC represented by the vth unit of the data matrix No. 1 is represented. Its cth component value represents the probability value of the cth interaction action. The interaction action with the largest probability value is selected as the output. The interaction action includes the interaction behavior between the NPC and the user and the task that the NPC can assign to the user. k v The second intermediate representation data of the vth unit of the data matrix No. 1, M (NPC) represents the set of all units of the NPC in the data matrix No. 1, W out is the first output weight parameter, b out is the first output bias parameter, and σ represents the sigmoid function.
8. The method for processing spatiotemporal AI data in a blockchain-integrated metaverse scenario according to claim 7 is characterized in that: The calculation formula of the second output layer is as follows: y v =σ(W y k v +b y ),v∈M (NPC) where y v The second output vector of the NPC represented by the vth unit of the data matrix No. 1, whose cth component value represents the probability value of the cth attribute strategy, selects the attribute strategy with the largest probability value as the output, and the attribute strategy includes the attribute values of the NPC, k v The second intermediate representation data of the vth unit of the data matrix No. 1, W y is the second output weight parameter, b y is the second output bias parameter, and σ represents the sigmoid function.
9. The method for processing spatiotemporal AI data in a blockchain-integrated metaverse scenario according to claim 1 is characterized in that: The steps of training the Metaverse interaction model include: Step 101, initializing the parameters of the metaverse interaction model; Step 102: Observe the metaverse scene information S at time t t , the strategy executed at time t (including interaction actions and attribute strategies) A t , Metaverse scene information S at time t+1 t+1 , execute strategy A t Rewards received R t ; Step 103, then calculate the policy error: where δ t represents the policy error at time t, γ represents the discount coefficient, γ∈[0,1], Represents the input S of the Metaverse interaction model t+1 The sum of the maximum probability values of the first output vector and the second output vector output when Represents the input S of the Metaverse interaction model t The first output vector and the second output vector corresponding to strategy A t The sum of the probability values of ; r q represents the mean satisfaction of all users with the qth NPC, b q Indicates the attribute change ratio of the qth NPC, F t Indicates execution strategy A t The per-unit value of the data processing amount at the time of γ1, γ2 and γ3 are the expected target weights, and their default values are 0.6, 0.2 and 0.3 respectively; Step 104, update the Metaverse interaction model, the updated formula is as follows: β∈[0,1], β represents the step size of deep learning, ← represents the transfer update; Step 105, iterate steps 102-104 until the metaverse interaction model converges or the number of iterations reaches a set value.
10. A computer storage medium, characterized in that: It is used to store computer-readable instructions, which, when read, can execute a spatiotemporal AI data processing method in a blockchain-fused metaverse scenario as described in any one of claims 1-9.