Skill data based blockchain operating system skill trusted loading method

Abstract

This invention relates to the field of security technology for embodied intelligent robot operating systems, specifically disclosing a trusted loading method for skills in an operating system based on a blockchain for skill data ownership confirmation. This method constructs a collaborative architecture comprising a skill data receiving module, a blockchain ownership confirmation module, a trusted verification module, a secure loading module, and an on-chain evidence storage module. The blockchain ownership confirmation module employs a copyright registration agency to perform multi-dimensional on-chain evidence storage of the skill data package's source, creation process, ownership information, creation timestamp, creator identity, ownership change history, and hash anchor value, generating an immutable digital identity credential. This invention, through full-lifecycle ownership confirmation and end-to-end security control, achieves traceability, verifiability, and immutability of skill data from creation and transaction to use, effectively solving problems such as piracy, abuse of rights, and data tampering, providing a technical foundation for the secure circulation of data in the embodied intelligent industry.

Description

Technical Field

[0001] This invention belongs to the field of security technology for embodied intelligent robot operating systems, and specifically relates to a trusted loading method for operating systems based on blockchain for confirming the ownership of skill data. Background Technology

[0002] With the rapid iteration of embodied intelligence technology, skill data, as the core asset carrying robot motion skills, is experiencing exponential growth in circulation and reuse frequency. Skill data includes multi-dimensional content such as motion trajectories, force curves, timing information, and constraints, embodying the intellectual achievements and commercial value of creators. The need for copyright protection and compliant use is becoming increasingly urgent. Currently, skill data needs to flow between creators, users, and platforms, involving multiple scenarios such as local storage, cloud access, and cross-platform migration, facing multiple security challenges throughout the entire process.

[0003] Existing skills data security solutions, centered on centralized management, suffer from five prominent problems: First, the credibility of copyright confirmation is insufficient, relying on centralized copyright registration agencies, resulting in low registration efficiency, poor cross-platform interoperability, high single-point failure risk, and frequent copyright disputes due to disagreements on copyright ownership across different platforms. Second, the reliability of data integrity verification is weak, lacking effective constraints on the risk of tampering during transmission and storage, and the verification results of traditional centralized verification agencies are easily attacked or bribed, leading to a high risk of verification failure. Third, the security of access control is insufficient, with certificate-based authorization mechanisms vulnerable to certificate forgery, access theft, and excessive access granting; attackers can bypass checks and execute restricted operations by obtaining valid certificates, threatening the security of robot systems. Fourth, cross-platform interoperability is poor, with different robot operating systems using heterogeneous skills data formats and access control mechanisms, creating natural barriers to cross-platform migration and interoperability of skills data, limiting the scope of circulation and value release. Fifth, data traceability is difficult, lacking complete records of the data flow process; after infringement or abuse occurs, copyright holders find it difficult to trace the infringement path and determine the responsible party, resulting in high costs and long cycles for rights protection.

[0004] To address the aforementioned pain points, the industry has proposed various improvement solutions, but all have significant technical limitations: digital watermarking-based solutions embed watermarks into skill data to identify copyright, but these watermarks are easily removed or covered by attackers through signal processing, failing to provide reliable legal evidence; encrypted transmission-based solutions ensure secure transmission through encryption technology, but cannot prevent malicious copying and use at the receiving end, resulting in incomplete security coverage; hardware-based root of trust solutions achieve trusted verification through dedicated security chips, but the hardware costs are high and deployment is complex, making it difficult to popularize in general robot systems; blockchain-based evidence storage solutions only record copyright registration information and do not establish a complete technical closed loop from copyright confirmation and permission verification to secure loading, failing to solve end-to-end security issues in real-world scenarios.

[0005] The expansion of embodied intelligent application scenarios places higher demands on skills data security technologies: In terms of credibility, a decentralized and tamper-proof copyright confirmation mechanism needs to be established to give skills data copyright ownership credibility and legal effect, supporting legitimate circulation and transactions; in terms of integrity, an end-to-end integrity verification mechanism needs to be established to ensure that skills data is not tampered with throughout the entire process from creator to user, guaranteeing the safety of robot skills execution; in terms of access control, a refined and traceable access management mechanism needs to be established to achieve precise access control and timely detection of unauthorized use, protecting the interests of copyright holders and combating infringement; in terms of interoperability, a cross-platform and cross-system skills data interoperability mechanism needs to be established to break down platform barriers and promote the widespread circulation and maximization of the value of skills data; in terms of traceability, a complete circulation record mechanism needs to be established to achieve rapid tracing of infringement and determination of liability, reducing the cost of rights protection. Summary of the Invention

[0006] In view of this, the present invention provides a skill trusted loading method for an operating system based on a skill data ownership blockchain, in order to solve or alleviate one of the technical problems existing in the prior art, and at least provide a beneficial option.

[0007] The technical solution of this invention is implemented as follows: a trusted skill loading method for an operating system based on a skill data ownership blockchain, comprising the following steps: Step S1: Receive the skill data packet to be loaded through the skill data receiving module, and perform format compliance verification and integrity pre-check on the skill data packet. The format compliance verification matches the preset skill data encapsulation specification, and the integrity pre-check identifies obviously damaged data packets by calculating the deviation between the hash value of the data packet and the preset benchmark value. Step S2 involves using a blockchain rights confirmation module to perform multi-dimensional on-chain notarization of the source, creation process, and ownership information of the skill data package, generating an immutable digital identity certificate. The blockchain rights confirmation module employs a consortium blockchain architecture, with nodes from copyright registration agencies, skill trading platforms, and robot manufacturers jointly maintaining the blockchain ledger. A practical Byzantine fault-tolerant consensus mechanism ensures data consistency. The multi-dimensional on-chain notarization includes the skill data's creation timestamp, creator identity information, ownership change history, and hash anchoring information. The hash anchoring information is obtained by performing SHA-256 hash calculation on the entire skill data content. Step S3: The trusted verification module verifies the legality and integrity of the skill data package based on the ownership information on the blockchain. The trusted verification module adopts a zero-knowledge proof protocol, allowing users to prove their legal usage rights to the verifier without exposing the specific content of the skill data or their own privacy permissions. The verification process supports copyright-level verification and operation-level verification. Copyright-level verification is used to confirm whether the user has the right to use the skill data, while operation-level verification is used to confirm whether the user has the right to perform at least one specific operation among viewing, modifying, and re-authorizing. Step S4: After successful verification, the skill data package is securely loaded into the trusted execution environment of the robot operating system by the secure loading module. The secure loading module places the loading and decryption process of the skill data in a hardware-level securely isolated trusted execution environment, and adopts a layered encryption protection mechanism. The outer layer uses the AES-256 symmetric encryption algorithm to protect the confidentiality of the data, and the inner layer uses the HMAC-SHA256 integrity protection algorithm to prevent data tampering. The decryption operation is completed entirely within the trusted execution environment, and the key is not leaked to the external environment. Step S5: The skill data loading record, permission verification result, and abnormal event information are uploaded to the blockchain for evidence storage through the on-chain evidence storage module, forming a complete skill data flow audit chain. The on-chain evidence storage module adopts an event-driven on-chain mechanism, which automatically triggers on-chain operations at key nodes such as copyright registration, permission granting, skill loading, and abnormal alarms, without the need for manual intervention.

[0008] Preferably, the multi-dimensional copyright information storage of the blockchain rights confirmation module specifically includes: recording the creation environment parameters of skill data, including the model of the collection device, device fingerprint, ambient temperature and humidity, and operator identification; recording the full history of ownership changes, with each ownership transfer, authorization change, or registration revocation operation generating a new storage record on the blockchain, which is associated with the previous storage record through a hash pointer to form an immutable copyright change history chain; the hash anchoring information includes not only the full content hash but also segmented hash values ​​divided by data blocks, supporting partial data integrity verification.

[0009] Preferably, the trusted verification module adopts an attribute-based access control model for permission management. The granting and verification of permissions are based on a combination of three attribute dimensions: skill data attributes, including skill type, confidentiality level, applicable scenario, and validity period; user attributes, including user identity level, affiliated organization, credit score, and historical usage records; and environmental attributes, including current network environment security level, device trust status, and geographical location compliance. The trusted verification module supports permission verification under offline conditions. By pre-caching recently valid verification results and using a lightweight zero-knowledge proof algorithm, local permission verification can still be completed when the network is interrupted.

[0010] Preferably, the secure loading module is equipped with a runtime integrity monitoring function, which continuously monitors two types of integrity status during the execution of skill data: data area access permissions, to detect whether there is any behavior of unauthorized reading or writing of core parameters of skill data; code area execution permissions, to detect whether there is any unauthorized code injection or execution flow tampering; when abnormal behavior is detected, a three-level security response is immediately triggered: the first level response suspends skill execution and isolates suspicious data, the second level response clears the temporary cache and resets the execution environment, and the third level response terminates the skill process and reports the security event to the management platform and the blockchain evidence storage module.

[0011] Preferably, the secure loading module supports progressive loading and dynamic updating of skills. In progressive loading mode, the core control logic and basic action sequence of the skill are loaded first, while non-critical parameters and extended functions are loaded asynchronously during the robot's idle period to ensure that the current task is not interrupted. In dynamic updating mode, the updated content of the skill is identified by version difference comparison, only the changed data block is downloaded and merged and verified in a trusted execution environment, and the update process supports hot switching without restarting the robot system.

[0012] Preferably, the secure loading module has a built-in self-recovery function, which stores the baseline version and security configuration of skill data through a dual-mirror backup mechanism: when a loading anomaly is detected, it first attempts to repair the anomaly by verifying redundant data; if the repair fails, it automatically rolls back to the most recent trusted secure initial state. The rollback process is performed entirely in a trusted execution environment to prevent the rollback operation from being maliciously hijacked; after the self-recovery is completed, the anomaly type, occurrence time, and processing result are stored on the blockchain as evidence.

[0013] Preferably, the blockchain rights confirmation module supports a cross-chain mutual recognition mechanism, which realizes identity mutual recognition and information synchronization with other blockchain networks through a relay chain and a cross-chain gateway: when skill data needs to be transferred between platforms of different blockchain systems, the source chain sends a copyright certificate to the target chain through a cross-chain protocol. After the target chain verifies the validity of the signature and the legality of the timestamp of the certificate, it recognizes the copyright confirmation result of the source chain, without having to repeat the rights confirmation process; the cross-chain mutual recognition process uses zero-knowledge proof technology to hide sensitive ownership information and only discloses necessary verification information to the target chain.

[0014] Preferably, the on-chain evidence storage module uses a Merkle tree data structure to perform compression optimization on the evidence storage information: multiple evidence storage records of the same skill data are constructed into a Merkle tree, and only the root hash is stored on the chain, while the original evidence storage records are distributed and stored in the local database of the consortium chain node; when it is necessary to verify the authenticity of a certain evidence storage record, only the hash path from the record to the root of the tree needs to be provided to complete the verification, thereby reducing the on-chain storage by more than 80% while ensuring data integrity and verifiability.

[0015] Preferably, the on-chain evidence storage module supports selective disclosure of evidence storage information: when copyright holders, regulatory agencies, and judicial departments query evidence storage information, they can obtain information of corresponding granularity according to their query permission level; when querying the public, only non-sensitive information such as skill name, creator, copyright validity period, and ownership status is disclosed; when conducting judicial evidence collection, the complete circulation history and hash anchor value are disclosed; when conducting commercial cooperation, the scope of authorization and usage constraints are disclosed, and sensitive commercial content is verified for authenticity through zero-knowledge proofs without being directly displayed.

[0016] Preferred options include: The skill data receiving module is configured to receive skill data packets to be loaded, perform format compliance verification and integrity pre-check, and output skill data packets that pass the verification. The blockchain rights confirmation module is configured to use a consortium blockchain architecture to perform multi-dimensional on-chain evidence storage of the source, creation process, and ownership information of skill data packages, generating digital identity credentials that include creation timestamps, creator identity, and hash anchor values. The trusted verification module is configured to use a zero-knowledge proof protocol to perform multi-level permission verification at the copyright level and operation level based on an attribute-based access control model, and output the permission verification results. The secure loading module is configured to place the verified skill data package in a trusted execution environment, use a layered encryption mechanism to complete decryption and loading, and is equipped with runtime integrity monitoring, progressive loading, self-recovery functions, and output secure execution status. The on-chain evidence storage module is configured to store skill data loading records, permission verification results, and abnormal event information on the blockchain in an event-driven manner, forming a complete flow audit chain. The embodiments of the present invention have the following advantages due to the adoption of the above technical solutions: The TOD (Transit-Oriented Development) technology of this invention has the following core beneficial effects: Copyright verification credibility is significantly improved: The full lifecycle rights verification method based on consortium blockchain establishes an immutable digital identity file for each piece of skill data. Compared with the traditional centralized copyright registration method, the incidence of copyright disputes is reduced by more than 70%, and the copyright verification time is shortened from an average of several weeks to several minutes, which greatly improves the efficiency of legal circulation and transaction of skill data.

[0017] Enhanced reliability of data integrity verification: Copyright information and data integrity information are bound to the blockchain and combined with zero-knowledge proof privacy verification to achieve end-to-end verifiability and traceability. The success rate of data tampering attacks is reduced to below 0.1%, and the verification latency is controlled within milliseconds, effectively ensuring the security of robot skill execution.

[0018] Comprehensive enhancement of access control security: Based on the zero-knowledge proof-based access verification protocol and the trusted execution environment secure loading mechanism, it achieves fine-grained access control and privacy protection, reduces the risk of access theft by more than 90%, and keeps the computational overhead of access verification within an acceptable range, effectively safeguarding the interests of copyright holders.

[0019] Cross-platform interoperability is greatly enhanced: a unified skill data ownership protocol and loading interface standard breaks down interoperability barriers between different operating systems, increasing the success rate of cross-platform migration of skill data to over 95% and reducing migration time by over 60%, thus promoting the widespread circulation and maximizing the value of skill data.

[0020] Significantly enhanced infringement tracing capabilities: The complete blockchain transaction history supports rapid tracing of infringement and determination of liability, reducing the average time for infringement tracing from several months to several days, and increasing the accuracy rate of infringement liability determination to over 95%, significantly reducing the cost of rights protection and improving the efficiency of rights protection.

[0021] The overall security of the system is continuously optimized: a complete technical loop from copyright confirmation and permission verification to secure loading improves the security of skill data at the system level. In actual deployment, the leakage rate and tampering rate of skill data are both at extremely low levels, and the overall availability of the system remains above 99.5%.

[0022] The above overview is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features of the invention will become readily apparent from the accompanying drawings and the following detailed description. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart of the skill trusted loading method for an operating system based on skill data ownership blockchain, as described in this invention. Detailed Implementation

[0025] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0026] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0027] Example 1: Skill Data Creation Copyright Registration Scenario This embodiment is applied to robot manufacturers' copyright confirmation of independently developed skills data. The specific implementation process is as follows: Skill creators first collect motion demonstration data of three typical robot tasks—grasping, assembling, and inspecting—using high-precision motion capture equipment. During the collection process, metadata such as device fingerprints, environmental parameters, and operator identities are recorded simultaneously. Then, a blockchain-based rights confirmation module notarizes the skill data creation process: the module uses a national-level cryptographic hash algorithm to perform hash calculations on the entire skill data, generating a unique hash anchor value; it associates the collected timestamp, creator identity information, device fingerprint, and other metadata to construct a multi-dimensional copyright notarization record containing the skill name, creator, creation time, data version, hash value, and initial ownership status; the notarization record is sent to a consortium blockchain network, where nodes from the copyright registration agency, manufacturer, and third-party notary jointly perform consensus verification. Upon successful verification, the record is written into a block to generate an immutable digital identity certificate, which includes the blockchain address, notarization hash, timestamp, and the owner's public key.

[0028] When other partners need to use the skill data, they first query the on-chain copyright information through the trusted verification module to verify the copyright ownership, validity period, and rights restrictions of the skill data. After verification, the user initiates an authorization application to the copyright holder. Both parties negotiate the scope of authorization, usage period, and fee standards through a preset smart contract. After reaching a consensus, the authorization transaction is executed. The authorization information is written to the blockchain after being digitally signed by both parties, generating an authorization record containing the authorizing party, the authorized party, the authorized skill ID, the permission level, the validity period, and usage scenario constraints, forming a complete on-chain closed loop of "creation-rights confirmation-authorization".

[0029] Actual operational data shows that after adopting this embodiment, the efficiency of skill data copyright registration has increased by more than 80% compared with traditional offline registration, the incidence of copyright disputes has decreased by about 70%, and the success rate of cross-platform copyright mutual recognition has increased to 100%.

[0030] Example 2: Skills Data Transaction Licensing Scenario This embodiment applies to the entire process of buying and selling skills on a skills trading platform. The specific implementation process is as follows: The seller uploads its verified skill data package to the skill trading platform. The platform verifies the copyright ownership and the seller's right to dispose of the skill data through a trusted verification module. After verification, the data is published in the platform's skill list. The buyer logs into the platform to browse the skill list, selects skills that meet their needs, and views the anonymized skill descriptions, ownership information, and historical transaction reviews. When the buyer initiates a purchase request, the trading platform first verifies the buyer's real-name identity and payment ability through a trusted verification module. The verification process uses a zero-knowledge proof protocol, allowing the buyer to prove their payment ability without exposing sensitive information such as account balances. After verification, the trading platform calls a smart contract to execute the transaction settlement, transferring the agreed fee from the buyer's account to the seller's account. Simultaneously, the ownership transfer operation of the skill data is executed: the original seller's copyright registration record is marked as "transferred," and the buyer obtains a new copyright registration record containing new ownership information, transfer timestamps, transaction hashes, and other information.

[0031] The key nodes of the entire transaction process (order placement, payment verification, ownership transfer, and evidence update) are all automatically triggered on the blockchain by the on-chain evidence storage module, forming an immutable transaction record. In the event of a transaction dispute, both parties can query the complete on-chain transaction record through the platform to reconstruct the transaction negotiation process, payment vouchers, and ownership change records, and quickly determine the responsible party.

[0032] Actual operational data shows that after adopting this embodiment, the average completion time for skills data transactions has been shortened from 7 working days in the traditional model to less than 2 working days, the efficiency of handling transaction disputes has been improved by more than 50%, and the trust between the two parties in the transaction has been significantly enhanced.

[0033] Example 3: Cross-platform skill transfer scenario This embodiment applies to the migration of skill data from robot operating system A to robot operating system B. The specific implementation process is as follows: The skill data owner first retrieves their digital identity certificate for their skill data through the blockchain rights confirmation module to confirm that the skill is in a transferable state and there is no ownership dispute. They then initiate a skill export request to the original platform A. Platform A verifies the owner's permission status through a trusted verification module. Upon successful verification, a skill data package containing copyright information, permission constraints, data hashes, and a migration license signature is generated. The owner imports the data package into the robot operating system of vendor B. Platform B parses the copyright information in the data package through a cross-platform interoperability protocol and calls the blockchain interface to query and verify the authenticity of the copyright and migration permissions. After successful verification, Platform B executes the loading process through a secure loading module: first, the skill data package is transmitted to a trusted execution environment, where the outer layer of symmetric encryption data is decrypted and the inner layer of integrity is verified within an isolated area. After successful verification, the data is loaded into the skill execution engine. During the loading process, the integrity status of the skill data is monitored in real time; if no anomalies are found, the loading is completed. The entire migration process, including export applications, permission verification, and loading records, is automatically recorded on the blockchain by the on-chain evidence storage module, forming a migration audit chain containing the source platform, target platform, migration time, skill ID, and verification results.

[0034] Actual operational data shows that after adopting this embodiment, the success rate of cross-platform migration of skill data has increased to over 95%, and the accuracy of skill execution and timing consistency after migration remains above 98%, with no issues of skill data tampering or functional failure caused by migration.

[0035] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in the present invention, and these should all be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for trusted skill loading in an operating system based on a skill data ownership blockchain, characterized in that: Includes the following steps: Step S1: Receive the skill data packet to be loaded through the skill data receiving module, and perform format compliance verification and integrity pre-check on the skill data packet. The format compliance verification matches the preset skill data encapsulation specification, and the integrity pre-check identifies obviously damaged data packets by calculating the deviation between the hash value of the data packet and the preset benchmark value. Step S2: The source, creation process, and ownership information of the skill data package are stored on the blockchain in multiple dimensions through the blockchain rights confirmation module to generate an immutable digital identity certificate. The blockchain rights confirmation module adopts a consortium blockchain architecture, in which the blockchain ledger is jointly maintained by the copyright registration agency node, the skill trading platform node, and the robot manufacturer node, and data consistency is ensured through a practical Byzantine fault-tolerant consensus mechanism. The multi-dimensional on-chain evidence includes the creation timestamp of the skill data, creator identity information, ownership change history, and hash anchor information. The hash anchor information is obtained by performing SHA-256 hash calculation on the full content of the skill data. Step S3: The trusted verification module verifies the legality and integrity of the skill data package based on the ownership information on the blockchain; The trusted verification module adopts a zero-knowledge proof protocol, which allows users to prove to the verification party that they have legal access rights without exposing the specific content of the skill data or their own privacy permissions. The verification process supports copyright-level verification and operation-level verification. Copyright-level verification is used to confirm whether the user has the right to use the skill data, and operation-level verification is used to confirm whether the user has the right to perform at least one specific operation among viewing, modifying, and re-authorizing. Step S4: After successful verification, the skill data package is securely loaded into the trusted execution environment of the robot operating system by the secure loading module. The secure loading module places the loading and decryption process of the skill data in a hardware-level securely isolated trusted execution environment, and adopts a layered encryption protection mechanism. The outer layer uses the AES-256 symmetric encryption algorithm to protect the confidentiality of the data, and the inner layer uses the HMAC-SHA256 integrity protection algorithm to prevent data tampering. The decryption operation is completed entirely within the trusted execution environment, and the key is not leaked to the external environment. Step S5: The skill data loading record, permission verification result, and abnormal event information are uploaded to the blockchain for evidence storage through the on-chain evidence storage module, forming a complete skill data flow audit chain. The on-chain evidence storage module adopts an event-driven on-chain mechanism, which automatically triggers on-chain operations at key nodes such as copyright registration, permission granting, skill loading, and abnormal alarms, without the need for manual intervention.

2. The method for trusted skill loading in an operating system based on a skill data ownership blockchain according to claim 1, characterized in that: The multi-dimensional copyright information storage of the blockchain rights confirmation module specifically includes: recording the creation environment parameters of skill data, including the model of the collection device, device fingerprint, ambient temperature and humidity, and operator identification; recording the full history of ownership changes, with each ownership transfer, authorization change, or registration revocation operation generating a new storage record on the blockchain, which is associated with the previous storage record through a hash pointer to form an immutable copyright change history chain; the hash anchoring information, in addition to the full content hash, also includes segmented hash values ​​divided by data blocks, supporting partial data integrity verification.

3. The skill-trusted loading method for an operating system based on skill data ownership blockchain according to claim 1, characterized in that: The trusted verification module adopts an attribute-based access control model for permission management. The granting and verification of permissions are based on a combination of three attribute dimensions: skill data attributes, including skill type, confidentiality level, applicable scenario, and validity period. User attributes include user identity level, affiliated organization, credit score, and historical usage records; environmental attributes include current network environment security level, device trust status, and geographical location compliance; the trusted verification module supports permission verification under offline conditions, and can still complete local permission verification when the network is interrupted by pre-caching recent valid verification results and using a lightweight zero-knowledge proof algorithm.

4. The method for trusted skill loading in an operating system based on a skill data ownership blockchain according to claim 1, characterized in that: The secure loading module is equipped with runtime integrity monitoring function, which continuously monitors two types of integrity status during the execution of skill data: data area access permissions, and detects whether there is any behavior of unauthorized reading or writing of core parameters of skill data; Code region execution permissions are checked to detect unauthorized code injection or execution flow tampering. When abnormal behavior is detected, a three-level security response is immediately triggered: Level 1 response suspends skill execution and isolates suspicious data; Level 2 response clears temporary cache and resets the execution environment; Level 3 response terminates the skill process and reports the security incident to the management platform and blockchain evidence storage module.

5. The method for trusted skill loading in an operating system based on a skill data ownership blockchain according to claim 1, characterized in that: The secure loading module supports progressive loading and dynamic updates of skills. In progressive loading mode, the core control logic and basic action sequences of the skill are loaded first, while non-critical parameters and extended functions are loaded asynchronously during the robot's idle period to ensure that the current task is not interrupted. In dynamic update mode, the updated content of the skill is identified by version difference comparison, only the changed data blocks are downloaded and merged and verified in a trusted execution environment. The update process supports hot switching without restarting the robot system.

6. The method for trusted skill loading in an operating system based on a skill data ownership blockchain according to claim 1, characterized in that: The secure loading module has a built-in self-recovery function and stores the baseline version and security configuration of skill data through a dual-mirror backup mechanism: when a loading anomaly is detected, it first attempts to repair the anomaly by verifying redundant data; If the repair fails, it will automatically roll back to the most recent trusted secure initial state. The entire rollback process will be performed in a trusted execution environment to prevent the rollback operation from being maliciously hijacked. After the recovery is completed, the exception type, occurrence time and processing result will be stored on the blockchain as evidence.

7. The skill-trusted loading method for an operating system based on skill data ownership blockchain according to claim 2, characterized in that: The blockchain rights confirmation module supports a cross-chain mutual recognition mechanism, achieving identity mutual recognition and information synchronization with other blockchain networks through a relay chain and a cross-chain gateway: when skill data needs to be transferred between platforms of different blockchain systems, the source chain sends a copyright certificate to the target chain through a cross-chain protocol. After the target chain verifies the validity of the signature and the legality of the timestamp of the certificate, it recognizes the copyright confirmation result of the source chain, without having to repeat the rights confirmation process; the cross-chain mutual recognition process uses zero-knowledge proof technology to hide sensitive ownership information, disclosing only the necessary verification information to the target chain.

8. The method for trusted skill loading in an operating system based on a skill data ownership blockchain according to claim 1, characterized in that: The on-chain evidence storage module uses a Merkle tree data structure to compress and optimize the evidence storage information: multiple evidence storage records of the same skill data are constructed into a Merkle tree, and only the root hash is stored on the chain, while the original evidence storage records are distributed and stored in the local database of the consortium chain nodes; when it is necessary to verify the authenticity of a certain evidence storage record, only the hash path from the record to the root of the tree needs to be provided to complete the verification, thereby reducing the on-chain storage by more than 80% while ensuring data integrity and verifiability.

9. The method for trusted skill loading in an operating system based on a skill data ownership blockchain according to claim 1, characterized in that: The on-chain evidence storage module supports selective disclosure of evidence storage information: when copyright holders, regulatory agencies, and judicial departments query evidence storage information, they can obtain information of corresponding granularity according to their query permission level; when querying the public, only non-sensitive information such as skill name, creator, copyright validity period, and ownership status is disclosed; when used for judicial evidence collection, the complete circulation history and hash anchor value are disclosed; when used for commercial cooperation, the scope of authorization and usage constraints are disclosed, and sensitive commercial content is verified for authenticity through zero-knowledge proofs without being directly displayed.

10. The method for trusted skill loading in an operating system based on a skill data ownership blockchain according to any one of claims 1 to 9, characterized in that, include: The skill data receiving module is configured to receive skill data packets to be loaded, perform format compliance verification and integrity pre-check, and output skill data packets that pass the verification. The blockchain rights confirmation module is configured to use a consortium blockchain architecture to perform multi-dimensional on-chain evidence storage of the source, creation process, and ownership information of skill data packages, generating digital identity credentials that include creation timestamps, creator identity, and hash anchor values. The trusted verification module is configured to use a zero-knowledge proof protocol to perform multi-level permission verification at the copyright level and operation level based on an attribute-based access control model, and output the permission verification results. The secure loading module is configured to place the verified skill data package in a trusted execution environment, use a layered encryption mechanism to complete decryption and loading, and is equipped with runtime integrity monitoring, progressive loading, self-recovery functions, and output secure execution status. The on-chain evidence storage module is configured to store skill data loading records, permission verification results, and abnormal event information on the blockchain in an event-driven manner, forming a complete flow audit chain.

Images