Composite robot heterogeneous collaborative operation system
By combining the task planning module and the collaborative control module with heterogeneous robot clusters and multimodal communication links, the problem of inflexible task allocation in composite robot systems is solved, enabling efficient and stable multi-robot collaborative operations and improving operational accuracy and efficiency.
Patent Information
- Application Number
- CN202512038220.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-31
- Publication Date
- 2026-02-24
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In existing composite robot systems, the fixed task allocation logic does not take into account real-time operation status and scene changes, resulting in untimely adjustment of the task priority of the visual inspection robot, leading to problems such as task backlog or resource waste.
Employing a task planning module, heterogeneous robot clusters, and collaborative control module, the system enables multi-robot collaborative operation through task decomposition, path planning and conflict avoidance, real-time data interaction, and dynamic adjustment. Utilizing the ROS2 communication protocol and a multimodal communication link of a 5G industrial private network + edge computing gateway, combined with an improved A* dynamic obstacle avoidance algorithm, the system ensures the real-time performance and collaborative nature of the operation.
It enables seamless collaboration among multiple robots, improves operational accuracy and stability, reduces the risk of overall task failure due to localized problems, shortens the operational cycle, reduces manpower and time costs, and forms a closed-loop operational execution and data optimization mechanism.
Smart Images
Figure CN121552323A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of robot control technology, specifically to a composite robot heterogeneous collaborative operation system. Background Technology
[0002] Heterogeneous collaborative operation of composite robots refers to a mode in which composite robots of different types and functions work together through communication interconnection and task scheduling systems to complete complex tasks that are difficult for a single robot to perform. Composite robots usually integrate modules such as mobile chassis, manipulators, and sensors. Heterogeneous collaboration emphasizes the complementarity of robots in terms of function, form, and motion characteristics. For example, a mobile manipulator is responsible for material grasping and transfer, a vision inspection robot completes accuracy verification, and a collaborative robotic arm performs assembly processes.
[0003] Existing systems mostly adopt fixed task allocation logic, which only allocates tasks based on the robot's initial function labels, without considering real-time operation status and scene changes. This makes it impossible to quickly adjust the operation priority of the visual inspection robot, resulting in task backlog or resource waste. To address this, we propose a composite robot heterogeneous collaborative operation system. Summary of the Invention
[0004] To address the shortcomings of existing technologies, this invention provides a heterogeneous collaborative operation system for composite robots. This system solves the problems of existing systems that mostly use fixed task allocation logic, assign tasks based solely on the robot's initial function labels, fail to consider real-time operation status and scene changes, and cannot quickly adjust the task priority of visual inspection robots, leading to task backlog or resource waste.
[0005] To achieve the above objectives, the present invention provides the following technical solution: a heterogeneous collaborative operation system for composite robots, comprising a task planning module, a heterogeneous robot cluster, and a collaborative control module. The collaborative control module establishes a bidirectional signal connection with the task planning module and the heterogeneous robot cluster, wherein the heterogeneous robot cluster includes robots of various functional types. The system achieves heterogeneous collaborative operation through the following steps:
[0006] S1. Task parsing and allocation: After receiving the task, the task planning module breaks it down into multiple sub-tasks and allocates them to the corresponding robots based on the functional characteristics of each robot.
[0007] S2. Path planning and conflict avoidance: Based on the map information of the work scenario, the optimal work path is planned, and the collaborative control module avoids path conflicts and work interference through dynamic obstacle avoidance algorithms.
[0008] S3. Heterogeneous collaborative execution: Each robot executes sub-tasks synchronously, achieves real-time data interaction through multimodal communication links, and performs collaborative linkage control in cross-operation links;
[0009] S4. Operation status monitoring and dynamic adjustment: The collaborative control module collects robot operation data in real time and dynamically adjusts sub-tasks and paths when abnormalities occur;
[0010] S5. Acceptance of work results and archiving of data: After the work is completed, the quality of the results is checked, and the relevant data is archived after acceptance.
[0011] Preferably, the collaborative control module establishes a bidirectional signal connection through the ROS2 communication protocol to ensure the compatibility and real-time performance of data transmission between the module and the robot cluster.
[0012] Preferably, the heterogeneous robot cluster includes at least an industrial robotic arm, an AGV handling robot, and an inspection drone, which are respectively adapted to sub-tasks such as precision assembly, grasping and handling, and path inspection. The system integrates multiple functional types of robots such as industrial robotic arms, AGV handling robots, and inspection drones. Through precise sub-task allocation and functional adaptation, it can simultaneously complete composite operations such as grasping and handling, precision assembly, and path inspection. Compared with single-type robot operations, there is no need to configure multiple independent systems. It can flexibly cope with complex operation scenarios in multiple fields such as intelligent manufacturing and warehousing logistics, and greatly expand the application scope.
[0013] Preferably, the sub-tasks in S1 include grasping and handling, path inspection, and precision assembly. The task planning module matches the sub-task allocation according to the robot's load capacity and working range. The grasping and handling sub-task is assigned to the AGV handling robot, the path inspection sub-task is assigned to the inspection drone, and the precision assembly sub-task is assigned to the industrial robotic arm. Relying on the ROS2 communication protocol and the multimodal communication link of 5G industrial private network + edge computing gateway, millisecond-level data interaction between the module and the robot and between robots is realized. The cross-operation link adopts the linkage sequence of AGV positioning, robotic arm grasping and assembly, and drone inspection and verification. With the improved A* dynamic obstacle avoidance algorithm, the operation interference and path conflict are effectively avoided, ensuring smooth connection of multi-robot collaborative actions and significantly improving operation accuracy and stability.
[0014] Preferably, the work scenario map in S2 is a 3D point cloud map, the optimal work path is a time window optimized path, and the dynamic obstacle avoidance algorithm adopts an improved A* algorithm. Conflict avoidance is achieved through the robot's real-time position coordinates and work sequence. The collaborative control module monitors the work progress, execution accuracy, and equipment status in real time. When task delays, accuracy deviations, or equipment failures occur, a backup robot can be quickly switched or the path can be replanned. Compared with a fixed work mode, abnormal handling can be completed without interrupting the overall work process, greatly reducing the risk of overall task failure due to local problems and improving the system's fault tolerance and continuous operation capability under complex working conditions.
[0015] Preferably, the multimodal communication link in S3 consists of a 5G industrial private network and an edge computing gateway, with data interaction latency controlled at the millisecond level to ensure real-time transmission of collaborative commands.
[0016] Preferably, the collaborative control sequence of the cross-operation links in S3 is AGV positioning, robotic arm grasping and assembly, and drone inspection and verification, ensuring the continuity of multi-robot operations. Through task decomposition and parallel execution, each robot carries out its own sub-tasks simultaneously, avoiding the time loss of a single robot's serial operation. At the same time, by using time windows to optimize the path and automated collaborative control, manual scheduling and intervention are reduced. Compared with traditional manual operation or single robot operation mode, the operation cycle is shortened by more than 30%, which greatly reduces labor input and time costs and improves the overall production efficiency.
[0017] Preferably, the abnormal situations mentioned in S4 include task delays exceeding a preset ratio, execution accuracy deviations exceeding a preset threshold, or equipment failures. The dynamic adjustment methods include switching the subtask to a backup robot and replanning the operation path.
[0018] Preferably, the quality inspection of the work results in S5 is achieved through the vision inspection unit carried by the inspection drone. The inspection range covers assembly accuracy, material transfer integrity and work area safety. After the work is completed, the work execution data, robot operation parameters, collaborative control logs and quality inspection reports are archived, providing accurate data support for subsequent task allocation strategy optimization, path planning algorithm iteration and robot performance debugging. This forms a closed-loop mechanism of work execution, data collection, analysis optimization and iterative application, so that the system's work capabilities can be continuously upgraded with the application scenario and maintain a highly efficient and stable operating state in the long term.
[0019] Preferably, the archived data includes job execution data, robot operating parameters, collaborative control logs, and quality inspection reports, providing data support for subsequent task allocation optimization and path planning iteration.
[0020] In summary, the technical effects and advantages of this invention are as follows:
[0021] 1. In this invention, the system integrates multiple types of robots such as industrial robotic arms, AGV handling robots, and inspection drones. Through precise sub-task allocation and functional adaptation, it can simultaneously complete complex operations such as grasping and handling, precise assembly, and path inspection. Compared with single-type robot operations, it does not require the additional configuration of multiple independent systems. It can flexibly cope with complex operation scenarios in multiple fields such as intelligent manufacturing and warehousing logistics, and greatly expand the application scope.
[0022] 2. In this invention, relying on the ROS2 communication protocol and the multimodal communication link of 5G industrial private network + edge computing gateway, millisecond-level data interaction between modules and robots and between robots is realized. In the cross-operation link, the linkage sequence of AGV positioning, robotic arm grasping and assembly, and drone inspection and verification is adopted. With the improved A* dynamic obstacle avoidance algorithm, the operation interference and path conflict are effectively avoided, ensuring smooth connection of multi-robot collaborative actions, and significantly improving the operation accuracy and stability.
[0023] 3. In this invention, the collaborative control module monitors the work progress, execution accuracy and equipment status in real time. When there is a task delay, accuracy deviation or equipment failure, it can quickly switch to a backup robot or replan the path. Compared with the fixed operation mode, it can complete the abnormal handling without interrupting the overall operation process, which greatly reduces the risk of overall task failure caused by local problems and improves the system's fault tolerance and continuous operation capability under complex working conditions.
[0024] 4. In this invention, by decomposing and executing tasks in parallel, each robot carries out its own sub-tasks simultaneously, avoiding the time loss of a single robot performing serial operations. At the same time, by using time windows to optimize paths and automated collaborative control, manual scheduling and intervention are reduced. Compared with traditional manual operation or single robot operation mode, the operation cycle is shortened by more than 30%, significantly reducing manpower input and time costs, and improving overall production efficiency.
[0025] 5. In this invention, after the task is completed, the task execution data, robot operating parameters, collaborative control logs and quality inspection reports are archived, providing accurate data support for subsequent task allocation strategy optimization, path planning algorithm iteration and robot performance debugging, forming a closed-loop mechanism of task execution, data collection, analysis optimization and iterative application, so that the system's operation capability can be continuously upgraded with the application scenario and maintain a highly efficient and stable operating state in the long term. Attached Figure Description
[0026] Figure 1 This is a schematic diagram of a heterogeneous collaborative operation system for composite robots according to the present invention;
[0027] Figure 2 This is a flowchart illustrating a heterogeneous collaborative operation system for composite robots according to the present invention. Detailed Implementation
[0028] 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.
[0029] refer to Figures 1-2 The illustrated heterogeneous collaborative operation system for composite robots includes a task planning module, a heterogeneous robot cluster, and a collaborative control module. The collaborative control module establishes a bidirectional signal connection with the task planning module and the heterogeneous robot cluster. The heterogeneous robot cluster comprises robots of various functional types. The system achieves heterogeneous collaborative operation through the following steps:
[0030] S1. Task parsing and allocation: After receiving the task, the task planning module breaks it down into multiple sub-tasks and allocates them to the corresponding robots based on the functional characteristics of each robot.
[0031] S2. Path planning and conflict avoidance: Based on the map information of the work scenario, the optimal work path is planned, and the collaborative control module avoids path conflicts and work interference through dynamic obstacle avoidance algorithms.
[0032] S3. Heterogeneous collaborative execution: Each robot executes sub-tasks synchronously, achieves real-time data interaction through multimodal communication links, and performs collaborative linkage control in cross-operation links;
[0033] S4. Operation status monitoring and dynamic adjustment: The collaborative control module collects robot operation data in real time and dynamically adjusts sub-tasks and paths when abnormalities occur;
[0034] S5. Acceptance of work results and archiving of data: After the work is completed, the quality of the results is checked, and the relevant data is archived after acceptance.
[0035] The collaborative control module establishes a bidirectional signal connection through the ROS2 communication protocol to ensure the compatibility and real-time performance of data transmission between the module and the robot cluster.
[0036] The heterogeneous robot cluster includes at least industrial robotic arms, AGV handling robots, and inspection drones, each adapted to sub-tasks such as precision assembly, grasping and handling, and path inspection. The system integrates multiple functional types of robots, such as industrial robotic arms, AGV handling robots, and inspection drones. Through precise sub-task allocation and functional adaptation, it can simultaneously complete composite operations such as grasping and handling, precision assembly, and path inspection. Compared with single-type robot operations, it does not require the additional configuration of multiple independent systems and can flexibly cope with complex operation scenarios in multiple fields such as intelligent manufacturing and warehousing logistics, greatly expanding the scope of application.
[0037] The S1 sub-tasks include grasping and handling, path inspection, and precision assembly. The task planning module matches the sub-task allocation based on the robot's load capacity and operating range. The grasping and handling sub-task is assigned to the AGV handling robot, the path inspection sub-task is assigned to the inspection drone, and the precision assembly sub-task is assigned to the industrial robotic arm. Relying on the ROS2 communication protocol and the multimodal communication link of 5G industrial private network + edge computing gateway, millisecond-level data interaction between the module and the robot, and between the robots, is realized. The cross-operation links adopt the linkage sequence of AGV positioning, robotic arm grasping and assembly, and drone inspection and verification. With the improved A* dynamic obstacle avoidance algorithm, the operation interference and path conflict are effectively avoided, ensuring smooth connection of multi-robot collaborative actions and significantly improving operation accuracy and stability.
[0038] In S2, the work scene map is a 3D point cloud map, the optimal work path is a time window optimized path, and the dynamic obstacle avoidance algorithm adopts the improved A* algorithm. Conflict avoidance is achieved by using the robot's real-time position coordinates and work sequence. The collaborative control module monitors the work progress, execution accuracy, and equipment status in real time. When task delays, accuracy deviations, or equipment failures occur, a backup robot can be quickly switched or the path can be replanned. Compared with the fixed work mode, abnormal handling can be completed without interrupting the overall work process, greatly reducing the risk of overall task failure due to local problems and improving the system's fault tolerance and continuous operation capability under complex working conditions.
[0039] The multimodal communication link in S3 consists of a 5G industrial private network and an edge computing gateway, with data interaction latency controlled at the millisecond level, ensuring the real-time transmission of collaborative commands.
[0040] In S3, the coordinated control sequence of cross-operation links is AGV positioning, robotic arm grasping and assembly, and drone inspection and verification. This ensures the continuity of multi-robot operations. Through task decomposition and parallel execution, each robot carries out its own sub-tasks simultaneously, avoiding the time loss of a single robot's serial operation. At the same time, by using time windows to optimize paths and automated collaborative control, manual scheduling and intervention are reduced. Compared with traditional manual operation or single robot operation mode, the operation cycle is shortened by more than 30%, which greatly reduces labor input and time costs and improves the overall production efficiency.
[0041] Among them, abnormal situations in S4 include task delays exceeding the preset ratio, execution accuracy deviations exceeding the preset threshold, or equipment failures. Dynamic adjustment methods include switching subtasks to backup robots and replanning operation paths.
[0042] In S5, the quality inspection of the work results is achieved through the vision inspection unit carried by the inspection drone. The inspection range covers assembly accuracy, material transfer integrity and work area safety. After the work is completed, the work execution data, robot operation parameters, collaborative control logs and quality inspection reports are archived, providing accurate data support for subsequent task allocation strategy optimization, path planning algorithm iteration and robot performance debugging. This forms a closed-loop mechanism of work execution, data collection, analysis optimization and iterative application, so that the system's work capabilities can be continuously upgraded with the application scenario and maintain a highly efficient and stable operating state in the long term.
[0043] The archived data includes job execution data, robot operating parameters, collaborative control logs, and quality inspection reports, providing data support for subsequent task allocation optimization and path planning iteration.
[0044] I. Preliminary Preparations
[0045] (I) System Hardware Deployment and Configuration
[0046] Core module deployment:
[0047] Task planning module: Deployed on the workshop control center server (CPU: Intel Xeon Gold 6330, memory 32GB), equipped with dedicated software for task parsing and allocation, supporting multi-task parallel processing.
[0048] Collaborative control module: It adopts an industrial-grade edge computing gateway (supporting ROS2 communication protocol) to establish a fixed connection with each robot and task planning module through a wired network to ensure the stability of signal transmission.
[0049] Heterogeneous robot cluster: configured with 6 AGV handling robots, 4 six-axis industrial robotic arms, and 2 inspection drones, while reserving 2 spare robotic arms as redundancy.
[0050] Communication and map building:
[0051] A 5G industrial private network was built to cover the entire workshop, and combined with edge computing gateways to form a multimodal communication link to ensure that data interaction latency is ≤5ms.
[0052] Workshop environmental data is collected using lidar and vision sensors to construct a 1:1 scale 3D point cloud map, marking key operational nodes and obstacle locations such as material areas, assembly areas, and testing areas.
[0053] (II) Software and Parameter Presets
[0054] Install system control software, integrating task allocation algorithms, improved A* dynamic obstacle avoidance algorithms, and collaborative linkage control programs, and preset sub-task types and robot function matching rules.
[0055] Set anomaly detection threshold: a task delay exceeding 10% or an execution accuracy deviation exceeding 0.5mm is considered anomaly; configure collaborative linkage timing parameters to ensure that the connection time of each link in "AGV positioning → robotic arm grasping and assembly → drone inspection and verification" is ≤2s.
[0056] II. Implementation of Core System Processes
[0057] (I) S1: Task Analysis and Allocation
[0058] The workshop control center issued the work task through the task planning module: "Complete the assembly of 100 sets of automotive gearbox housings and gears within 3 hours, and simultaneously complete the material transfer trajectory inspection and assembly quality inspection."
[0059] The task planning module breaks down the sub-tasks: "grabbing and transporting gearbox housing and gears" is assigned to 6 AGV transport robots, "precise assembly of gears and housings" is assigned to 4 industrial robotic arms, and "material transfer path inspection + assembly quality inspection" is assigned to 2 inspection drones.
[0060] Based on the operating range and real-time status of each robot, the AGV is responsible for the round-trip transportation from the material area to the assembly area, the robotic arm is fixed at the assembly station to perform assembly operations, and the drone covers the inspection nodes of the material area and the assembly area according to the preset route.
[0061] (ii) S2: Path planning and conflict avoidance
[0062] The collaborative control module calls up the workshop's 3D point cloud map and, in conjunction with the time requirements of the task, plans the optimal path for each AGV within a time window, avoiding fixed obstacles such as workshop pillars and conveyor belts, to ensure that multiple AGVs can transfer without conflict.
[0063] By using an improved A* dynamic obstacle avoidance algorithm, the GPS positioning coordinates of each robot are obtained in real time (update frequency 10Hz), and a low-altitude inspection path (height 3m, speed 2m / s) is planned for the UAV to avoid spatial interference with AGV and robotic arm.
[0064] The planned paths and obstacle avoidance rules are distributed to each robot to complete the path preloading before the operation.
[0065] (III) S3: Heterogeneous Collaborative Execution of Tasks
[0066] All robots start working simultaneously: AGVs pick up gearbox housings and gears from the material area according to the planned path, transfer them to the assembly station, and achieve ±0.1mm accuracy positioning through the laser positioning module, sending a "positioning signal" to the collaborative control module.
[0067] After receiving signals through the 5G industrial private network, the collaborative control module sends assembly instructions to the corresponding industrial robotic arm. The robotic arm completes the precise assembly of the gear and the housing based on the preset program. During the assembly process, the force control sensor provides real-time feedback of pressure data to ensure assembly accuracy.
[0068] Once assembly is complete, the robotic arm sends an "assembly complete signal," and the drone immediately flies to the workstation, uses the vision inspection unit to capture assembly details, and transmits the images to the collaborative control module for real-time image analysis to complete the inspection and verification.
[0069] Throughout the process, the multimodal communication link ensures millisecond-level signal interaction at each stage, guaranteeing the seamless timing of AGV arrival, robotic arm assembly, and drone verification.
[0070] (iv) S4: Operation status monitoring and dynamic adjustment
[0071] The collaborative control module collects real-time operational data from each robot: the AGV's transfer progress, the robotic arm's assembly accuracy, and the drone's inspection coverage.
[0072] When a motor malfunctions in an industrial robotic arm, the system immediately identifies it as an equipment malfunction, quickly switches the remaining assembly tasks of that robotic arm to a backup robotic arm, and adjusts the transfer rhythm of the corresponding AGV to avoid material accumulation and prevent interruption of the overall operation process.
[0073] When a AGV experiences a delay of more than 10% due to low battery power, the collaborative control module replans the path of another idle AGV, takes over its transfer task, and adjusts the drone inspection path to temporarily cover the original transfer node of the AGV, ensuring that no material transfer trajectory is missed.
[0074] (v) S5: Acceptance of work results and archiving of data
[0075] After all sub-tasks are completed, two inspection drones will conduct a comprehensive quality inspection on 100 assembled gearboxes. The inspection scope includes gear meshing clearance, housing assembly flatness, and surface scratches. The inspection data will be transmitted back to the control center in real time.
[0076] The task planning module summarizes and analyzes the test results, determines that all 100 sets of products meet the quality standards, and completes the work acceptance.
[0077] The system automatically archives job data, robot operating parameters, and collaborative control logs to form a complete job file.
[0078] III. Evaluation of Implementation Results
[0079] This implementation successfully completed the integrated operation of 100 automotive transmissions, with an overall operation cycle of 2 hours and 50 minutes, a 35% reduction compared to the traditional manual + single-robot operation mode. Assembly accuracy was controlled within ±0.05mm, and the quality inspection pass rate was 100%. The response time for handling anomalies was ≤30 seconds, and no operation was interrupted due to equipment failure or path conflicts. The system operated stably, with smooth collaboration among the robots, significantly reducing labor input and time costs, and fully meeting the high-efficiency operation requirements of a smart manufacturing workshop.
[0080] Finally, it should be noted that the above description is only 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 heterogeneous collaborative operation system for composite robots, characterized in that: The system includes a task planning module, a heterogeneous robot cluster, and a collaborative control module. The collaborative control module establishes a bidirectional signal connection with the task planning module and the heterogeneous robot cluster. The heterogeneous robot cluster contains robots of various functional types. The system achieves heterogeneous collaborative operation through the following steps: S1. Task parsing and allocation: After receiving the task, the task planning module breaks it down into multiple sub-tasks and allocates them to the corresponding robots based on the functional characteristics of each robot. S2. Path planning and conflict avoidance: Based on the map information of the work scenario, the optimal work path is planned, and the collaborative control module avoids path conflicts and work interference through dynamic obstacle avoidance algorithms. S3. Heterogeneous collaborative execution: Each robot executes sub-tasks synchronously, achieves real-time data interaction through multimodal communication links, and performs collaborative linkage control in cross-operation links; S4. Operation status monitoring and dynamic adjustment: The collaborative control module collects robot operation data in real time and dynamically adjusts sub-tasks and paths when abnormalities occur; S5. Acceptance of work results and archiving of data: After the work is completed, the quality of the results is checked, and the relevant data is archived after acceptance.
2. The heterogeneous collaborative operation system for composite robots according to claim 1, characterized in that: The collaborative control module establishes a bidirectional signal connection through the ROS2 communication protocol to ensure the compatibility and real-time performance of data transmission between the module and the robot cluster.
3. The heterogeneous collaborative operation system for composite robots according to claim 1, characterized in that: The heterogeneous robot cluster includes at least industrial robotic arms, AGV handling robots, and inspection drones, which are adapted to sub-tasks such as precision assembly, grasping and handling, and path inspection, respectively.
4. The heterogeneous collaborative operation system of composite robots according to claim 1, characterized in that: The sub-tasks described in S1 include grasping and handling, path inspection, and precision assembly. The task planning module matches the sub-task allocation according to the robot's load capacity and working range. The grasping and handling sub-task is assigned to the AGV handling robot, the path inspection sub-task is assigned to the inspection drone, and the precision assembly sub-task is assigned to the industrial robotic arm.
5. A heterogeneous collaborative operation system for composite robots according to claim 1, characterized in that: The work scenario map described in S2 is a 3D point cloud map, the optimal work path is a time window optimized path, and the dynamic obstacle avoidance algorithm adopts the improved A* algorithm, which achieves conflict avoidance by using the robot's real-time position coordinates and work sequence.
6. The heterogeneous collaborative operation system of composite robots according to claim 1, characterized in that: The multimodal communication link described in S3 consists of a 5G industrial private network and an edge computing gateway, with data interaction latency controlled at the millisecond level to ensure real-time transmission of collaborative commands.
7. A heterogeneous collaborative operation system for composite robots according to claim 1, characterized in that: In S3, the coordinated control sequence for cross-operation links is AGV positioning, robotic arm grasping and assembly, and drone inspection and verification, ensuring the continuity of multi-robot operations.
8. A heterogeneous collaborative operation system for composite robots according to claim 1, characterized in that: The abnormal situations described in S4 include task delays exceeding a preset ratio, execution accuracy deviations exceeding a preset threshold, or equipment failures. Dynamic adjustment methods include switching subtasks to backup robots and replanning work paths.
9. A heterogeneous collaborative operation system for composite robots according to claim 1, characterized in that: In S5, the quality inspection of work results is achieved through the visual inspection unit carried by the inspection drone, and the inspection range covers assembly accuracy, material transfer integrity and work area safety.
10. A heterogeneous collaborative operation system for composite robots according to claim 1, characterized in that: The archived data includes job execution data, robot operating parameters, collaborative control logs, and quality inspection reports, providing data support for subsequent task allocation optimization and path planning iteration.