Robot state intelligent anchoring method, device, computer equipment and storage medium
By assigning unique identifiers to robots and utilizing multi-source data processing and the consensus adjudication mechanism of AI oracle nodes, the problem of reliable anchoring of robot states to digital assets is solved, realizing a reliable and interpretable mapping of robot states and automated business control.
Patent Information
- Application Number
- CN202611053057.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-15
- Publication Date
- 2026-11-10
AI Technical Summary
Existing technologies lack a reliable and stable mechanism to connect on-chain digital states with offline physical facts, resulting in distorted robot asset states, difficulty in forming unified access and settlement rules, severe separation between digital assets and physical systems, and difficulty in defining responsibilities.
By assigning a unique identifier to the robot, acquiring multi-source state data, preprocessing it to form a standardized evidence package, and using multiple AI oracle nodes for time-series analysis and consensus confirmation, the adjudication result is output, mapped to digital asset state, written into a trusted ledger, and combined with pre-set strategies for business event access determination.
It achieves a reliable anchoring of robot state to digital assets, reduces the risk of state distortion, ensures the reliability and accuracy of asset state, supports automated business control, and reduces the risk of digital actions deviating from physical reality.
Smart Images

Figure CN123074751A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of trusted digitalization and automated governance technology of Internet of Things assets, and in particular to a method, device, computer equipment and storage medium for intelligent anchoring of robot status. Background Technology
[0002] In robot trading platforms, robot assets are represented as a set of digital rights continuously bound to physical robots, including ownership, usage rights, lease rights, maintenance responsibilities, capability certificates, and performance certificates. The authenticity and validity of these dynamic rights depend on the accurate reflection of the robot's physical state. However, current technologies generally lack a reliable and stable mechanism to connect the on-chain digital state with offline physical facts.
[0003] On the one hand, the controller logs, sensor data, actuator feedback, battery status, positioning information, maintenance records, and external observation data generated by robots in real time are not only diverse in origin, frequency, and reliability, but also susceptible to interference from sensor drift, communication interruptions, missing records, human tampering, or forged playback. On the other hand, blockchain and digital asset systems can typically only process abstract state flags and events, and cannot directly understand raw physical data.
[0004] This is why, even with platforms collecting vast amounts of data, a crucial intermediate adjudication mechanism is still lacking to integrate multi-source, heterogeneous physical evidence into a single, clear, and auditable state conclusion, and further map it into a standard state object that the digital asset system can directly invoke. This deficiency often results in assets in the digital world having been transferred or settled, but the corresponding robots in the physical world lacking the actual conditions for delivery, leasing, or performance.
[0005] The typical problems currently faced include, but are not limited to: over-reliance on a single sensor or log leading to distorted state; lack of standardization of raw data, making it difficult to form unified access and settlement rules; lack of multi-source evidence fusion and state adjudication layer; severe separation between digital assets and physical systems; and difficulty in determining responsibility after a dispute occurs due to the lack of replayable and traceable evidence and version chains.
[0006] Therefore, how to transform multi-source physical evidence into interpretable robot state vectors, state levels, and causes of change, rather than merely remaining at the level of raw data, has become a core technical problem that urgently needs to be solved. Summary of the Invention
[0007] This invention provides a method, apparatus, computer device, and storage medium for intelligent anchoring of robot states, in order to solve the problem of the lack of a reliable and interpretable anchoring mechanism between the state of digital assets and the physical facts of robots.
[0008] A method for intelligent anchoring of robot state, comprising: Initialize the robot's identity: Assign a unique identifier to the robot and bind the unique identifier to the robot's hardware identity identifier, cryptographic security credentials, ownership and traceability metadata; Acquire multi-source state data of the robot; the multi-source state data includes internal state data and external observation data. The multi-source state data is preprocessed to form a standardized state evidence package; The state evidence package is input to multiple AI oracle nodes. The multiple AI oracle nodes perform time-series analysis, state adjudication, and consensus confirmation on events within a specified time window based on the state evidence package, historical state data, and a preset rule base, and output the adjudication result. The adjudication result includes a robot state vector and a state level. The robot state vector represents at least two of the following: hardware health, spatial location or work area state, and energy state. According to predefined mapping rules, the robot state vector and the state level are mapped to the digital asset state associated with the robot; Obtain the evidence digest of the state evidence package, write the digital asset state and the evidence digest into a trusted ledger to form an on-chain state object, and record the state version change relationship; When a business event involving the robot occurs, the digital asset status registered in the on-chain state object is invoked, and the business event is subject to access determination or automated control according to preset policy rules; wherein, the business event includes at least one of ownership transfer, lease handover, task settlement, maintenance responsibility switching or guarantee release; The process involves inputting the state evidence package into multiple AI oracle nodes, whereby the multiple AI oracle nodes, based on the state evidence package, historical state data, and a preset rule base, perform time-series analysis, state adjudication, and consensus confirmation on events within a specified time window, and output an adjudication result, including: Based on the collection timestamp, trusted time source, device serial number, and message causal relationship in the state evidence package, the temporal relationship of events within the specified time window is constructed; Based on the aforementioned temporal relationship, deterministic rule verification, machine learning model inference, and historical window analysis are performed in parallel to obtain node analysis results and generate candidate state conclusions. By using the Byzantine fault-tolerant consensus mechanism, candidate state conclusions generated by multiple AI oracle nodes are exchanged and verified. When more than a preset threshold of nodes reach a consensus on the same candidate state conclusion, it is confirmed as the ruling result.
[0009] Optionally, the internal status data includes at least one of controller logs, sensor data, actuator feedback, current, voltage, temperature, task progress, and fault codes; The external observation data includes at least one of the following: data reported by the positioning base station, camera verification results, access control records, maintenance work order records, and results uploaded by third-party detection nodes; The preprocessing includes time alignment, format standardization, source identification, signature verification, and outlier removal.
[0010] Optionally, the candidate state conclusion may further include a comprehensive confidence level and a causal explanation; The deterministic rules are used to verify the logical consistency, legality, and temporal validity of evidence; the machine learning model is used to identify operating states and abnormal patterns; and the historical window analysis is used to assess the continuity of state changes and identify unreasonable mutations.
[0011] Optionally, the digital asset status is mapped to a status label including at least one of tradable, leaseable, deliverable, frozen, pending review, pending repair, and dispute pending.
[0012] Optionally, it also includes: During the state adjudication process through the AI oracle node, if it is identified that the robot's identity is inconsistent, its location is abnormal, its components are seriously faulty, its security status is abnormal, the credibility of the evidence is below the threshold, or there is a substantial conflict between multiple sources of evidence, then the digital asset status in the adjudication result will be set to a frozen state, a pending review state, or a dispute pending state.
[0013] Optionally, it also includes: In response to a dispute query request, trusted audit data within a specified time window is obtained; the trusted audit data includes the status evidence package, the status adjudication log of the AI oracle node, the version record of the digital asset status, and related business action records; Based on the trusted audit data, extract the smallest verifiable evidence unit and / or key anomaly coordinates that led to the state change, and replay and verify the event sequence.
[0014] Optionally, the step of invoking the digital asset status registered in the on-chain state object and performing access determination or automated control on the business event according to preset policy rules includes: Retrieve the digital asset state from the on-chain state object and obtain its contained state level and state label; Obtain the status level threshold and status label admission conditions associated with business events; The status level is compared with the status level threshold, and the status label is verified to meet the status label admission conditions of the current business event. If the status level is lower than the status level threshold, or if the status label does not meet the status label admission conditions for the current business event, the execution process of the business event will be automatically blocked.
[0015] A robot state intelligent anchoring device, comprising: The identity registration and binding module is used to initialize the robot's identity: assigning a unique identifier to the robot and binding the unique identifier to the robot's hardware identity identifier, cryptographic security credentials, ownership and traceability metadata; A state data acquisition module is used to acquire multi-source state data of the robot; the multi-source state data includes internal state data and external observation data. An evidence standardization and encapsulation module is used to preprocess the multi-source state data to form a standardized state evidence package; The time-series analysis and consensus adjudication module is used to input the state evidence package to multiple AI oracle nodes. The multiple AI oracle nodes perform time-series analysis, state adjudication, and consensus confirmation on events within a specified time window based on the state evidence package, historical state data, and a preset rule base, and output the adjudication result. The adjudication result includes a robot state vector and a state level. The robot state vector represents at least two of the following: hardware health, spatial location or work area state, and energy state. The digital asset state mapping module is used to map the robot state vector and the state level to the digital asset state associated with the robot according to predefined mapping rules. The on-chain evidence storage and version control module is used to obtain the evidence digest of the state evidence package, write the digital asset state and the evidence digest into a trusted ledger to form an on-chain state object, and record the state version change relationship. The state-driven business control module is used to call the digital asset status registered in the on-chain state object when a business event involving the robot occurs, and to perform access judgment or automated control on the business event according to the preset policy rules; wherein, the business event includes at least one of ownership transfer, lease handover, task settlement, maintenance responsibility switching or guarantee release; The timing analysis and consensus adjudication module is also used for: Based on the collection timestamp, trusted time source, device serial number, and message causal relationship in the state evidence package, the temporal relationship of events within the specified time window is constructed; Based on the aforementioned temporal relationship, deterministic rule verification, machine learning model inference, and historical window analysis are performed in parallel to obtain node analysis results and generate candidate state conclusions. By using the Byzantine fault-tolerant consensus mechanism, candidate state conclusions generated by multiple AI oracle nodes are exchanged and verified. When more than a preset threshold of nodes reach a consensus on the same candidate state conclusion, it is confirmed as the ruling result.
[0016] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described intelligent robot state anchoring method.
[0017] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described intelligent robot state anchoring method.
[0018] The aforementioned intelligent anchoring method, device, computer equipment, and storage medium for robot states establish a stable one-to-one correspondence between the robot entity and digital assets by binding a unique identifier to the robot and acquiring multi-source state data. This enables the transformation of physical entities into computable and writable blockchain digital asset objects. Combined with time-series analysis and consensus adjudication mechanisms using multiple AI oracle nodes, it effectively reduces the risk of state distortion caused by single-point forgery, log tampering, or sensor drift, ensuring a high degree of trustworthiness of the asset state. By writing the digital asset state and evidence digest into a trusted ledger to form on-chain state objects, and using pre-defined strategies to determine the access of business events such as transfers, leases, and settlements, it achieves automatic updates and constraints of digital actions based on physical states, greatly reducing the risk of digital actions deviating from physical reality. This invention solves the problem of the lack of a trustworthy and interpretable anchoring mechanism between digital asset states and robot physical facts. Attached Figure Description
[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0020] Figure 1 This is a flowchart of a robot state intelligent anchoring method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a robot state intelligent anchoring device in one embodiment of the present invention; Figure 3 This is a schematic diagram of a computer device according to an embodiment of the present invention. Detailed Implementation
[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0022] In one embodiment, such as Figure 1 As shown, a robot state intelligent anchoring method is provided, including the following steps S10~S70.
[0023] S10. Initialize the robot's identity: Assign a unique identifier to the robot and bind the unique identifier to the robot's hardware identity identifier, cryptographic security credentials, ownership and traceability metadata.
[0024] S20. Obtain multi-source state data of the robot; the multi-source state data includes internal state data and external observation data.
[0025] Understandably, in this embodiment, a six-layer architecture is preferably adopted, namely: a robot-side trusted data collection layer, an evidence aggregation and standardization layer, an AI oracle adjudication layer, a digital asset mapping layer, an on-chain registration and version control layer, and a transaction and governance interface layer. These layers can be centrally deployed on the same server or distributed across the robot, edge nodes, platform servers, and blockchain nodes. Steps S10 and S20 are primarily executed by the robot-side trusted data collection layer. It should be noted that the robot described in this embodiment is not limited to the complete robot body, but can also be used for digital asset objects such as key robot components, robot operational capability certificates, robot maintenance certificates, task fulfillment certificates, and robot logistics delivery status certificates.
[0026] Specifically, in step S10, the robot's identity is initialized. First, a globally unique Robot Identity Identifier (RID) is assigned to the robot. That is, the Robot Identity Identifier is an identifier used to uniquely identify the robot entity. Subsequently, this RID is bound to the robot's hardware identity identifier (such as the main control board serial number, key component serial number), cryptographic security credentials (such as a trusted hardware public key), and ownership and traceability metadata (such as manufacturer identifier, model, manufacturing date, and initial owner), together forming an auditable and tamper-proof digital identity object. This identity object is securely stored and serves as the root of trust anchored to all subsequent state data, ensuring that any digital asset and operation record can be uniquely and reliably traced back to this specific physical robot entity.
[0027] In step S20, the robot's multi-source state data is continuously or on-demand acquired through the robot-side trusted acquisition layer. Specifically, the multi-source state data includes internal state data acquired from within the robot itself and external observation data acquired from independent nodes. In some examples, data acquisition covers three main sources: perception data sources, including various sensor readings, positioning system (such as GPS / BeiDou) information, and third-party observation data from external observation nodes (such as cameras, access control); decision and control data sources, including controller internal logs, decision command streams, and actual action feedback from actuators; and maintenance data sources, including voltage, current, and temperature information from the battery management system (BMS), as well as health status and alarm information reported by the maintenance system.
[0028] To ensure the authenticity and integrity of the collected data, the acquisition layer preferably adds acquisition timestamps, device signatures, or hardware proofs based on a Trusted Execution Environment (TEE) to key data while acquiring the raw data. This layer extracts comprehensive and heterogeneous runtime evidence from the robot itself and its operating environment, providing raw materials for the subsequent formation of standardized state evidence packages.
[0029] The technical effects of steps S10 and S20 are as follows: by establishing a unique and tamper-proof identity anchor, it ensures that all subsequent state data and assets can be accurately and reliably mapped back to the specific physical robot entity, eliminating identity impersonation and confusion; at the same time, by systematically collecting multi-source runtime data from dimensions such as perception, decision-making, and protection, a complete and original evidence set covering the entire "sensing-control-management" link of the robot is constructed for subsequent state adjudication, laying a data foundation for highly reliable state assessment.
[0030] Optionally, the internal status data includes at least one of controller logs, sensor data, actuator feedback, current, voltage, temperature, task progress, and fault codes; The external observation data includes at least one of the following: data reported by the positioning base station, camera verification results, access control records, maintenance work order records, and results uploaded by third-party detection nodes; Understandably, multi-source state data specifically includes internal state data collected from within the robot itself and external observation data collected from independent nodes. Internal state data can be selected from controller logs, sensor data such as joint angles / wheel speeds, current / voltage / temperature readings, task progress, fault codes, and serial numbers of critical components. External observation data can be selected from data reported by positioning base stations, visual verification results from cameras, access records from the site access control system, initiated maintenance work orders, or results uploaded by third-party detection nodes. In actual system deployment, at least two of the aforementioned internal and external data sets are selected for combined collection to constitute cross-validation of the robot's state.
[0031] This embodiment constructs a cross-validation mechanism that combines internal and external data by simultaneously collecting internal state data generated by the robot itself and external observation data from independent nodes, fundamentally improving the authenticity, tamper resistance, and credibility of state evidence.
[0032] S30. The multi-source state data is preprocessed to form a standardized state evidence package.
[0033] Understandably, the preprocessing process first performs time alignment and cleaning operations on the acquired multi-source heterogeneous state data. The system calibrates data streams with different sampling frequencies or transmission delays by introducing a unified time synchronization benchmark, and performs necessary denoising and outlier removal on the original signal to eliminate high-frequency noise caused by physical environmental interference or communication jitter, ensuring the temporal consistency and purity of the input data.
[0034] Subsequently, the preprocessing module performs unified format encoding and source identifier mapping on the cleaned data. The system converts data collected using different protocols into a standard machine-readable format and assigns a clear source label to each data point using preset rules. Based on this, and combining data verification status and context hashing, the system calculates a preliminary credibility score for each piece of evidence according to a preset weighting strategy, ultimately encapsulating it into a standardized state evidence package containing complete metadata and the preliminary credibility score. It refers to a standardized object formed within time t or window T. In some examples, the state evidence package... This includes the following fields: robot identifier, data collection source identifier, evidence category, data collection time, original value, standard value, unit, context hash, signature status, data collection link identifier, and initial credibility score. Step S30 can be executed at the evidence collection and standardization layer.
[0035] This embodiment effectively solves the analysis obstacles caused by the dispersed data sources, inconsistent formats, misaligned time sequences, and varying quality by performing standardized preprocessing on multi-source, heterogeneous raw state data. It not only eliminates the consistency problem at the data level, but also assigns a preliminary credibility score to each piece of evidence, thereby laying a unified and high-quality input foundation for the subsequent automated and quantifiable state adjudication of AI oracles.
[0036] Optionally, the preprocessing includes time alignment, format standardization, source identification, signature verification, and outlier removal.
[0037] Understandably, firstly, data from different acquisition nodes can be time-aligned based on a high-precision time source, uniformly mapped to a standard time axis, and the data format can be uniformly encoded to eliminate protocol heterogeneity and ensure format uniformity. Simultaneously, a unique source identifier for the generating device or system is appended to each piece of data.
[0038] Subsequently, the system performs signature verification on the digitally signed data to confirm its integrity and the sender's identity, and performs outlier detection and removal on the data stream based on statistical methods or rule bases. These steps together ensure the temporal consistency, format standardization, traceability, security, and data quality of the input data, laying the foundation for generating standardized state evidence packages.
[0039] S40. The state evidence package is input to multiple AI oracle nodes. The multiple AI oracle nodes perform time-series analysis, state adjudication, and consensus confirmation on events within a specified time window based on the state evidence package, historical state data, and a preset rule base, and output the adjudication result. The adjudication result includes a robot state vector and a state level. The robot state vector represents at least two of the following: hardware health, spatial location or work area state, and energy state.
[0040] Understandably, step S40 can be executed at the AI oracle adjudication layer. Standardized state evidence packages can be used. Historical state data and a pre-defined rule base are used as inputs and distributed to a distributed network consisting of multiple AI oracle nodes. The core function of each AI oracle node is to adjudicate multi-source physical state evidence as digital asset state, and its implementation is not limited to a single technical path.
[0041] In a preferred embodiment of the present invention, each node is not a single model, but a composite adjudication system integrating event sequencing, deterministic rules, an analysis and inference engine, and multi-node consensus. The analysis and inference engine can employ machine learning models, temporal models, graph models, anomaly detection models, large language models, multi-agent decision-making mechanisms, or any combination thereof, to achieve the identification and state inference of complex patterns. The core task of this system is to generate a credible, interpretable, and consensus-confirmed adjudication result for multi-source, potentially conflicting evidence within a specified time window. This result must at least include the robot's state vector. (Describes multi-dimensional states such as operation, health, and location) and status levels (e.g., excellent, caution, faulty, etc.), and may optionally include overall reliability. and explanation of the reasons The robot state vector represents at least two of the following: hardware health, spatial location or work area state, and energy state.
[0042] In one example, the robot's state vector It can be represented as:
[0043] in, This indicates consistency in identity; Indicates hardware health status; Indicates availability; Indicates the location or status of the work area; Indicates energy state; Indicates the status of task fulfillment; Indicates the maintenance lifecycle status; Indicates the safety and compliance status; This indicates the quality and credibility of the evidence. It should be understood that the above component composition, naming, and threshold calculation methods are merely non-limiting examples. In practical applications, the dimensions of the state vector, the specific meaning of each component, the calculation method, and the threshold can all be adjusted and expanded according to the type of robot, application scenario, and adjudication requirements. As long as its core function is to provide a multi-dimensional, unified, and quantitative expression of the robot's true operating state, it falls within the scope of protection of this invention.
[0044] The adjudication process begins with chronological sequencing of events. When conflicts exist within an evidence package, the system integrates the collection timestamps, reliable time sources, device serial numbers, message causal relationships, and preset priorities of the evidence sources to construct the temporal relationship of all events within that time window. This aims to identify the order and concurrency of events and detect temporal anomalies caused by data playback or tampering, establishing a unified and reliable time benchmark for subsequent analysis.
[0045] Subsequently, each oracle node executes multi-engine collaborative adjudication in parallel based on the sorted event sequence. Deterministic rules are used to quickly filter and exclude obviously invalid evidence (such as signature verification failures, time reversals, and missing required fields); machine learning models are responsible for in-depth analysis of valid evidence to identify complex patterns such as health status, failure modes, and task completion rates; historical window analysis compares current evidence with recent historical states to determine the continuity of state changes and identify unreasonable mutations. Each node generates its own candidate state conclusion based on this analysis.
[0046] Finally, the decision can be definitively confirmed using a Byzantine fault-tolerant consensus mechanism. Each node broadcasts its locally generated candidate conclusions to the entire network, and through multiple rounds of message interaction and verification, a consensus is reached on the unique final decision within the distributed network. This mechanism ensures that even if some nodes fail or act maliciously, the entire system can still output correct and consistent decisions, thus combining the intelligent judgment of AI oracle nodes with the fault-tolerant reliability of a distributed system.
[0047] Optionally, step S40, namely, inputting the state evidence package to multiple AI oracle nodes, whereby the multiple AI oracle nodes, based on the state evidence package, historical state data, and a preset rule base, perform time-series analysis, state adjudication, and consensus confirmation on events within a specified time window, and output an adjudication result, including: S401. Based on the collection timestamp, trusted time source, device serial number, and message causal relationship in the state evidence package, construct the temporal relationship of events within the specified time window; S402. Based on the aforementioned temporal relationship, perform deterministic rule verification, machine learning model inference, and historical window analysis in parallel to obtain node analysis results and generate candidate state conclusions. S403. Through the Byzantine fault-tolerant consensus mechanism, candidate state conclusions generated by multiple AI oracle nodes are exchanged and verified. When more than a preset threshold of nodes reach a consensus on the same candidate state conclusion, it is confirmed as the ruling result.
[0048] Understandably, in step S401, event timing sorting is performed first. When the state evidence package... When the data contains potentially conflicting evidence from multiple sources such as controllers and external cameras, the temporal sequence of all events within a specified time window is constructed by integrating the acquisition timestamp of each piece of evidence, the reliable time source (such as the signal from the National Time Service Center), the serial number of the device that generated the data, and the causal relationship between messages (such as alarm-triggered task abort). The core purpose of this step is to form a logically consistent sequence of events, thereby clearly identifying which events occurred first, which occurred later, and which were concurrent events. It also enables the detection of temporal anomalies inserted due to data replay or malicious tampering, laying a reliable time baseline for subsequent in-depth analysis.
[0049] In step S402, each AI oracle node performs multi-engine collaborative analysis in parallel based on the aforementioned event sequence. First, deterministic rules are applied to quickly filter and verify the evidence. This rule base consists of explicit logical conditions used to eliminate invalid evidence such as invalid signatures, reversed time sequences, missing required fields, or evidence that clearly conflicts with physical laws. Then, valid evidence is input into a machine learning model for inference. This model, after training, can identify complex patterns such as device health status, potential failure modes, and task completion rates from the data. Simultaneously, historical window analysis is performed, comparing the current event sequence with recent historical state data to assess whether the robot's state changes smoothly and continuously, and whether there are any illogical abrupt changes. Each node integrates the results of these three aspects of node analysis to form a preliminary, comprehensive judgment of the robot's current state.
[0050] Each AI oracle node generates a candidate state conclusion based on its independent node analysis results. This conclusion is a structured decision output that must contain the robot's state vector. (A multidimensional vector used to quantify the robot's operation, health, position, and other multidimensional states), comprehensive credibility. (Overall credibility score of this ruling calculated based on the quality of evidence and the degree of conflict), Status Level (e.g., tiered labels such as "Normal," "Warning," "Fault," etc.) and an explanation of the reasons for reaching this conclusion. (For example, indicate which key sensor data was used or which conflicting evidence was excluded).
[0051] In step S403, the candidate state conclusions generated by all nodes are finally confirmed through a Byzantine fault-tolerant consensus mechanism. Each node broadcasts its candidate conclusions to the oracle network, and the consistency of the conclusions is checked through multiple rounds of message exchange and mutual verification. This mechanism can tolerate a certain number of node failures, delays, or malicious behavior. Only when more than a preset threshold (e.g., more than two-thirds) of the valid nodes in the network reach a consensus on the same specific state conclusion will the conclusion be confirmed as the final and tamper-proof decision. This step ensures the objectivity of the decision and resistance to single points of fraud, achieving a deep integration of AI intelligent judgment and the fault-tolerant reliability of distributed systems.
[0052] This embodiment establishes a rigorous event timeline, providing a unified and reliable time baseline for subsequent analysis and effectively resolving the temporal conflicts and discrepancies of multi-source heterogeneous evidence. By combining deterministic rules, machine learning models, and historical window analysis into a collaborative adjudication mechanism, it achieves rapid verification of evidence, complex pattern recognition, and assessment of the continuity of change, significantly improving the comprehensiveness and accuracy of state determination. Finally, by leveraging the Byzantine fault-tolerant consensus mechanism to confirm the consistency of conclusions from distributed nodes, it ensures the objectivity, reliability, and resilience against single-point malicious attacks or failures in the final adjudication result.
[0053] S50. According to predefined mapping rules, map the robot state vector and the state level to the digital asset state associated with the robot.
[0054] Understandably, a predefined mapping rule can be represented by a predefined mapping function M. Through the mapping function M, the robot state vector output by the AI oracle node can be mapped... With status level This is transformed into a digital asset state that is associated with the robot and can be directly used at the business level. Digital asset status The functional relationship can be expressed as: ,in, This represents the set of scenario strategy rules. Indicates overall credibility. Scenario strategy rule set. It is a set of pre-configured logical rules that are bound to specific business scenarios (such as leasing, transactions, and guarantees). It defines the conditions for mapping which technical state to which business state.
[0055] The core logic of the mapping function M is based on The rules in the middle, for , and A comprehensive judgment is made. For example, a leasing strategy rule might stipulate that: when the identity consistency identifier in the state vector... Health score Usage rate Location compliance and safety status All are above the set threshold, and the status level is... When the rating is "Excellent," the corresponding digital asset status for the robot's usage rights should be mapped to "Leaseable." Conversely, if... Inconsistent identity or key quality indicators If the value is below the threshold, the asset status may be directly mapped to "frozen" or "pending review".
[0056] Mapped output digital asset state object It is a structured data object that not only contains status labels such as "leasable" and "frozen," but also rich fields such as status effective time, status version number, applicable business scenario, set of prohibited / allowed actions, summary reference of the evidence package on which it is based, and a flag indicating whether manual review is required. This object encapsulates the underlying, multi-dimensional technical status parameters into a high-level status assertion that is understandable and directly processed by the business system. Step S50 can be executed at the digital asset mapping layer.
[0057] This embodiment realizes the transformation from the state description of the "physical-information space" to the state definition of the "digital-asset space". Through this mapping, the complex operation of the robot is abstracted into a clear and definite asset state, thereby providing a unique and authoritative input basis for subsequent on-chain registration and state-based automated business control (such as automatic authorization of transactions or triggering insurance claims).
[0058] Optionally, the digital asset status is mapped to a status label including at least one of tradable, leaseable, deliverable, frozen, pending review, pending repair, and dispute pending.
[0059] Understandably, in one alternative implementation, the state of digital assets... These are specifically mapped to a set of explicit state labels with business semantics. These labels constitute a direct definition of the robot's related asset permissions and availability, and their specific values include, but are not limited to, at least one of the following: tradable, leaseable, deliverable, frozen, pending review, pending maintenance, and dispute pending. Each state label is mapped by the mapping function M based on the robot's state vector. Status Level Overall credibility and specific scenario strategy rule sets The calculations are directly linked to different business control logics. For example, a leaseable state allows the initiation of a lease contract, while a frozen state prohibits any transfer of ownership or usage rights.
[0060] S60. Obtain the evidence digest of the state evidence package, write the digital asset state and the evidence digest into a trusted ledger to form an on-chain state object, and record the state version change relationship.
[0061] Understandably, an evidence digest (such as a hash value) of the standardized state evidence package can be obtained, and this digest can be compared with the digital asset state generated in step S50. Together, they are written to a trusted ledger. Here, a trusted ledger refers to a data storage system with immutable and traceable characteristics, including blockchain, consortium blockchain, or auditable distributed ledger. After being written, the two are jointly encapsulated into a new on-chain state object, which serves as the authoritative and immutable anchor point for the robot's asset state in the digital space.
[0062] Simultaneously, the state version change relationships need to be recorded. Specifically, when creating a new on-chain state object, the system generates an incrementing state version number for it and explicitly records its predecessor state reference (such as a hash value or version identifier pointing to the previous state object). Furthermore, metadata such as update time and update trigger reason can also be recorded. In this way, all historical state changes form a complete and verifiable version chain. As a preferred solution to optimize storage costs, the original full state evidence package... Evidence is typically stored off-chain in a repository, with only an immutable summary stored on-chain for verification. Step S60 can be performed on-chain at the registration and version control layer.
[0063] In some examples, the state of digital assets is determined before being written to the final trusted ledger. The evidence summary can first be submitted to a state cache and review layer. This layer allows authorized nodes (such as regulators or asset custodians) to manually review or audit the ruling when necessary. Only when the review is passed or the cache times out will the state be formally and irreversibly written to the target trusted ledger, thereby ensuring finality while adding a layer of risk control for critical operations.
[0064] In some examples, the state of digital assets It can serve as a standardized input for multi-layered real-time monitoring interfaces. Specifically, platform monitoring nodes, industry monitoring nodes, regional monitoring nodes, or other RegTech systems (regulatory technology systems) can, based on... It can efficiently perform penetrating audits, risk warnings, compliance verifications, and accountability tracking by controlling version iterations, anomaly markers, freeze markers, review markers, and evidence summary citations, without having to directly parse all the original robot data.
[0065] In addition, the state of digital assets It can also serve as a standardized input for the adaptive update process of contract rules. For example, when When specific device health declines, evidence quality deteriorates, specific anomaly patterns occur frequently, or regulatory policies change within multiple consecutive time windows, the system can automatically trigger contract rule version switching, parameter updates, supplementary review, or a self-evolving verification process within a sandbox. Preferably, before such updates officially take effect, they still need to undergo preset verification, simulation testing, permission approval, or manual review to ensure that adaptive updates are strictly based on verifiable state facts.
[0066] S70. When a business event involving the robot occurs, the digital asset status registered in the on-chain state object is invoked, and the business event is subject to access determination or automated control according to preset policy rules; wherein, the business event includes at least one of ownership transfer, lease handover, task settlement, maintenance responsibility switching or guarantee release.
[0067] Understandably, step S70 can be executed at the transaction and governance interface layer. A state-driven business control mechanism can be adopted. When a critical business event involving the bot occurs in the platform, the control logic will proactively query and retrieve the state of the digital assets stored on the chain. This will serve as the highest authoritative basis for determining access. Here, a business event can refer to a key stage in the change of ownership or right to use robot assets, including at least one of the following: ownership transfer, lease handover, task settlement, switch of maintenance responsibilities, and release of guarantees.
[0068] The core logic of access determination is to assess the status of digital assets based on pre-defined policy rules for different business scenarios. The encapsulated specific state assertion is validated. This state assertion originates from the state vector. The mapping results of each component, including identity consistency Hardware health Location or work area status Energy state Task performance status Maintain lifecycle status Safety and compliance status and the quality and credibility of evidence For example, in cases of ownership transfer, the strategy rule might require... Must be "consistent" Higher than the safety threshold, and No abnormalities, at this time the corresponding The system will only allow the transaction process to continue if the status label is "tradable".
[0069] The system performs differentiated judgments for different business events. For lease handover (start), the strategy not only checks whether the robot is in a "leasable" state, but also further verifies its current energy status. Does it meet the requirements for mission endurance and location? Is it within the permitted area? If so... If the delivery is insufficient, the system can automatically trigger a "pause delivery and prompt for charging" action. For task settlement, the system will strictly adhere to the task fulfillment status recorded on the blockchain. In combination with the quality of evidence Make a judgment: if "Completed" and If it is "high", payment settlement will be automatically triggered; if If the status is "under dispute", the settlement process will be automatically suspended and a manual review will be notified.
[0070] Through the above mechanism, the execution of any business action is based on the real-time and reliable on-chain asset status, thereby achieving a strong binding and precise synchronization between business flow and the robot's actual operating status, security status, and compliance. While improving automation efficiency, it effectively prevents asset and security risks caused by misoperation when the device status does not meet the conditions.
[0071] Steps S10-S70 establish a stable one-to-one correspondence between the robot entity and digital assets by binding a unique identifier to the robot and acquiring multi-source state data. This transforms the physical entity into a computable, writable blockchain digital asset object. Combined with the time-series analysis and consensus adjudication mechanism of multiple AI oracle nodes, the risk of state distortion caused by single-point forgery, log tampering, or sensor drift is effectively reduced, ensuring high reliability of the asset state. The digital asset state and evidence digest are written into a trusted ledger to form an on-chain state object. Pre-defined strategies are used to determine the accessibility of business events such as transfers, leases, and settlements, enabling automatic updates and constraints of digital actions based on physical state, significantly reducing the risk of digital actions deviating from physical reality. This embodiment addresses the problem of the lack of a reliable and interpretable anchoring mechanism between the digital asset state and the robot's physical reality.
[0072] Optionally, step S70, which involves calling the digital asset status registered in the on-chain state object and performing admission determination or automated control on the business event according to preset policy rules, includes: S701. Retrieve the digital asset status from the on-chain state object and obtain its state level and state label; S702. Obtain the status level threshold and status label admission conditions associated with business events; S703. Compare the status level with the status level threshold, and verify whether the status label meets the status label admission conditions of the current business event. S704. If the status level is lower than the status level threshold, or the status label does not meet the status label admission conditions of the current business event, the execution process of the business event is automatically blocked.
[0073] Understandably, in step S701, when the system detects a key business event involving the robot (such as ownership transfer, lease handover, task settlement, etc.) being triggered, it first initiates a query to the trusted ledger to extract the current digital asset state from the corresponding on-chain state object. Because this state object not only contains macro-level state labels but also encapsulates the asset's state level at a specific moment, this state level is a quantitative assessment of the robot's overall operational health, safety compliance, and the credibility of evidence. The state labels are macro-level business status identifiers such as "tradeable" and "leasable." The system must first obtain this state level and state labels as the basis for subsequent judgments.
[0074] Subsequently, in step S702, the system dynamically matches and retrieves the state level threshold and state label admission conditions set in the preset policy rules based on the type of the currently occurring business event. Different business scenarios have different requirements: for example, in the "ownership transfer" scenario, to protect asset value, the policy rules will focus on verifying identity consistency. Hardware health Safety and compliance status and the quality and credibility of evidence The first scenario sets a higher threshold requirement and may require the status label to be "clear ownership"; while the "right to use lease" scenario focuses more on availability. (Idle, online, available for order taking, or operational status), battery level and energy supply status. Maintain lifecycle status and the status of the work area These indicators may be required, and the status label may need to be "idle" or "available".
[0075] In step S703, the system accurately compares the status level obtained in step S701 with the threshold and verifies whether the obtained status label meets the status label admission conditions of the current business event, so as to assess whether the current asset meets the business execution conditions.
[0076] After comparison and verification, if the current status level fails to meet the threshold required by the business event (i.e., the status level is below the threshold), or the current status label does not meet the access conditions, the system will automatically trigger the risk control interception mechanism to suspend or block the subsequent execution process of the business event. For example, if the robot's current safety and compliance status... An anomaly causing the overall status level to be low will result in the system refusing access and maintaining or changing the asset to a "frozen" state when a "transfer of ownership" request occurs; in the "task settlement" scenario, if the task fulfillment status... or quality of evidence If the requirements are not met, the system will block fund settlement and suspend the process for manual review. Only when the status level meets or exceeds the threshold and the status label fully meets the admission criteria will the business process be allowed to proceed, thus ensuring that every operation is carried out in a real and compliant asset state.
[0077] This embodiment constructs a real-time and reliable automated business access gate based on objective data by calling the state and state level of immutable digital assets on the blockchain in real time and automatically comparing them with the state level threshold of specific business scenarios. Through this mechanism, the feasibility of business flow is directly and dynamically anchored to the robot's current real operating state, realizing state-driven decision-making. When the state level does not meet the security threshold, the system can instantly and automatically block high-risk business operations, thereby effectively preventing asset losses and compliance risks caused by critical operations such as ownership transfer and fund settlement when the equipment is in an abnormal state, there is insufficient evidence, or there are security risks.
[0078] Optionally, it also includes: S80. During the state adjudication process through the AI oracle node, if it is identified that the robot's identity is inconsistent, its location is abnormal, its components are seriously faulty, its security status is abnormal, the credibility of the evidence is lower than the threshold, or there is a substantial conflict between multiple sources of evidence, then the digital asset status in the adjudication result is set to frozen status, pending review status, or dispute pending status.
[0079] Understandably, when AI oracle nodes execute decisions based on state evidence packages, they monitor core risk dimensions in real time according to a pre-defined rule base. Once any pre-defined anomaly is identified, such as inconsistency in identity (e.g., hardware identifier does not match the on-chain RID binding), abnormal location (deviating from the pre-defined geofence), major component failure (critical sensors or actuators report unrecoverable errors), abnormal security status (detection of physical collisions or unauthorized instructions), evidence credibility below a threshold (e.g., missing key signatures), or substantial conflict between multiple sources of evidence, the system will immediately determine that it is currently in a high-risk or uncertain state.
[0080] At this point, the AI oracle node will include the state of the digital asset in the ruling result. The status can be directly set to "Frozen," "Pending Review," or "Dispute Suspended." The "Frozen" status is a strict restrictive state that immediately triggers the business control layer (S70) to halt all sensitive operations such as ownership transfer and leasing. The "Pending Review" status is a transitional state, allowing only limited operation during authorized personnel review and suspending automated business transactions. The "Dispute Suspended" status is applicable to disputes that have entered formal arbitration or judicial proceedings; in this state, all changes in asset ownership and high-value business operations are strictly locked until the dispute is resolved.
[0081] This embodiment, as a key link in risk closed-loop control, can transform core anomalies discovered during the adjudication process into protective actions at the asset level in real time and automatically, thereby immediately triggering a circuit breaker at the business level after a technical risk occurs, effectively preventing the expansion of losses or improper status transitions.
[0082] Optionally, it also includes: S901. In response to a dispute query request, obtain trusted audit data within a specified time window; the trusted audit data includes the status evidence package, the status adjudication log of the AI oracle node, the version record of the digital asset status, and related business action records; S902. Based on the trusted audit data, extract the smallest verifiable evidence unit and / or key anomaly coordinates that caused the state change, and replay and verify the event sequence.
[0083] Understandably, in step S901, the system responds to a dispute inquiry request from an arbitrator, regulator, or asset-related party. This request must specify the time window involved in the dispute. Based on this time window, the system retrieves and aggregates complete trusted audit data from the trusted ledger (blockchain) and its associated off-chain evidence repository. The trusted audit data is not the raw data stream, but rather a key structured record constituting a verifiable historical trajectory, specifically including: a state evidence package that triggered the ruling. This includes its cryptographic index, AI oracle node state adjudication logs recording the consensus process, digital asset state version records documenting all historical changes (i.e., state version chains), and records of related business actions driven by state (such as admission decisions and freeze orders). These data are interlocked through cryptographic hashes, forming a complete, immutable, and trustworthy audit trail from evidence to state to business actions.
[0084] In step S902, instead of inefficiently replaying all the data, the system intelligently analyzes the audit trail to pinpoint abrupt changes in state. Specifically, the system analyzes state version records to identify key shifts in the state of digital assets (e.g., from...). Become The system then extracts the minimum verifiable unit of evidence or key anomaly coordinates that led to the change from the adjudication log and state evidence package associated with that version. The minimum verifiable unit of evidence refers to the most concise subset of evidence that can independently and completely prove the rationality of the state change, such as a location boundary violation report confirmed by multiple nodes. Key anomaly coordinates are metadata used to precisely locate the problem and may include the time of the anomaly event, the identifier of the source node that generated the anomaly, conflicting evidence pairs, the specific boundary violation threshold, and the boundary of the state change (from...). arrive (and the rule identifier that triggers the ruling).
[0085] Based on the extracted smallest verifiable unit of evidence or key anomaly coordinates, the system performs focused event sequence replay and verification. The replay process revolves around the state transition point, reloading and verifying the validity of related evidence, rule triggering logic, and consensus signatures. This design allows parties investigating disputes to quickly obtain the reasons for state transitions without having to search through massive amounts of sensor data and logs one by one.
[0086] This embodiment not only significantly reduces the computational workload and time cost of dispute review, but also makes the final determination of responsibility more accurate, clear and easy to explain.
[0087] Application Example 1: Trading of the right to use delivery robots (lease handover and automated control) In this example, for a business scenario where delivery robots are purchased on an hourly or per-task basis, the platform first initializes the robot's identity by binding its unique identifier with hardware identity identifiers and cryptographic security credentials, such as the main control board serial number, battery number, drive motor number, and security chip public key. When a third party initiates a usage right purchase, the system collects multi-source status data in real time, including location, remaining battery power, wheel set fault codes, and maintenance records. This data is preprocessed to form a standardized status evidence package, which is then analyzed by multiple AI oracle nodes based on the evidence package and preset rules. After confirming that the robot's location is compliant, its battery power is sufficient, and it is fault-free, the system outputs a "leasable" ruling. This status level is then mapped to a tradable digital asset status and written to a trusted ledger along with an evidence digest. Finally, the platform automatically allows the usage right transaction by calling the "leasable" status object registered on the blockchain for admission determination. Conversely, if the oracle detects abnormal battery temperature or location deviation (i.e., the status level changes to "suspended order taking" or "frozen"), the platform will block new usage right transactions according to preset strategies, achieving automated risk control.
[0088] Application Example 2: Delivery and Maintenance Responsibility Switching for Industrial Robotic Arms (Ownership Transfer and Liability Transfer) This example applies to a two-tier workflow scenario for industrial robotic arms. The platform first initializes the robotic arm's identity and binds hardware identifiers such as the controller and core joints. When ownership transfer and maintenance responsibility switching are involved, the system collects multi-source status data, including controller logs, joint current, temperature rise, visual calibration, and maintenance work orders, and generates a status evidence package. This package is then fused and analyzed by an AI oracle node to verify identity consistency, core joint wear, and firmware legitimacy. If key indicators are compliant, the system outputs a ruling of "deliverable and maintenance responsibility switchable." This status level is then mapped to a digital asset status and written to a trusted ledger. Finally, when the buyer confirms receipt, the platform directly calls the "deliverable" status registered on the blockchain, automatically triggering the transfer of ownership of the digital asset and the update of maintenance responsibility, thereby achieving automated delivery control based on physical facts. If the oracle detects abnormal joint wear or inconsistent firmware signatures, the status level will be ruled "pending review," and the system will block the delivery action to ensure the security of asset transfer.
[0089] Application Example 3: Task Settlement and Guarantee Release for AGVs in the Industrial Park (Task Settlement and Guarantee Release) For the task settlement and guarantee release business of AGV robots in the park logistics scenario, the system first collects multi-source status data in real time based on the bound hardware identity, including task trajectory, cargo sensor data, access control records, instruction receipts, and on-site camera verification results. Then, the AI oracle node performs time-series analysis on this data to comprehensively determine the task completion degree, path compliance, and external receipt consistency. When the overall credibility exceeds a preset threshold, it outputs a ruling result of "settlement and release of guarantee deposit". This status is then mapped to the corresponding digital asset status, and an evidence package containing trajectory summary and receipt information is written into the trusted ledger. Finally, the platform automatically triggers the settlement and guarantee deposit release process by calling the status registered on the chain. Conversely, if there is a path deviation or receipt conflict, the status level will be updated to "dispute pending", and the platform will automatically suspend the relevant settlement and release actions until the dispute is resolved.
[0090] The three application examples above collectively demonstrate the core logic of the real-time traffic of this invention: it is not simply about predicting the business revenue of robots, but rather about constructing a truth layer of digital asset states supported by physical facts through a closed loop of "multi-source data acquisition - AI oracle adjudication - on-chain state mapping", thereby providing accurate automated access control basis for complex business events such as ownership transfer, leasing, settlement and maintenance.
[0091] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0092] In one embodiment, a robot state intelligent anchoring device is provided, which corresponds one-to-one with the robot state intelligent anchoring method described in the above embodiments. For example... Figure 2 As shown, the robot's intelligent state anchoring device includes: The identity registration and binding module 10 is used to initialize the robot's identity: assign a unique identifier to the robot and bind the unique identifier to the robot's hardware identity identifier, cryptographic security credentials, ownership and traceability metadata; The state data acquisition module 20 is used to acquire multi-source state data of the robot; the multi-source state data includes internal state data and external observation data; The evidence standardization and encapsulation module 30 is used to preprocess the multi-source state data to form a standardized state evidence package; The time-series analysis and consensus adjudication module 40 is used to input the state evidence package to multiple AI oracle nodes. The multiple AI oracle nodes perform time-series analysis, state adjudication, and consensus confirmation on events within a specified time window based on the state evidence package, historical state data, and a preset rule base, and output an adjudication result. The adjudication result includes a robot state vector and a state level. The robot state vector represents at least two of the following: hardware health, spatial location or work area state, and energy state. The digital asset state mapping module 50 is used to map the robot state vector and the state level to the digital asset state associated with the robot according to a predefined mapping rule. The on-chain evidence storage and version control module 60 is used to obtain the evidence digest of the state evidence package, write the digital asset state and the evidence digest into a trusted ledger to form an on-chain state object, and record the state version change relationship. The state-driven business control module 70 is used to call the digital asset status registered in the on-chain state object when a business event involving the robot occurs, and to perform access judgment or automated control on the business event according to the preset policy rules; wherein, the business event includes at least one of ownership transfer, lease handover, task settlement, maintenance responsibility switching or guarantee release; The timing analysis and consensus adjudication module is also used for: Based on the collection timestamp, trusted time source, device serial number, and message causal relationship in the state evidence package, the temporal relationship of events within the specified time window is constructed; Based on the aforementioned temporal relationship, deterministic rule verification, machine learning model inference, and historical window analysis are performed in parallel to obtain node analysis results and generate candidate state conclusions. By using the Byzantine fault-tolerant consensus mechanism, candidate state conclusions generated by multiple AI oracle nodes are exchanged and verified. When more than a preset threshold of nodes reach a consensus on the same candidate state conclusion, it is confirmed as the ruling result.
[0093] Optionally, the internal status data includes at least one of controller logs, sensor data, actuator feedback, current, voltage, temperature, task progress, and fault codes; The external observation data includes at least one of the following: data reported by the positioning base station, camera verification results, access control records, maintenance work order records, and results uploaded by third-party detection nodes; The preprocessing includes time alignment, format standardization, source identification, signature verification, and outlier removal.
[0094] Optionally, the candidate state conclusion may further include a comprehensive confidence level and a causal explanation; The deterministic rules are used to verify the logical consistency, legality, and temporal validity of evidence; the machine learning model is used to identify operating states and abnormal patterns; and the historical window analysis is used to assess the continuity of state changes and identify unreasonable mutations.
[0095] Optionally, the digital asset status is mapped to a status label including at least one of tradable, leaseable, deliverable, frozen, pending review, pending repair, and dispute pending.
[0096] Optionally, the state-driven business control module 70 is also used for: During the state adjudication process through the AI oracle node, if it is identified that the robot's identity is inconsistent, its location is abnormal, its components are seriously faulty, its security status is abnormal, the credibility of the evidence is below the threshold, or there is a substantial conflict between multiple sources of evidence, then the digital asset status in the adjudication result will be set to a frozen state, a pending review state, or a dispute pending state.
[0097] Optionally, it also includes an audit playback and verification module, which is used for: In response to a dispute query request, trusted audit data within a specified time window is obtained; the trusted audit data includes the status evidence package, the status adjudication log of the AI oracle node, the version record of the digital asset status, and related business action records; Based on the trusted audit data, extract the smallest verifiable evidence unit and / or key anomaly coordinates that led to the state change, and replay and verify the event sequence.
[0098] Optionally, the state-driven business control module 70 is also used for: Retrieve the digital asset state from the on-chain state object and obtain its contained state level and state label; Obtain the status level threshold and status label admission conditions associated with business events; The status level is compared with the status level threshold, and the status label is verified to meet the status label admission conditions of the current business event. If the status level is lower than the status level threshold, or if the status label does not meet the status label admission conditions for the current business event, the execution process of the business event will be automatically blocked.
[0099] Specific limitations regarding the intelligent robot state anchoring device can be found in the limitations of the intelligent robot state anchoring method described above, and will not be repeated here. Each module in the aforementioned intelligent robot state anchoring device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.
[0100] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 3 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores data related to the intelligent robot state anchoring method. The network interface communicates with external terminals via a network connection. When the processor executes the computer program, it implements an intelligent robot state anchoring method.
[0101] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the robot state intelligent anchoring method described in the above embodiment; to avoid repetition, this will not be repeated here. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the robot state intelligent anchoring device embodiment; to avoid repetition, this will not be repeated here.
[0102] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program implements the robot state intelligent anchoring method described in the above embodiment; to avoid repetition, it will not be described again here. Alternatively, when executed by a processor, the computer program implements the functions of each module / unit in this embodiment of the robot state intelligent anchoring device; to avoid repetition, it will not be described again here.
[0103] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0104] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0105] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.
Claims
1. A method for intelligent anchoring of robot state, characterized in that, include: Initialize the robot's identity: Assign a unique identifier to the robot and bind the unique identifier to the robot's hardware identity identifier, cryptographic security credentials, ownership and traceability metadata; Acquire multi-source state data of the robot; the multi-source state data includes internal state data and external observation data; The multi-source state data is preprocessed to form a standardized state evidence package; The state evidence package is input to multiple AI oracle nodes. The multiple AI oracle nodes perform time-series analysis, state adjudication, and consensus confirmation on events within a specified time window based on the state evidence package, historical state data, and a preset rule base, and output the adjudication result. The adjudication result includes a robot state vector and a state level. The robot state vector represents at least two of the following: hardware health, spatial location or work area state, and energy state. According to predefined mapping rules, the robot state vector and the state level are mapped to the digital asset state associated with the robot; Obtain the evidence digest of the state evidence package, write the digital asset state and the evidence digest into a trusted ledger to form an on-chain state object, and record the state version change relationship; When a business event involving the robot occurs, the digital asset status registered in the on-chain state object is invoked, and the business event is subject to access determination or automated control according to preset policy rules; wherein, the business event includes at least one of ownership transfer, lease handover, task settlement, maintenance responsibility switching or guarantee release.
2. The intelligent robot state anchoring method according to claim 1, characterized in that, The internal status data includes at least one of the following: controller logs, sensor data, actuator feedback, current, voltage, temperature, task progress, and fault codes. The external observation data includes at least one of the following: data reported by the positioning base station, camera verification results, access control records, maintenance work order records, and results uploaded by third-party detection nodes; The preprocessing includes time alignment, format standardization, source identification, signature verification, and outlier removal.
3. The intelligent robot state anchoring method according to claim 1, characterized in that, The process involves inputting the state evidence package into multiple AI oracle nodes, which then perform time-series analysis, state adjudication, and consensus confirmation on events within a specified time window based on the state evidence package, historical state data, and a preset rule base, to output an adjudication result, including: Based on the collection timestamp, trusted time source, device serial number, and message causal relationship in the state evidence package, the temporal relationship of events within the specified time window is constructed; Based on the aforementioned temporal relationship, deterministic rule verification, machine learning model inference, and historical window analysis are performed in parallel to obtain node analysis results and generate candidate state conclusions. The deterministic rules are used to verify the logical consistency, legality, and temporal validity of the evidence; the machine learning model is used to identify the running state and abnormal patterns; and the historical window analysis is used to evaluate the continuity of state changes and identify unreasonable mutations. The Byzantine fault-tolerant consensus mechanism is used to exchange and verify candidate state conclusions generated by multiple AI oracle nodes. When more than a preset threshold of nodes reach a consensus on the same candidate state conclusion, it is confirmed as the ruling result. The candidate state conclusion also includes comprehensive credibility and causal explanation.
4. The intelligent robot state anchoring method according to claim 1, characterized in that, The status of the digital assets is mapped to a status label including at least one of the following: tradable, leaseable, deliverable, frozen, pending review, pending maintenance, and dispute pending.
5. The intelligent robot state anchoring method according to claim 1, characterized in that, Also includes: During the state adjudication process through the AI oracle node, if it is identified that the robot's identity is inconsistent, its location is abnormal, its components are seriously faulty, its security status is abnormal, the credibility of the evidence is below the threshold, or there is a substantial conflict between multiple sources of evidence, then the digital asset status in the adjudication result will be set to a frozen state, a pending review state, or a dispute pending state.
6. The intelligent robot state anchoring method according to claim 1, characterized in that, Also includes: In response to a dispute query request, retrieve trusted audit data within a specified time window; The trusted audit data includes the status evidence package, the status decision log of the AI oracle node, the version record of the digital asset status, and related business action records; Based on the trusted audit data, extract the smallest verifiable evidence unit and / or key anomaly coordinates that led to the state change, and then replay and verify the event sequence.
7. The intelligent robot state anchoring method according to claim 1, characterized in that, The process of invoking the digital asset status registered in the on-chain state object and performing access determination or automated control on the business event according to preset policy rules includes: Retrieve the digital asset state from the on-chain state object and obtain its contained state level and state label; Obtain the status level threshold and status label admission conditions associated with business events; The status level is compared with the status level threshold, and the status label is verified to meet the status label admission conditions of the current business event. If the status level is lower than the status level threshold, or if the status label does not meet the status label admission conditions for the current business event, the execution process of the business event will be automatically blocked.
8. A robot state intelligent anchoring device, characterized in that, include: The identity registration and binding module is used to initialize the robot's identity: assigning a unique identifier to the robot and binding the unique identifier to the robot's hardware identity identifier, cryptographic security credentials, ownership and traceability metadata; The state data acquisition module is used to acquire multi-source state data of the robot; the multi-source state data includes internal state data and external observation data. An evidence standardization and encapsulation module is used to preprocess the multi-source state data to form a standardized state evidence package; The time-series analysis and consensus adjudication module is used to input the state evidence package to multiple AI oracle nodes. The multiple AI oracle nodes perform time-series analysis, state adjudication, and consensus confirmation on events within a specified time window based on the state evidence package, historical state data, and a preset rule base, and output the adjudication result. The adjudication result includes a robot state vector and a state level. The robot state vector represents at least two of the following: hardware health, spatial location or work area state, and energy state. The digital asset state mapping module is used to map the robot state vector and the state level to the digital asset state associated with the robot according to predefined mapping rules. The on-chain evidence storage and version control module is used to obtain the evidence digest of the state evidence package, write the digital asset state and the evidence digest into a trusted ledger to form an on-chain state object, and record the state version change relationship. The state-driven business control module is used to call the digital asset status registered in the on-chain state object when a business event involving the robot occurs, and to perform access judgment or automated control on the business event according to the preset policy rules; wherein, the business event includes at least one of ownership transfer, lease handover, task settlement, maintenance responsibility switching or guarantee release.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the robot state intelligent anchoring method according to any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the robot state intelligent anchoring method according to any one of claims 1 to 7.
