A Human-Machine Collaborative Scheduling Method and System Based on Multidimensional Task Protocol

Abstract

This invention relates to the field of artificial intelligence technology, particularly to a human-machine hybrid collaborative scheduling method and system based on multidimensional task protocolization. The method includes: establishing a collaborative system architecture supporting access from heterogeneous entities such as humans and intelligent agents; providing corresponding access ports for human users and digital intelligent agents; receiving unstructured task requests submitted by task initiators; performing multidimensional semantic parsing and feature extraction on the unstructured task requests through a multidimensional task protocolization conversion engine, converting them into machine-readable standardized protocol tasks; completing multidimensional protocolization and dimensionality reduction processing of tasks; and using preset cross-modal unified intelligent workload measurement rules. This invention achieves unified value measurement of human cognitive labor and intelligent agent computing power consumption, and automated settlement of human-machine hybrid collaborative workload, solving the problems of pricing dimension barriers between human physical attributes and digital intelligent agent computing power attributes, and the incommensurability of cross-modal subject workloads.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a human-machine hybrid collaborative scheduling method and system based on multi-dimensional task protocolization. Background Technology

[0002] Traditional task scheduling often relies on single-dimensional decision-making, lacks unified standards for human-machine collaboration, and suffers from ambiguous task decomposition, unbalanced resource matching, and insufficient dynamic adaptability in complex scenarios, making it difficult to guarantee collaborative efficiency and control precision. To address the problems of insufficient task standardization, chaotic human-machine division of labor, and delayed scheduling response, this scheduling method and system are proposed.

[0003] Existing technologies have long relied on human physical working hours as the core basis for pricing and measurement, with task settlement entirely anchored to human biological time consumption. However, the execution of digital intelligent agents depends on computing power and the frequency of computing resource calls. The two have completely different measurement dimensions and logics, and there is no unified value conversion standard. At the same time, existing collaborative systems are all designed around human physical identities and lack standardized access protocols that can be read by machines. They cannot incorporate digital intelligent agents into a unified collaborative system, nor have they built a universal workload measurement model that is compatible with both human and machine subjects. In addition, traditional platforms use two independent and isolated ledger systems for human currency settlement and machine computing power billing. Reconciliation is cumbersome and cannot be automated when collaborating across modalities. Ultimately, this leads to a natural pricing dimension barrier between humans and intelligent agents, and the workload of cross-modal subjects cannot be uniformly measured or is incommensurable. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides the following technical solution: A human-machine hybrid collaborative scheduling method and system based on multidimensional task protocolization includes the following steps: Building a collaborative system architecture that supports access from heterogeneous subjects such as humans and intelligent agents, providing corresponding access ports for human users and digital intelligent agents respectively, receiving unstructured task requests submitted by task initiators, performing multidimensional semantic parsing and feature extraction on unstructured task requests through a multidimensional task protocolization conversion engine, converting them into machine-readable standardized protocol tasks, completing multidimensional protocolization dimensionality reduction processing of tasks, performing standardized value measurement on tasks that have completed protocolization dimensionality reduction through preset cross-modal unified intelligent workload measurement rules, realizing unified pricing of human cognitive labor and intelligent agent computing power consumption, completing budget locking based on the unified pricing result of tasks, and completing task matching and distribution with execution nodes by combining the dimensional requirements of standardized protocol tasks and the adaptability of candidate execution nodes. The execution nodes include human execution nodes and intelligent agent execution nodes. According to the multidimensional acceptance rules agreed upon in the standardized protocol tasks, the task results output by the matched execution nodes are accepted and judged, generating corresponding acceptance results. Based on the acceptance results and the unified pricing results of tasks, the automated clearing and full-process closed-loop management of task value is completed through smart contracts.

[0005] As an improvement to the above technical solution: It provides a visual graphical console access port for human users and a standardized communication protocol API access port for digital intelligent agents, enabling native, seamless access for heterogeneous subjects such as humans and intelligent agents.

[0006] As an improvement to the above technical solution: It clarifies four core dimensions of the task: input requirements, output specification boundaries, time limit constraints, and quality acceptance standards. It uses a finite state machine to complete the flow control of the entire task process and achieve machine-level mutual recognition between the task initiator and the executor.

[0007] As an improvement to the above technical solution: Based on the scale of computing power consumption of the intelligent agent corresponding to the comprehensive task, the complexity of the task logic arrangement, the cognitive load and skill scarcity of human intervention required for the task, the standardized value conversion of the completed task is used to generate the task value amount corresponding to the system's native unified accounting unit.

[0008] As an improvement to the above technical solution: S41: Based on the historical performance data of all nodes in the network, predict the success rate and delivery risk of candidate execution nodes in completing standardized protocol tasks, and with the goal of global scheduling optimization, match the execution node with the highest adaptability to complete the task distribution.

[0009] As an improvement to the above technical solution: If the task result meets the multi-dimensional acceptance rules agreed upon in the standardized protocol task, the acceptance is deemed successful, triggering the subsequent clearing and settlement process; If the acceptance rules are not met, the acceptance is deemed unsuccessful, triggering a task rollback and rescheduling process, and synchronously updating the performance records and credit data of the corresponding execution node.

[0010] As an improvement to the above technical solution: The entire process of budget locking, node bidding, and value clearing within the system is completed through the system's native unified accounting unit. After acceptance, the corresponding value amount is automatically transferred to the execution node through smart contracts. Fiat currency top-ups and final value clearing outside the system are only achieved through a preset gateway.

[0011] As an improvement to the above technical solution: The human-machine hybrid collaborative scheduling system based on multidimensional task protocolization is characterized by including: a heterogeneous subject access module, a multidimensional task protocolization conversion module, a cross-modal unified workload measurement module, a task scheduling and matching module, a multidimensional acceptance and control module, and a smart contract automated settlement module. The heterogeneous subject access module is used to provide standardized access ports for humans and intelligent agents, receive unstructured task requests, and support access registration of human and intelligent agent execution nodes. The multidimensional task protocol conversion module is used to perform multidimensional parsing and feature extraction on unstructured task requests, convert them into standardized protocol tasks that can be read by machines, and complete the task protocol dimensionality reduction. The cross-modal unified workload measurement module has built-in cross-modal unified workload measurement rules, which are used to comprehensively consider the computing power consumption of intelligent agents, task arrangement complexity, human cognitive load and skill scarcity, to standardize the value conversion of protocol-based tasks, generate the task value amount of the system's native unified accounting unit, realize the unified value measurement of human and machine workload, and break down the pricing dimension barrier of heterogeneous human and machine subjects. The task scheduling and matching module is used to lock the budget based on the unified value measurement result of the task, and combine the standardized protocol task requirements and node adaptability to complete the matching and distribution of tasks and execution nodes. The multi-dimensional acceptance control module is used to judge the task execution results according to the acceptance rules of the standardized protocol tasks, generate acceptance results and synchronize them to the smart contract automated settlement module. The smart contract automated settlement module is used to automatically clear and settle the value of the task through smart contracts based on the acceptance results and the unified value measurement results of the task.

[0012] The beneficial effects of this invention are: 1. In this invention, by adopting a cross-modal unified intelligent workload measurement rule to standardize the value conversion of human and machine workload, and using the system's native unified accounting unit to complete automated clearing and settlement, a unified value measurement of human cognitive labor and intelligent agent computing power consumption and automated clearing of human-machine hybrid collaborative workload are realized. This solves the problems of pricing dimension barriers between human physical attributes and digital intelligent agent computing power attributes, and the incommensurability of cross-modal subject workload.

[0013] 2. In this invention, by providing native, undifferentiated access ports for humans and intelligent agents, and by reducing the dimensionality of unstructured tasks to multidimensional protocols and completing dynamic scheduling and standardized acceptance, barrier-free access for heterogeneous human-machine entities and quantitative control of the entire task process are achieved, solving the problems of machine access failure, low scheduling efficiency and collaborative trust gaps in traditional collaborative systems. Attached Figure Description

[0014] Figure 1 This is the task lifecycle state machine transition diagram of the present invention; Figure 2 This is a flowchart of the global task protocol conversion engine of the present invention; Figure 3This is a schematic diagram of the predictive pricing and dynamic routing engine of the present invention; Figure 4 This is a flowchart of the cross-modal workload measurement and settlement process of the present invention; Figure 5 This is a structural diagram of the system device module of the present invention. Detailed Implementation

[0015] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

[0016] See appendix Figures 1-5 A human-machine hybrid collaborative scheduling method and system based on multi-dimensional task protocolization includes the following steps: S1: Build a collaborative system architecture that supports access from heterogeneous entities such as humans and intelligent agents, providing corresponding access ports for human users and digital intelligent agents respectively, and receiving unstructured task requests submitted by task initiators. S2: Through the multi-dimensional task protocol transformation engine, unstructured task requests are subjected to multi-dimensional semantic parsing and feature extraction, and transformed into machine-readable standardized protocol tasks, thus completing the multi-dimensional protocol dimensionality reduction processing of tasks. S3: By using preset cross-modal unified intelligent workload measurement rules, the value of tasks that complete protocol-based dimensionality reduction is standardized and measured, thereby achieving unified pricing of human cognitive labor and intelligent agent computing power consumption. S4: Based on the unified pricing results of the task, the budget is locked. Combining the dimensional requirements of the standardized protocol task with the adaptability of the candidate execution nodes, the task and execution nodes are matched and distributed. The execution nodes include human execution nodes and intelligent agent execution nodes. S5: In accordance with the multi-dimensional acceptance rules agreed upon in the standardized protocol task, the task results output by the matched execution nodes are accepted and judged, and corresponding acceptance results are generated; S6: Based on the acceptance results and the unified pricing results of the task, the task value is automatically cleared and the entire process is closed-loop controlled through smart contracts.

[0017] In one embodiment, the pricing dimension barrier between human physical attributes and digital intelligent agent computing power attributes is broken down, solving the problem of incommensurability of cross-modal agent workload. By providing native, undifferentiated access ports for humans and digital intelligent agents, the two types of agents are abstracted into equivalent general capability nodes, eliminating the pricing anchoring differences between physical identity and computing power attributes, and providing subject access support for unified value measurement and automated settlement of human-machine hybrid collaborative workload.

[0018] See appendix Figures 1-5 S11: Provides a visual graphical console access port for human users and a standardized communication protocol API access port for digital intelligent agents, enabling native, seamless access for heterogeneous subjects such as humans and intelligent agents.

[0019] In one embodiment, unstructured tasks are analyzed from four core dimensions: input requirements, output specifications, time constraints, and quality acceptance. These tasks are then mapped to standardized protocol tasks in the form of finite state machines. Deterministic rules are used to achieve quantitative control over the entire task process, enabling machine-level mutual recognition between the task initiator and the executor.

[0020] See appendix Figures 1-5 S21: Define the four core dimensions of the task: input requirements, output specification boundaries, time limit constraints, and quality acceptance standards. Use a finite state machine to complete the flow control of the entire task process and achieve machine-level mutual recognition between the task initiator and the executor.

[0021] In one embodiment, the computational power consumption of the intelligent agent, the complexity of task logic arrangement, the human cognitive load and skill scarcity indicators are combined with the system's endogenous conversion coefficient to complete the value conversion, generate the task value amount corresponding to the system's native unified accounting unit, and construct a cross-modal standardized value measurement model.

[0022] See appendix Figures 1-5 S31: Based on the computing power consumption of the intelligent agent corresponding to the comprehensive task, the complexity of the task logic arrangement, the cognitive load and skill scarcity of human intervention required for the task, the standardized value conversion of the completed task is used to generate the task value amount corresponding to the system's native unified accounting unit.

[0023] In one embodiment, based on historical performance data of all network nodes, the success rate of execution and delivery risk are predicted. With the goal of optimal global scheduling, the standardized protocol tasks are accurately matched with the appropriate execution nodes, thereby achieving efficient task distribution.

[0024] See appendix Figures 1-5 S41: Based on the historical performance data of all network nodes, predict the success rate and delivery risk of candidate execution nodes in completing standardized protocol tasks, and with the goal of global scheduling optimization, match the execution node with the highest adaptability to complete the task distribution.

[0025] In one embodiment, the compliance of the task result is determined according to the multi-dimensional acceptance rules agreed upon in the agreement. Once the acceptance is passed, an automated clearing and settlement process is triggered to ensure the immediacy and certainty of the task value transfer.

[0026] See appendix Figures 1-5S51: If the task result meets the multi-dimensional acceptance rules agreed upon in the standardized protocol task, the acceptance is deemed successful, triggering the subsequent clearing and settlement process; S52: If the acceptance rules are not met, the acceptance is deemed unsuccessful, triggering the task rollback and rescheduling process, and synchronously updating the performance records and credit data of the corresponding execution node.

[0027] In one embodiment, a rollback and rescheduling process is initiated for tasks that fail acceptance, and the performance records and credit data of the execution nodes are updated synchronously to complete the closed-loop management of tasks and the dynamic update of the node trust system.

[0028] See appendix Figures 1-5 S61: The entire process of budget locking, node bidding, and value clearing within the system is completed through the system's native unified accounting unit. After acceptance, the corresponding value amount is automatically transferred to the execution node through smart contracts. Only the fiat currency recharge and final value clearing outside the system are realized through the preset gateway.

[0029] In one embodiment, the entire process of budget locking, node bidding, and value clearing is completed using the system's native unified accounting unit. Value is automatically transferred through smart contracts, and fiat currency is only used as an external recharge and final settlement interface to achieve automated cross-modal value clearing.

[0030] See appendix Figures 1-5 The human-machine hybrid collaborative scheduling system based on multidimensional task protocolization is characterized by including: a heterogeneous subject access module, a multidimensional task protocolization conversion module, a cross-modal unified workload measurement module, a task scheduling and matching module, a multidimensional acceptance and control module, and a smart contract automated settlement module. The heterogeneous subject access module is used to provide standardized access ports for humans and intelligent agents, receive unstructured task requests, and support access registration of human and intelligent agent execution nodes. The multidimensional task protocol conversion module is used to perform multidimensional parsing and feature extraction on unstructured task requests, converting them into standardized protocol tasks that can be read by machines, thus completing the task protocol dimensionality reduction. The cross-modal unified workload measurement module has built-in cross-modal unified workload measurement rules. It is used to comprehensively consider the computing power consumption of intelligent agents, task orchestration complexity, human cognitive load and skill scarcity, to standardize the value conversion of protocol-based tasks, generate task value quotas in the system's native unified accounting unit, realize unified value measurement of human and machine workload, and break down the pricing dimension barriers of heterogeneous human and machine subjects. The task scheduling and matching module is used to lock the budget based on the unified value measurement results of the task, and combine the standardized protocol task requirements and node adaptability to complete the matching and distribution of tasks and execution nodes. The multi-dimensional acceptance and control module is used to judge the task execution results according to the acceptance rules of the standardized protocol tasks, generate acceptance results and synchronize them to the smart contract automated settlement module. The smart contract automated settlement module is used to automatically clear and settle the value of tasks based on the acceptance results and the unified value measurement results of the tasks through smart contracts.

[0031] In one embodiment, this human-machine hybrid collaborative scheduling system based on multi-dimensional task protocolization is supported by a four-layer decoupled network topology: identity and trust layer, economy and pricing layer, task protocol layer, and multimodal interaction layer. It achieves fully automated scheduling and settlement of the entire human-machine hybrid collaborative process through six modules: heterogeneous subject access, multi-dimensional task protocolization conversion, cross-modal unified workload measurement, task scheduling and matching, multi-dimensional acceptance control, and automated smart contract settlement. The core operating principle is as follows: The heterogeneous subject access module provides a visual interactive console for humans and a standardized protocol API access port for digital intelligent agents, completing unified registration of human and machine execution nodes and receiving unstructured task requests; the multi-dimensional task protocolization conversion module performs multi-dimensional analysis and feature extraction on unstructured tasks, transforming them into machine-readable standardized protocol tasks and generating a core task state vector. in Input a feature tensor to the task, representing the context, materials, and prerequisite dependent data required for the task; The structured paradigm and specification boundaries for the expected output of the task; Define the task's absolute time constraint function and the task's deadline and timeout penalty gradient. For the quality acceptance judgment hyperplane, it supports deterministic binary classification or continuous scoring logic; The cross-modal unified workload measurement module uses built-in measurement rules to: The protocol-based tasks are converted into a unified accounting unit of the system's native Work Token, where W(T) is the scalar value of the standard workload of the task; Estimated lexical computational power consumption for the agent to complete the task; The topological complexity and dependency depth of the task decomposition graph; Neurocognitive load and skill scarcity index for performing tasks for humans; These are the system's endogenous conversion coefficients for computing power consumption, orchestration complexity, and human cognitive intervention, respectively. The task scheduling and matching module is based on After budget locking is completed, the posterior probability of acceptance of task T executed by candidate node ni is calculated using Bayesian inference: Through: Construct an expected delivery cost model, in which For nodes Execute the task Expected delivery costs; For nodes Pre-response basic quotation; Sunk costs for rollback and rescheduling in the event of a failed task acceptance; This is the system time sensitivity weight adjustment factor; A nonlinear delay risk function characterizing the time delay loss; The system is minimized To achieve global optimization, dynamic routing and distribution of tasks and optimal human-machine execution nodes are implemented, and the multi-dimensional acceptance and control module follows... Acceptance hyperplane The rules define and validate the task execution results, synchronizing the findings. The smart contract automated settlement module relies on smart contracts and an immutable ledger, and processes the settlement based on the validation results. The system automatically completes task value clearing and settlement, and synchronously links with the identity trust layer to complete behavior recording on the blockchain, credit update and rights arbitration, realizing efficient scheduling, reliable acceptance and unified value clearing through human-machine hybrid collaboration.

[0032] The above embodiments are merely illustrative of the technical solutions of the present invention and are not intended to limit it. Anyone skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A human-machine hybrid collaborative scheduling method based on multi-dimensional task protocolization, characterized in that, Includes the following steps: S1: Build a collaborative system architecture that supports access from heterogeneous entities such as humans and intelligent agents, providing corresponding access ports for human users and digital intelligent agents respectively, and receiving unstructured task requests submitted by task initiators. S2: Through the multi-dimensional task protocol transformation engine, unstructured task requests are subjected to multi-dimensional semantic parsing and feature extraction, and transformed into machine-readable standardized protocol tasks, thus completing the multi-dimensional protocol dimensionality reduction processing of tasks. S3: By using preset cross-modal unified intelligent workload measurement rules, the value of tasks that complete protocol-based dimensionality reduction is standardized and measured, thereby achieving unified pricing of human cognitive labor and intelligent agent computing power consumption. S4: Based on the unified pricing result of the task, the budget is locked. Combining the dimensional requirements of the standardized protocol task with the adaptability of the candidate execution nodes, the task and execution nodes are matched and distributed. The execution nodes include human execution nodes and intelligent agent execution nodes. S5: In accordance with the multi-dimensional acceptance rules agreed upon in the standardized protocol task, the task results output by the matched execution nodes are accepted and judged, and corresponding acceptance results are generated; S6: Based on the acceptance results and the unified pricing results of the task, the task value is automatically cleared and the entire process is closed-loop controlled through smart contracts.

2. The human-machine hybrid collaborative scheduling method based on multi-dimensional task protocolization according to claim 1, characterized in that, The collaborative system architecture in S1 also includes the following steps: S11: Provides a visual graphical console access port for human users and a standardized communication protocol API access port for digital intelligent agents, enabling native, seamless access for heterogeneous subjects such as humans and intelligent agents.

3. The human-machine hybrid collaborative scheduling method based on multi-dimensional task protocolization according to claim 2, characterized in that, The standardization protocol task in S2 also includes the following steps: S21: Define the four core dimensions of the task: input requirements, output specification boundaries, time limit constraints, and quality acceptance standards. Use a finite state machine to complete the flow control of the entire task process and achieve machine-level mutual recognition between the task initiator and the executor.

4. The human-machine hybrid collaborative scheduling method based on multi-dimensional task protocolization according to claim 3, characterized in that, The cross-modal unified intelligent workload measurement rule in S3 also includes the following steps: S31: Based on the computing power consumption of the intelligent agent corresponding to the comprehensive task, the complexity of the task logic arrangement, the cognitive load and skill scarcity of human intervention required for the task, the standardized value conversion of the completed task is used to generate the task value amount corresponding to the system's native unified accounting unit.

5. The human-machine hybrid collaborative scheduling method based on multi-dimensional task protocolization according to claim 4, characterized in that, The matching and distribution of tasks and execution nodes in S4 also includes the following steps: S41: Based on the historical performance data of all nodes in the network, predict the success rate and delivery risk of candidate execution nodes in completing standardized protocol tasks, and with the goal of global scheduling optimization, match the execution node with the highest adaptability to complete the task distribution.

6. The human-machine hybrid collaborative scheduling method based on multi-dimensional task protocolization according to claim 4, characterized in that, The acceptance criteria in S5 also include the following steps: S51: If the task result meets the multi-dimensional acceptance rules agreed upon in the standardized protocol task, the acceptance is deemed successful, and the subsequent clearing and settlement process is triggered; S52: If the acceptance rules are not met, the acceptance is deemed unsuccessful, triggering the task rollback and rescheduling process, and synchronously updating the performance records and credit data of the corresponding execution node.

7. The human-machine hybrid collaborative scheduling method based on multi-dimensional task protocolization according to claim 4, characterized in that, The automated clearing and closed-loop management of the entire process in S6 also includes the following steps: S61: The entire process of budget locking, node bidding, and value clearing within the system is completed through the system's native unified accounting unit. After acceptance, the corresponding value amount is automatically transferred to the execution node through smart contracts. Fiat currency top-ups and final value clearing outside the system are only achieved through a preset gateway.

8. A human-machine hybrid collaborative scheduling system based on multi-dimensional task protocolization, characterized in that: include: Heterogeneous subject access module, multi-dimensional task protocol conversion module, cross-modal unified workload measurement module, task scheduling and matching module, multi-dimensional acceptance control module, and smart contract automated settlement module; The heterogeneous subject access module is used to provide standardized access ports for humans and intelligent agents, receive unstructured task requests, and support access registration of human and intelligent agent execution nodes. The multidimensional task protocol conversion module is used to perform multidimensional parsing and feature extraction on unstructured task requests, convert them into standardized protocol tasks that can be read by machines, and complete the task protocol dimensionality reduction. The cross-modal unified workload measurement module has built-in cross-modal unified workload measurement rules, which are used to comprehensively consider the computing power consumption of intelligent agents, task arrangement complexity, human cognitive load and skill scarcity, to standardize the value conversion of protocol-based tasks, generate the task value amount of the system's native unified accounting unit, realize the unified value measurement of human and machine workload, and break down the pricing dimension barrier of heterogeneous human and machine subjects. The task scheduling and matching module is used to lock the budget based on the unified value measurement result of the task, and combine the standardized protocol task requirements and node adaptability to complete the matching and distribution of tasks and execution nodes. The multi-dimensional acceptance control module is used to judge the task execution results according to the acceptance rules of the standardized protocol tasks, generate acceptance results and synchronize them to the smart contract automated settlement module. The smart contract automated settlement module is used to automatically clear and settle the value of the task through smart contracts based on the acceptance results and the unified value measurement results of the task.

Images