Robot calling and man-machine cooperation method and system based on remote link

Abstract

The invention relates to the technical field of robots, and discloses a robot calling and man-machine cooperation method and system based on remote linking, and the method comprises the steps: a robot registers to a platform as a callable node; the platform is a task matching robot and establishes a remote link; the robot automatically executes a task firstly; through real-time state evaluation, a remote cooperation process is triggered when an abnormal condition is met; the remote cooperation main body performs cooperation or takeover control through the operation interface; and after ending, recovering the automatic mode and recording data. The system comprises corresponding function units for implementing the steps. According to the method and the system, intelligent switching of'automatic execution is dominated and abnormal cooperation is taken as the bottom 'is realized, the task success rate of a complex scene is improved, and a complete technical closed loop supporting robot resource sharing and commercial operation is constructed through session vouchers, resource isolation, consumption measurement and an automatic settlement and accounting mechanism.

Description

Technical Field

[0001] This invention relates to the field of robotics, specifically to a method and system for robot invocation and human-machine collaboration based on remote links. Background Technology

[0002] With the continuous advancement of automation and artificial intelligence technologies, the application of robots in logistics warehousing, security inspection, facility maintenance, industrial operation and maintenance, and modern agriculture is becoming increasingly sophisticated. These complex, dynamic, and partially unstructured work scenarios place higher demands on the reliability, adaptability, and economy of robots in performing tasks.

[0003] Currently, robots operating in the aforementioned environments primarily rely on two technological approaches: The first is a fully autonomous operation mode, where the robot independently completes tasks based entirely on its local perception, decision-making, and control systems. However, faced with high environmental uncertainty, ambiguous task boundaries, and unforeseen extreme situations (such as complex terrain, dynamic obstacles, and occasional equipment malfunctions), the perception and cognitive capabilities of purely autonomous systems are often limited, easily leading to task interruptions, execution failures, and even safety accidents. The second approach is a purely remote teleoperation mode, where a human operator directly and remotely controls the robot throughout the entire process via real-time video streams. While this method can handle complex emergencies, it is highly dependent on human intervention, resulting in inherent drawbacks such as heavy operator workload, high skill requirements, sensitivity to communication latency, high labor costs, and difficulty in large-scale deployment under a "one person, one robot" model.

[0004] Furthermore, with the increasing prevalence of robot assets, achieving efficient sharing of robot resources, on-demand services, and commercial operation has become an urgent need. Existing robot management systems primarily focus on monitoring the status and managing tasks of individual or fixed-formation robots, lacking a universal method to abstract heterogeneous robots into standardized "execution nodes" that can be uniformly scheduled and remotely invoked by a platform. More importantly, existing solutions lack a complete, secure, reliable, and sustainable service loop. Specifically, no system can yet support a dynamic and flexible model of "primarily automatic execution, supplemented by remote collaboration" while simultaneously achieving refined management of robot usage permissions, secure isolation of resources and data between different users or task sessions, and accurate measurement of service usage. Based on this, a commercial service model can be formed that features automatic settlement, flexible revenue distribution, and full-process traceability and auditability. This problem is particularly prominent in fields such as agricultural machinery, which are characterized by high risk, high value, and long-distance operations, limiting the large-scale promotion and commercialization of robot technology.

[0005] To address the aforementioned issues, this application proposes a method and system for robot invocation and human-machine collaboration based on remote links. Summary of the Invention

[0006] This invention aims to provide a method and system for robot invocation and human-robot collaboration based on remote links, primarily addressing the high failure rate of existing purely autonomous robots in complex, unstructured environments; the high cost and difficulty in scaling purely remote operation; and the lack of a system method that abstracts robots as callable nodes, enables secure switching between automatic execution and external collaboration, and supports shared rental, metering, and settlement.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0008] A method for robot invocation and human-robot collaboration based on remote links includes the following steps:

[0009] S1: Robot registration steps: Register the robot with the platform as a remotely invoked execution node and report the robot's capability parameters;

[0010] S2: Task matching and link establishment steps: In response to a task request, the platform acts as the task matching robot execution node and establishes a remote link with the matched robot.

[0011] S3: Automatic execution steps: After the remote connection is established, control the robot to perform preset actions related to the task in automatic execution mode;

[0012] S4: Status assessment and collaboration triggering steps: During the automatic execution of the robot, its execution status is assessed in real time, and a remote collaboration process is triggered when the assessment result meets the preset abnormal conditions.

[0013] S5: Remote collaboration step: Provide the robot's operation interface to the remote collaboration subject, enabling the remote collaboration subject to perform collaborative control or takeover control of the robot through the remote link;

[0014] S6: State recovery and recording steps: After the collaboration of the remote collaboration subject is completed, switch the control of the robot back to automatic execution mode or standby state, and record the execution data of this task.

[0015] Preferably, in step S4, the preset abnormal conditions include at least one of the following: environmental complexity exceeding a threshold, consecutive execution failures exceeding a threshold, detection of a security risk event, and receipt of a user's proactive request for collaboration.

[0016] Preferably, in step S5, the robot operates in a human-machine co-driving mode, where the robot continues to perform basic actions, and the remote collaborative entity provides auxiliary control commands or takeover control commands.

[0017] Preferably, before step S2, a session credential generation step is included, in which the platform generates a unique session credential for the task request or subscription request. The session credential is bound to usage permissions, usage duration, and task scope. The platform only allows the execution of step S2 and subsequent steps after verifying the validity of the session credential.

[0018] Preferably, in step S5, the platform provides a multi-tenant isolation strategy for different sessions, the isolation strategy including at least isolating the control command channel and the sensor data media channel.

[0019] Preferably, in steps S5 and / or S6, the platform measures the usage during the session, and the measurement dimensions include at least one of session duration, number of control commands, data throughput and computing power consumption, and generates a billing record based on the measurement results.

[0020] A robot invocation and human-robot collaboration system based on remote linking, comprising:

[0021] Robot execution unit, used to perform the automatic execution steps;

[0022] The platform scheduling unit is used to execute the robot registration step and the task matching and link establishment step;

[0023] The remote link management unit is used to establish and maintain the remote link between the platform and the robot;

[0024] A state assessment and switching unit is used to execute the state assessment and collaboration triggering steps and the state recovery and recording steps;

[0025] A remote human-computer interaction unit is used to implement the remote collaboration steps and provide an operation interface and control channel for the remote collaboration subject.

[0026] The shared rental and metering unit is used to generate session credentials, implement multi-tenant isolation strategies, perform session usage metering, and generate billing records.

[0027] Preferably, the remote human-machine interaction unit receives and presents multi-source sensor data from the robot, including visual images, positioning information, robot posture information, and workload information.

[0028] Preferably, the shared rental and metering unit is further configured to: perform automatic settlement based on the billing records, and distribute the settlement revenue to at least one of the robot asset provider, remote collaboration entity, platform operator and computing power node provider according to preset allocation rules, while generating audit records containing key session operations and events.

[0029] Preferably, the robot execution unit is deployed on an agricultural machinery robot, which includes a large harvester, a seeder, or a tractor; the state assessment and switching unit is configured to trigger the remote collaboration process when the agricultural machinery robot detects abnormal terrain, dense obstacles, or abnormal workload during the execution of harvesting, traveling, or working tasks; the multi-source sensor data received and presented by the remote human-machine interaction unit includes at least visual images, satellite positioning information, body attitude information, and workload information.

[0030] Beneficial effects:

[0031] (1) Through a dynamic switching mechanism of "automatic execution as the main method and abnormal triggering collaboration", the high efficiency of robot autonomous operation and the flexibility of human (or advanced decision-making body) in dealing with complex situations are effectively integrated. When the robot encounters anomalies that exceed its autonomous processing capabilities, remote and precise intervention can be seamlessly introduced, thereby significantly improving the task completion rate and the robustness of the overall system in unstructured and dynamic environments;

[0032] (2) By introducing session credentials, multi-tenant isolation, usage metering, and automatic settlement and revenue sharing mechanisms, this solution abstracts robots from independent devices into a unified security service that can be scheduled by the platform. This not only realizes the shared rental of robot resources and refined permission management, but also forms a complete business closed loop, supports service metering, fee settlement, and operation auditing, and provides a key technical foundation for the large-scale and commercial operation of robot services;

[0033] (3) The “human-machine co-driving” mode of this solution is not a simple transfer of control, but allows the robot and the remote subject to collaborate in a hierarchical manner. The robot is continuously responsible for the underlying stability control and basic actions, while the remote subject focuses on high-level decision-making and abnormal handling. This collaborative mode not only ensures operational safety and reduces the cognitive load and skill requirements of the remote operator, but also makes the “one person, multiple machines” scheduling mode possible, which greatly improves human resource efficiency;

[0034] (4) Taking agricultural machinery as an example, this solution provides a practical and feasible remote support method for solving problems such as changing terrain, sudden obstacles, and abnormal loads in field operations. By integrating multi-source sensor data and supporting real-time intervention by remote experts, it can effectively cope with complex field conditions while ensuring operational safety, and ensure the continuous and efficient operation of high-value agricultural machinery and equipment, demonstrating the huge application potential of the technology in vertical fields such as agricultural modernization. Attached Figure Description

[0035] Figure 1 This is a flowchart of the steps of the robot invocation and human-machine collaboration method based on remote link of the present invention;

[0036] Figure 2This is a flowchart of the robot invocation and human-machine collaboration method based on remote link of the present invention;

[0037] Figure 3 This is a schematic diagram of the robot invocation and human-machine collaboration system based on remote link according to the present invention;

[0038] Figure 4 This is a flowchart illustrating the remote human-computer interaction unit of the present invention. Detailed Implementation

[0039] 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 embodiments of the present invention, and not all embodiments. 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.

[0040] Please see Figure 1-4 A method for robot invocation and human-machine collaboration based on remote links includes the following steps:

[0041] S1: Robot registration steps: Register the robot with the platform as a remotely invoked execution node and report the robot's capability parameters;

[0042] S2: Task matching and link establishment steps: In response to a task request, the platform acts as the task matching robot execution node and establishes a remote link with the matched robot.

[0043] S3: Automatic execution steps: After the remote connection is established, control the robot to perform preset actions related to the task in automatic execution mode;

[0044] S4: Status assessment and collaboration triggering steps: During the automatic execution of the robot, its execution status is assessed in real time, and a remote collaboration process is triggered when the assessment result meets the preset abnormal conditions.

[0045] S5: Remote collaboration step: Provide the robot's operation interface to the remote collaboration subject, enabling the remote collaboration subject to perform collaborative control or takeover control of the robot through the remote link;

[0046] S6: State recovery and recording steps: After the collaboration of the remote collaboration subject is completed, switch the control of the robot back to automatic execution mode or standby state, and record the execution data of this task;

[0047] By virtualizing the physical robot into a platform-manageable resource (S1), and after task scheduling and connection establishment (S2), the system enters the execution phase centered on robot autonomy (S3). The core intelligence of the system is reflected in S4, which dynamically switches from "automatic mode" to "remote collaboration mode" when a preset anomaly occurs through real-time status evaluation (S5). After the task or collaboration is completed, the system will properly restore and record the data (S6), forming a complete task lifecycle management closed loop. This establishes the basic paradigm of "registration-connection-autonomous execution-intelligent evaluation-collaboration switching-restore recording," breaking down the barriers between autonomous operation and remote control, and realizing the dynamic and orderly integration of the two based on actual working conditions, providing a framework for all subsequent specific features.

[0048] More specifically, in step S4, the preset abnormal conditions include at least one of the following: environmental complexity exceeding a threshold, consecutive execution failures exceeding a threshold, detection of a security risk event, and receipt of a user's proactive request for collaboration. This clarifies the specific dimensions of the evaluation: environmental complexity (e.g., terrain, obstacle density), self-performance (consecutive failures), security (risk events), and user intent (proactive request). The platform or robot's local evaluation module will continuously monitor data in these dimensions and compare it with the thresholds. This improves the rationality and timeliness of the system response, avoids unnecessary collaboration (reducing labor costs) or overlooks necessary interventions (ensuring task success and safety), and makes the timing of human-machine switching more intelligent and objective.

[0049] More specifically, in step S5, the robot operates in a human-robot co-driving mode. The robot continues to perform basic actions, while the remote collaborating entity provides auxiliary control commands or takeover control commands. In this mode, control is not completely transferred: the robot's underlying control loop (such as balance, obstacle avoidance, and basic trajectory tracking) continues to operate, responsible for the stability and safety of "driving"; the remote collaborating entity (human or AI) provides higher-level commands (such as "go around that puddle" or "go to point A") or direct takeover commands (such as "turn the steering wheel 10 degrees to the left"). This allows the operator to focus on high-level decision-making and handling of anomalies without needing to perform millimeter-level full-process control, making "one person, multiple robots" possible and greatly improving human resource efficiency. Simultaneously, the robot's underlying autonomous safety capabilities are preserved, enhancing the safety of the collaborative process.

[0050] More specifically, prior to step S2, a session credential generation step is included. The platform generates a unique session credential for each task or subscription request. This session credential is bound to usage permissions, usage duration, and task scope. The platform only allows the execution of step S2 and subsequent steps after verifying the validity of the session credential. The platform generates a unique, digital "key" (session credential) for each task or subscription, encoding policies such as "who can use it," "which robot to use," "how long to use it," and "what it can do." All subsequent steps (S2-S6) require verification of this "key" before proceeding, achieving refined and dynamic authorization management of robot resource access. This is crucial for operating robots as "services" rather than "devices," providing the technical prerequisite for pay-per-use, pay-per-duration, and pay-per-function billing models, and ensuring the security and controllability of the service.

[0051] More specifically, in step S5, the platform provides a multi-tenant isolation strategy for different sessions. This isolation strategy includes at least isolating the control command channel and the sensor data media channel. When multiple tasks or users are concurrent, the platform isolates the control command streams and sensor data media streams (such as video and point clouds) of different sessions at the network or virtual channel level, ensuring that they do not interfere with, spy on, or tamper with each other. This solves the core security and privacy issues when multiple users use the platform simultaneously. Isolation ensures accurate delivery of commands and prevents data leakage, which is the core technical guarantee for the platform to securely and reliably support high-concurrency shared services, meeting the security requirements of enterprise-level applications.

[0052] More specifically, in steps S5 and / or S6, the platform measures the usage during the session. The measurement dimensions include at least one of session duration, number of control commands, data throughput, and computing power consumption. Based on the measurement results, a billing record is generated. The platform acts like an "electricity meter," statistically analyzing the resources consumed during a session from multiple dimensions such as time (session duration), interaction depth (number of commands), data cost (throughput), and computing resources (computing power consumption), ultimately aggregating and generating a structured billing record. This achieves precise quantification and visualization of service costs. This transforms the business model from extensive "equipment leasing" to refined "service metering," making billing more fair and transparent. This provides the data foundation for automated settlement and diversified business models (such as billing by command count or by data flow).

[0053] A robot invocation and human-robot collaboration system based on remote linking, comprising:

[0054] Robot execution unit, used to perform the automatic execution steps;

[0055] The platform scheduling unit is used to execute the robot registration step and the task matching and link establishment step;

[0056] The remote link management unit is used to establish and maintain the remote link between the platform and the robot;

[0057] A state assessment and switching unit is used to execute the state assessment and collaboration triggering steps and the state recovery and recording steps;

[0058] A remote human-computer interaction unit is used to implement the remote collaboration steps and provide an operation interface and control channel for the remote collaboration subject.

[0059] The shared rental and metering unit is used to generate session credentials, implement multi-tenant isolation strategies, perform session usage metering, and generate billing records.

[0060] Each "unit" corresponds to one or a set of steps in the method: the "robot execution unit" corresponds to S3, the "platform scheduling unit" corresponds to S1 and S2, and so on. The "shared rental and metering unit" integrates and implements the functions of claims 4-6. Through collaborative work, each unit instantiates the method flow, thus providing a clear product protection scope corresponding to the method. It clarifies the components of the system and their functional divisions, ensuring that patent protection is not limited to the "flow" idea but also covers the specific system product that implements the flow, enhancing the practical value and strength of the patent.

[0061] More specifically, the remote human-machine interaction unit receives and presents multi-source sensor data from the robot. This multi-source sensor data includes visual images, positioning information, robot posture information, and workload information. By fusing, encoding, and transmitting the raw data (images, position, posture, and workload) from the robot's sensors, the data is ultimately presented to the operator in an intuitive form (such as video footage overlaid with posture instruments and electronic maps) on the remote operation interface. By providing multi-source, heterogeneous, but complementary sensor information, the remote operator's perception and understanding of the situation on-site is greatly enhanced, enabling them to make accurate judgments similar to those made on-site. This is the information foundation for achieving effective remote collaboration.

[0062] More specifically, the shared rental and metering unit is further configured to: perform automatic settlement based on the billing records, and distribute the settlement revenue to at least one of the robot asset provider, remote collaboration entity, platform operator, and computing power node provider according to preset allocation rules. Simultaneously, it generates audit records containing key session operations and events. Based on the billing records, the "shared rental and metering unit" automatically executes the revenue sharing logic, distributing revenue to each participant according to preset rules. Furthermore, it extracts key events in the session (such as collaboration trigger times, takeover instructions, and abnormal events) to generate immutable audit logs, completing the final link from "technical metering" to "business closed loop" and "responsibility traceability." Automatic settlement reduces operating costs and clarifies profit distribution; audit records provide a basis for dispute resolution and process traceability, establishing the platform's credibility and compliance, making them an indispensable part of commercial operation.

[0063] More specifically, the robot execution unit is deployed on an agricultural machinery robot, which includes large harvesters, seeders, or tractors. The state assessment and switching unit is configured to trigger the remote collaboration process when the agricultural machinery robot detects abnormal terrain, dense obstacles, or abnormal workload during harvesting, movement, or other tasks. The multi-source sensor data received and presented by the remote human-machine interaction unit includes at least visual images, satellite positioning information, machine posture information, and workload information. By limiting the execution unit to large agricultural equipment, the triggering conditions for state assessment are specified as "abnormal terrain, dense obstacles, and abnormal workload" specific to field operations. The interactive data also corresponds to the visual, satellite positioning, posture, and workload information relevant to agricultural operations. By deeply integrating a general platform solution with a vertical field with clear pain points (high-value equipment, complex field environment, and high requirements for operational continuity), the invention's practicality and industrial value are strongly supported. This demonstrates that the solution is not merely an abstract theory, but a key technology capable of solving practical problems such as remote operation and maintenance and reducing reliance on machine operators in agricultural modernization.

[0064] As can be seen from the above, the specific embodiments of the present invention are as follows:

[0065] This technical solution constructs a platform-based, service-oriented ecosystem for remote robot operation. Its core working principle begins with abstracting heterogeneous robots from the physical world into standardized "execution nodes" managed uniformly by the platform through a registration process, and reporting their capability parameters to form a virtualized resource pool. When the platform receives a task request, the scheduling unit matches the task requirements with the robot's capabilities and establishes a secure remote link between the successfully matched task and the robot. This link typically includes independent control command channels and sensor data media channels. After the link is established, the robot enters a primarily autonomous operation mode, executing preset tasks using its own algorithms, while continuously transmitting multi-source sensor data back to the platform for monitoring via the media channels. The system's key intelligence lies in its built-in state evaluation mechanism. This mechanism continuously analyzes the robot's execution status based on real-time data (such as environmental complexity, action execution success rate, and safety sensor signals). Once the evaluation results meet preset abnormal conditions (such as overly complex terrain, continuous failures, or user-initiated requests), it automatically triggers a dynamic switch in the working mode, initiating a remote collaboration process. During the collaboration phase, authorized remote operators or automated decision-making entities (remote collaboration subjects) can access the system through a dedicated interactive interface. This interface integrates and presents multi-dimensional information such as the robot's real-time video, pose, and load, providing a strong sense of remote presence. In the "human-robot co-driving" collaboration mode, the robot's underlying controller continues to operate to handle basic actions and ensure safety, while the remote collaboration subject sends higher-level auxiliary guidance instructions or direct, precise takeover instructions based on the situation. The two work together to complete complex operations. To support commercial sharing services, the platform generates a unique session credential bound to permissions and duration before the task begins. During collaboration, it implements network channel isolation for different sessions to ensure security, and accurately measures resource consumption such as session duration, number of instructions, and data traffic to generate billing data. Finally, after collaboration, the system smoothly restores the robot to autonomous mode and automatically completes settlement and accounting and generates audit records based on the measurement results, thus achieving a complete technical closed loop from robot resource virtualization, intelligent scheduling, adaptive human-robot collaboration to secure metering and settlement. This solution is particularly suitable for remote operation support of high-risk, high-value equipment such as agricultural machinery. Through the above mechanism, it can effectively cope with complex working conditions in the field and improve the success rate of operations and equipment utilization.

[0066] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for robot invocation and human-machine collaboration based on remote links, comprising the following steps: S1: Robot registration steps: Register the robot with the platform as a remotely invoked execution node and report the robot's capability parameters; S2: Task matching and link establishment steps: In response to a task request, the platform acts as the task matching robot execution node and establishes a remote link with the matched robot. S3: Automatic execution steps: After the remote connection is established, control the robot to perform preset actions related to the task in automatic execution mode; S4: Status assessment and collaboration triggering steps: During the automatic execution of the robot, its execution status is assessed in real time, and a remote collaboration process is triggered when the assessment result meets the preset abnormal conditions. S5: Remote collaboration step: Provide the robot's operation interface to the remote collaboration subject, enabling the remote collaboration subject to perform collaborative control or takeover control of the robot through the remote link; S6: State recovery and recording steps: After the collaboration of the remote collaboration subject is completed, switch the control of the robot back to automatic execution mode or standby state, and record the execution data of this task.

2. The method for robot invocation and human-machine collaboration based on remote link according to claim 1, characterized in that: In step S4, the preset abnormal conditions include at least one of the following: environmental complexity exceeding a threshold, consecutive execution failures exceeding a threshold, detection of a security risk event, and receipt of a user's proactive request for collaboration.

3. The method and system for robot invocation and human-machine collaboration based on remote link as described in claim 1, characterized in that: In step S5, the robot operates in human-machine co-driving mode, where the robot continues to perform basic actions, and the remote collaborating entity provides auxiliary control commands or takeover control commands.

4. The method and system for robot invocation and human-machine collaboration based on remote link as described in claim 1, characterized in that: Before step S2, there is also a session credential generation step, in which the platform generates a unique session credential for the task request or subscription request. The session credential is bound to the usage permissions, usage duration and task scope. The platform will only allow the execution of step S2 and subsequent steps after verifying that the session credential is valid.

5. The method for robot invocation and human-machine collaboration based on remote link according to claim 4, characterized in that: In step S5, the platform provides a multi-tenant isolation strategy for different sessions, the isolation strategy including at least isolating the control command channel and the sensor data media channel.

6. A method for robot invocation and human-machine collaboration based on remote linking according to claim 4 or 5, characterized in that: In steps S5 and / or S6, the platform measures the usage during the session. The measurement dimensions include at least one of session duration, number of control commands, data throughput, and computing power consumption, and generates billing records based on the measurement results.

7. A remote link-based robot invocation and human-robot collaboration system for implementing the method of any one of claims 1 to 6, characterized in that, include: Robot execution unit, used to perform the automatic execution steps; The platform scheduling unit is used to execute the robot registration step and the task matching and link establishment step; The remote link management unit is used to establish and maintain the remote link between the platform and the robot; A state assessment and switching unit is used to execute the state assessment and collaboration triggering steps and the state recovery and recording steps; A remote human-computer interaction unit is used to implement the remote collaboration steps and provide an operation interface and control channel for the remote collaboration subject. The shared rental and metering unit is used to generate session credentials, implement multi-tenant isolation strategies, perform session usage metering, and generate billing records.

8. A robot invocation and human-machine collaboration system based on remote link according to claim 7, characterized in that: The remote human-machine interaction unit receives and presents multi-source sensor data from the robot, including visual images, positioning information, robot posture information, and workload information.

9. A robot invocation and human-machine collaboration system based on remote link according to claim 7, characterized in that: The shared rental and metering unit is further configured to: perform automatic settlement based on the billing records, and distribute the settlement revenue to at least one of the robot asset provider, remote collaboration entity, platform operator and computing power node provider according to preset allocation rules, while generating audit records containing key session operations and events.

10. A robot invocation and human-machine collaboration system based on remote link according to claim 7, characterized in that: The robot execution unit is deployed on the agricultural machinery robot, which includes a large harvester, a seeder, or a tractor; the state assessment and switching unit is configured to trigger the remote collaboration process when the agricultural machinery robot detects abnormal terrain, dense obstacles, or abnormal workload during the harvesting, traveling, or operation tasks; the multi-source sensor data received and presented by the remote human-machine interaction unit includes at least visual images, satellite positioning information, body attitude information, and workload information.

Images