Multi-robot collaborative building operation method and system

Abstract

The invention aims to provide a multi-robot cooperative building operation method and system, and belongs to the technical field of intelligent building robot cooperative control, and the method comprises the steps: firstly obtaining a building operation task and a building three-dimensional model, decomposing the task into a plurality of sub-tasks based on the model, and generating an initial task allocation scheme according to the operation type and the spatial position of the subtask, and allocating the initial task allocation scheme to a plurality of robots. In the operation process, the operation state and environment data of each robot are collected in real time, and a dynamic conflict detection matrix is constructed. And when a path conflict or a resource conflict is detected, dynamically adjusting the conflict task according to a preset priority rule, generating an updated task allocation instruction, sending the updated task allocation instruction to the corresponding robot, and controlling the corresponding robot to execute the adjusted sub-task. According to the invention, through a dynamic conflict detection and intelligent adjustment mechanism, the problem of path and resource conflict of multiple robots in a dynamic building construction environment is effectively solved, and the construction efficiency and safety are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of collaborative control of intelligent construction robots, and in particular to a multi-robot collaborative construction operation method and system. Background Art

[0002] In the field of construction, with the introduction of robotics, multi-robot collaborative operations have become an important means of improving construction efficiency and quality. However, existing technologies for multi-robot collaborative construction operations have significant shortcomings, especially in dynamic path conflict detection and adjustment.

[0003] Traditional methods typically rely on pre-set, fixed paths, with robots executing tasks along these predetermined paths. This approach struggles with dynamic changes on the construction site, such as the presence of moving obstacles (such as moving construction materials or temporary scaffolding). This can lead to conflicts between robots. For example, if two robots attempt to pass through the same narrow passage at the same time, a collision or blockage can easily occur, severely impacting construction efficiency and safety.

[0004] Furthermore, existing technologies lag in detecting conflict paths. Typically, they only implement simple avoidance measures after a conflict occurs, failing to predict and avoid it in advance. This passive conflict handling approach not only delays construction schedules but also increases costs.

[0005] In summary, existing technologies for multi-robot collaborative construction operations are unable to effectively address path conflicts in dynamic environments and lack a real-time dynamic detection and adjustment mechanism. Therefore, a technical solution that can dynamically detect path conflicts in real time and perform intelligent adjustments is urgently needed to improve the efficiency and safety of construction. Summary of the Invention

[0006] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a multi-robot collaborative construction operation method. By decomposing the construction operation task into subtasks based on a three-dimensional model and assigning them to robots, the operation status and environmental data are collected in real time to build a dynamic conflict detection matrix, and the conflicting tasks are dynamically adjusted according to priority rules. The robots are controlled to execute the adjusted subtasks, so as to efficiently solve the path and resource conflict problems in construction and improve the efficiency and safety of construction operations.

[0007] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, the present invention provides a multi-robot collaborative construction operation method, comprising the following steps: Acquire a construction task and a three-dimensional building model, and decompose the construction task into a plurality of subtasks based on the three-dimensional building model; Generate an initial task allocation plan based on the operation type and spatial location of each subtask, and allocate the subtasks to multiple robots; Collect the operating status and environmental data of each robot in real time, and build a dynamic conflict detection matrix based on the operating status and environmental data; When a path conflict or a resource conflict is detected in the dynamic conflict detection matrix, dynamically adjusting the conflicting tasks based on a preset priority rule to generate an updated task allocation instruction; The task assignment instructions are sent to the corresponding robots to control each robot to execute the adjusted subtasks.

[0008] Beneficial effects: By decomposing construction work tasks into subtasks based on three-dimensional models and assigning them to robots, real-time collection of work status and environmental data is used to build a dynamic conflict detection matrix. Conflicting tasks are dynamically adjusted according to priority rules, and robots are controlled to execute the adjusted subtasks, so as to efficiently solve path and resource conflicts in construction and improve construction work efficiency and safety.

[0009] Furthermore, the three-dimensional building model includes structural component coordinates, work area boundaries, and obstacle distribution information; the decomposition basis of the subtasks includes the matching degree between the work type and the robot function, and the Euclidean distance between the spatial position and the current position of the robot.

[0010] Beneficial effects: The three-dimensional building model clearly contains the coordinates of structural components, the boundaries of the working area and the distribution of obstacles, making the basis for subtask decomposition more comprehensive. Taking into account the matching degree between the work type and the robot function, and the Euclidean distance between the spatial position and the robot's current position, the rationality and accuracy of task allocation are further optimized, thereby improving work efficiency.

[0011] Furthermore, the preset priority rules include: When there is a path conflict, the robot performing the emergency task has priority over the robot performing the routine task; When multiple robots apply for the same resource, the robot with a remaining battery power lower than a first preset threshold will be given priority in obtaining the resource; When the remaining time of the task is less than the second preset threshold, the corresponding robot prioritizes executing path planning.

[0012] Beneficial effects: Preset priority rules are established. When there is a conflict in routes, robots with urgent tasks are given priority. When multiple robots request the same resource, robots with low battery power are given priority. Robots with the shortest remaining time on their tasks are prioritized for path planning. These rules ensure the proper adjustment of robot tasks in conflicting situations, guaranteeing efficient and safe operations.

[0013] Furthermore, the method for generating the dynamic conflict detection matrix includes: Discretize the planned paths of the robots into time-space node sequences and calculate the node overlap; If the node overlap of two or more robots exceeds the threshold within the same time interval, it is determined to be a path conflict; If multiple robots apply for the same operating tool or material resource, it is determined to be a resource conflict.

[0014] Beneficial effects: The method for generating a dynamic conflict detection matrix is described in detail. Path conflicts are determined by discretizing the robot planning path into a time-space node sequence and calculating the node overlap. Resource conflicts are also determined by judging whether multiple robots apply for the same operating tool or material resource, thereby improving the accuracy and timeliness of conflict detection.

[0015] Furthermore, when a dynamic obstacle is detected, the environmental data is updated and the dynamic conflict detection matrix is recalculated. The dynamic adjustment includes: Predicting the coverage area of the dynamic obstacle within a preset time window based on the position and movement speed of the dynamic obstacle; Screening out robots whose paths intersect with the coverage area as affected robots; Planning a detour path for the affected robot and coordinating other robots to slow down or wait based on the priority rules, specifically including: generating at least two candidate detour paths based on obstacle distribution information in the three-dimensional building model; Calculate the path priority score based on the length, number of turns, and minimum distance to dynamic obstacles of each candidate detour path; Select the candidate detour path with the highest priority score as the final detour path; Sending avoidance instructions to other robots that have a path conflict with the affected robot; If the priority of the task performed by the other robot is lower than that of the affected robot, controlling the other robot to decelerate to a first speed threshold or enter a waiting state; If the priority of the task executed by the other robot is higher than that of the affected robot, the affected robot is controlled to suspend movement until the conflict is resolved.

[0016] Beneficial Effect: When a dynamic obstacle is detected, the system updates environmental data and recalculates the dynamic conflict detection matrix. Based on the location and speed of the dynamic obstacle, the system predicts the coverage area, identifies the affected robots, and plans a detour for them. It also coordinates other robots to slow down or wait. This dynamic adjustment mechanism effectively responds to dynamic environmental changes, further improving the system's adaptability and safety.

[0017] Furthermore, the dynamic adjustment is achieved through a reinforcement learning model, and the input of the reinforcement learning model includes: Conflict type, remaining battery power of the robot, task deadline, and output as task delay time or path replanning solution.

[0018] Furthermore, the types of path conflicts include: Intersection conflicts, conflicts involving vehicles traveling in opposite directions, and same-direction rear-end collisions. The types of resource conflicts include tool occupancy conflicts and material allocation conflicts. Clearly and subtly classifying conflict types helps the system more accurately identify and handle different types of conflicts, improving the pertinence and effectiveness of conflict resolution.

[0019] Furthermore, when the robot's communication is interrupted, it autonomously performs obstacle avoidance and task continuation operations based on the locally stored three-dimensional building model and historical task allocation instructions. This mechanism ensures that the robot can continue to perform tasks in the event of communication interruption, thereby improving the reliability and stability of the system.

[0020] In a second aspect, the present invention provides a multi-robot collaborative construction operation system for executing any of the multi-robot collaborative construction operation methods described in the first aspect, comprising: Task decomposition module, used to parse construction work tasks and generate subtask sets; Conflict detection module, used to build a dynamic conflict detection matrix and identify conflict types; Dynamic scheduling module, used to generate task allocation instructions based on priority rules; Communication control module, used to send instructions to each robot and receive status feedback; The modules work together to realize the complete process from task decomposition to conflict detection, dynamic adjustment and command sending, providing comprehensive technical support for multi-robot collaborative construction operations.

[0021] In the third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored on the memory. When the processor executes the computer program, it implements the method as described in any one of the first aspects. The electronic device provides a hardware implementation basis for the multi-robot collaborative construction operation method, enabling the method to be applied in actual construction scenarios, which has important practical significance.

[0022] In summary, compared with existing technologies, this invention achieves efficient collaborative operation of multiple robots in complex construction environments by constructing a dynamic conflict detection matrix and dynamically adjusting tasks based on priority rules. This method not only detects and resolves path and resource conflicts in real time, improving construction efficiency and safety, but also allows for autonomous execution of tasks even when communication is interrupted, significantly enhancing the system's reliability and adaptability, thus providing a more intelligent and efficient solution for the construction industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present invention and, together with the description, serve to explain the principles of the present invention.

[0024] The present invention can be more clearly understood from the following detailed description with reference to the accompanying drawings, in which: Figure 1 This is a general flow chart of the multi-robot collaborative construction operation method provided by an embodiment of the present invention; it shows the overall operation logic of the system; Figure 2 This is a task decomposition flow chart of the multi-robot collaborative construction operation method provided by an embodiment of the present invention, which specifically illustrates how to extract information from a three-dimensional model and decompose tasks; Figure 3 This is a diagram of a conflict detection mechanism for a multi-robot collaborative construction operation method provided by an embodiment of the present invention, illustrating a method for identifying path and resource conflicts; Figure 4 This is a dynamic task adjustment diagram of the multi-robot collaborative construction operation method provided by an embodiment of the present invention, showing the strategy for adjusting tasks according to different conflict types. DETAILED DESCRIPTION

[0025] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical solution of the present application, rather than limitations of the technical solution of the present application. In the absence of conflict, the technical features in the embodiments of the present application and the embodiments can be combined with each other. The following embodiments are only used to more clearly illustrate the technical solution of the present invention and are not intended to limit the scope of protection of the present invention.

[0026] The term "and / or" in this document simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can mean: A exists alone, A and B exist simultaneously, or B exists alone. Additionally, the character " / " in this document generally indicates an "or" relationship between the related objects.

[0027] Example 1 like Figure 1-Figure 2 As shown, an embodiment of the present invention provides a multi-robot collaborative construction operation method, comprising the following steps: S1. Obtain a construction task and a 3D building model, and decompose the construction task into multiple subtasks based on the 3D building model; S2. Generate an initial task allocation plan based on the job type and spatial location of each subtask, and assign the subtasks to multiple robots; S3. Collect the operating status and environmental data of each robot in real time, and build a dynamic conflict detection matrix based on the operating status and environmental data; S4. When a path conflict or resource conflict is detected in the dynamic conflict detection matrix, dynamically adjust the conflicting tasks based on preset priority rules and generate updated task allocation instructions; S5. Send the task assignment instructions to the corresponding robots, and control each robot to execute the adjusted subtasks.

[0028] Furthermore, the three-dimensional building model includes the coordinates of structural components, the boundaries of the work area, and the distribution of obstacles; the decomposition of subtasks is based on the matching degree between the work type and the robot function, and the Euclidean distance between the spatial position and the robot's current position.

[0029] As shown in the figure, the preset priority rules include: When there is a path conflict, the robot performing the emergency task has priority over the robot performing the routine task; When multiple robots apply for the same resource, the robot with a remaining battery power lower than a first preset threshold will be given priority in obtaining the resource; When the remaining time of the task is less than the second preset threshold, the corresponding robot prioritizes executing path planning.

[0030] Specifically, the method for generating the dynamic conflict detection matrix includes the following steps: S4.1. Path discretization processing: S4.1.1. Set the planned path of each robot at a preset time interval Perform discrete sampling to generate a time-space node sequence, where each node contains the robot ID, timestamp and three-dimensional coordinates ; S4.1.2. Based on the spatial resolution of the three-dimensional building model, map the three-dimensional coordinates to a grid space, where the grid side length is L; S4.2. Calculation of node overlap: S4.2.1. Construct a hash table to store the spatial node set corresponding to each timestamp, with the key being , the value is the grid coordinate of all robots at that moment; S4.2.2. For each timestamp , traverse the grid coordinate set. If there are more than two robot IDs in the same grid, it is determined to be spatial overlap; S4.2.3. Count the number of spatial overlaps within N consecutive time stamps and calculate the overlap ratio ; S4.3. Path conflict determination: S4.3.1, if ( is a preset overlap threshold), it is determined as a path conflict, and the conflict type is recorded as intersection conflict, opposite-direction driving conflict or same-direction rear-end collision conflict; S4.4. Resource conflict detection: S4.4.1. Monitor resource request instructions sent by each robot, where the resources include construction tools and material numbers; S4.4.2. If it is detected that the same resource is requested by two or more robots within the time window T, and the resource is currently in the "occupied" state, it is considered a resource conflict; S4.5. Conflict Matrix Construction: S.4.5.1. Generate a two-dimensional matrix , where the row index m represents the conflict type and the column index n represents the robot ID combination associated with the conflict; S4.5.2 Matrix Elements The value of is the conflict urgency, and the calculation formula is:

[0031] in is the weight coefficient, and .

[0032] Among them, the preset overlap threshold It can be a fixed empirical value or a dynamic value. For example, the types of robots in the construction operation area are periodically detected (based on the task division). When there are more types of robots, the road conditions are more complex, and the probability of unexpected robots arriving at the corresponding spatial nodes at the same time is higher. In this case, a preset overlap threshold is set accordingly. The smaller the value, the higher the sensitivity of path conflict determination. On the contrary, the preset overlap threshold is set accordingly. The bigger.

[0033] Furthermore, when a dynamic obstacle is detected, the environment data is updated and the dynamic conflict detection matrix is recalculated. The dynamic adjustment includes: Based on the position and movement speed of dynamic obstacles, predict the coverage area of dynamic obstacles within a preset time window; Filter out robots whose paths intersect with the coverage area as affected robots; Plan a detour path for the affected robot and coordinate other robots to slow down or wait based on priority rules, including: Generate at least two candidate detour paths based on obstacle distribution information in the three-dimensional building model; Calculate the path priority score based on the length, number of turns, and minimum distance to dynamic obstacles of each candidate detour path; Select the candidate detour path with the highest priority score as the final detour path; Send avoidance instructions to other robots that have a path conflict with the affected robot; If the task performed by other robots has a lower priority than the affected robot, the other robots are controlled to decelerate to the first speed threshold or enter a waiting state; If the task executed by other robots has a higher priority than the affected robot, the affected robot will be controlled to suspend movement until the conflict is resolved.

[0034] Furthermore, if Figure 4 As shown in Figure 2, dynamic adjustment is achieved through a reinforcement learning model, and the input of the reinforcement learning model includes: Conflict type, remaining battery power of the robot, task deadline, and output as task delay time or path replanning solution; The specific steps to achieve dynamic adjustment through reinforcement learning model include the following steps: S6.1. Construct the state space, action space, and reward function of the reinforcement learning model; wherein the state space includes: conflict type, remaining robot battery power, remaining task time, and distance between the current position and the target point; the action space includes: task delay adjustment, path replanning instructions, and resource reallocation instructions; and the reward function is calculated based on a weighted combination of task completion rate, conflict resolution efficiency, and energy consumption indicators. S6.2. Collect historical task execution data and conflict resolution records, and perform offline training on the reinforcement learning model until the model converges; S6.3. When a dynamic conflict is detected, the current state is input into the reinforcement learning model, and the optimal action is output; S6.4. Perform dynamic adjustment based on the optimal action, specifically including: If the output is a task delay adjustment, the allowed delay interval is calculated based on the robot's remaining battery power and the remaining time of the task, and the delay duration is allocated within the interval; If the output is a path replanning instruction, generating at least two alternative paths according to the obstacle distribution in the three-dimensional building model, and selecting the optimal path based on the path priority score; If the output is a resource reallocation instruction, resources are allocated preferentially to the high-priority task robots according to the priority rules; S6.5. Record the adjusted task execution results and update the training data set of the reinforcement learning model.

[0035] Furthermore, the types of path conflicts include: Intersection conflicts, conflicts of driving in opposite directions, and conflicts of rear-end collisions in the same direction. Types of resource conflicts include tool occupancy conflicts and material allocation conflicts. The detection and handling of path conflicts and resource conflicts include the following steps: S7.1. Path conflict detection: For intersection conflicts, the estimated time for each robot to arrive at the intersection is calculated. If the time difference is less than the first threshold, it is determined to be a conflict; For conflicts with the robots moving in opposite directions, the robot motion direction vectors are analyzed. If the path segments of the two robots intersect and move in opposite directions, a conflict is determined. For same-direction rear-end collisions, the relative speed and distance between the rear robot and the front robot are calculated. If the braking distance is less than the current distance, a collision is considered. S7.2 Resource Conflict Handling: When a tool occupancy conflict is detected, the usage status data of the tool is obtained. If the tool is in an idle timeout state, it is forcibly released and assigned to a high-priority task robot; When a material allocation conflict is detected, an alternative material collection path is planned for the conflicting robot based on the material storage point locations in the three-dimensional building model, and the dynamic conflict detection matrix is updated; S7.3. Conflict resolution verification: After the adjustment, the conflict detection matrix is recalculated. If the conflict node elimination rate does not reach the preset ratio, a manual intervention instruction is triggered.

[0036] Furthermore, if Figure 2-Figure 3 As shown in the figure, when the robot communication is interrupted, it autonomously performs obstacle avoidance and task continuation operations based on the locally stored building 3D model and historical task allocation instructions. The autonomous processing when the robot communication is interrupted includes the following steps: S8.1. Communication status monitoring: Periodically sending heartbeat packets to the central controller, if the number of consecutive heartbeat packets exceeds a third preset threshold, it is determined that the communication is interrupted; S8.2, Local Mode Activation: Loading the local area data related to the current task in the locally stored three-dimensional building model; Retrieve the unfinished subtask sequences and corresponding path plans in the historical task assignment instructions; S8.3. Autonomous obstacle avoidance execution: Activate the backup obstacle avoidance sensor set, including lidar and ultrasonic sensors, to update the obstacle map at 20Hz. Adopting the rolling horizon control method, the moving path within the next 5 seconds is dynamically planned based on the latest obstacle map; S8.4, Task Continuation Control: After communication is restored, the locally executed job data is compared with the central controller for differences; If the difference rate is lower than the fourth preset threshold, continue to execute subsequent subtasks; if the difference rate exceeds the limit, trigger the task to roll back to the most recent consistent state.

[0037] As an embodiment, a multi-robot collaborative construction operation system is used to perform a multi-robot collaborative construction operation method, including: Task decomposition module, used to parse construction work tasks and generate subtask sets; Conflict detection module, used to build a dynamic conflict detection matrix and identify conflict types; Dynamic scheduling module, used to generate task allocation instructions based on priority rules; The communication control module is used to send instructions to each robot and receive status feedback.

[0038] Example 2 An embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein when the computer program is executed by the processor, the steps of the following method are implemented: Obtaining a construction task and a three-dimensional building model, and decomposing the construction task into multiple subtasks based on the three-dimensional building model; Generate an initial task allocation plan based on the job type and spatial location of each subtask, and assign the subtasks to multiple robots; Collect the operating status and environmental data of each robot in real time, and build a dynamic conflict detection matrix based on the operating status and environmental data; When a path conflict or resource conflict is detected in the dynamic conflict detection matrix, the conflicting tasks are dynamically adjusted based on the preset priority rules to generate updated task allocation instructions; Send the task assignment instructions to the corresponding robots to control each robot to execute the adjusted subtasks.

[0039] The preset priority rules include: When there is a path conflict, the robot performing the emergency task has priority over the robot performing the routine task; When multiple robots apply for the same resource, the robot with a remaining battery power lower than a first preset threshold will be given priority in obtaining the resource; When the remaining time of the task is less than the second preset threshold, the corresponding robot prioritizes executing path planning.

[0040] When a dynamic obstacle is detected, the environment data is updated and the dynamic conflict detection matrix is recalculated. Dynamic adjustments include: Based on the position and movement speed of dynamic obstacles, predict the coverage area of dynamic obstacles within a preset time window; Filter out robots whose paths intersect with the coverage area as affected robots; Plan a detour path for the affected robot and coordinate other robots to slow down or wait based on priority rules, including: Generate at least two candidate detour paths based on obstacle distribution information in the three-dimensional building model; Calculate the path priority score based on the length, number of turns, and minimum distance to dynamic obstacles of each candidate detour path; Select the candidate detour path with the highest priority score as the final detour path; Send avoidance instructions to other robots that have a path conflict with the affected robot; If the task performed by other robots has a lower priority than the affected robot, the other robots are controlled to decelerate to the first speed threshold or enter a waiting state; If the task executed by other robots has a higher priority than the affected robot, the affected robot will be controlled to suspend movement until the conflict is resolved.

[0041] An electronic device provided by an embodiment of the present invention stores a computer program thereon, which can execute a multi-robot collaborative construction operation method provided by any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0042] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A multi-robot collaborative construction operation method, characterized in that: The following steps are involved: Acquire a construction task and a three-dimensional building model, and decompose the construction task into a plurality of subtasks based on the three-dimensional building model; Generate an initial task allocation plan based on the operation type and spatial location of each subtask, and allocate the subtasks to multiple robots; Collect the operating status and environmental data of each robot in real time, and build a dynamic conflict detection matrix based on the operating status and environmental data; When a path conflict or a resource conflict is detected in the dynamic conflict detection matrix, dynamically adjusting the conflicting tasks based on a preset priority rule to generate an updated task allocation instruction; The task assignment instructions are sent to the corresponding robots to control each robot to execute the adjusted subtasks.

2. The multi-robot collaborative construction operation method according to claim 1, characterized in that: The three-dimensional building model includes the coordinates of structural components, the boundaries of the working area, and obstacle distribution information; the decomposition basis of the subtasks includes the matching degree between the working type and the robot function, and the Euclidean distance between the spatial position and the current position of the robot.

3. The multi-robot collaborative construction operation method according to claim 1, characterized in that: The preset priority rules include: When there is a path conflict, the robot performing the emergency task has priority over the robot performing the routine task; When multiple robots apply for the same resource, the robot with a remaining battery power lower than a first preset threshold will be given priority in obtaining the resource; When the remaining time of the task is less than the second preset threshold, the corresponding robot prioritizes executing path planning.

4. The multi-robot collaborative construction operation method according to claim 1, characterized in that: The method for generating the dynamic conflict detection matrix includes: Discretize the planned paths of the robots into time-space node sequences and calculate the node overlap; If the node overlap of two or more robots exceeds the threshold within the same time interval, it is determined to be a path conflict; If multiple robots apply for the same operating tool or material resource, it is determined to be a resource conflict.

5. The multi-robot collaborative construction operation method according to claim 4, characterized in that: When a dynamic obstacle is detected, the environment data is updated and the dynamic conflict detection matrix is recalculated. Dynamic adjustment includes: Predicting the coverage area of the dynamic obstacle within a preset time window based on the position and movement speed of the dynamic obstacle; Screening out robots whose paths intersect with the coverage area as affected robots; Planning a detour path for the affected robot and coordinating other robots to slow down or wait based on the priority rules, specifically including: generating at least two candidate detour paths based on obstacle distribution information in the three-dimensional building model; Calculate the path priority score based on the length, number of turns, and minimum distance to dynamic obstacles of each candidate detour path; Select the candidate detour path with the highest priority score as the final detour path; Sending avoidance instructions to other robots that have a path conflict with the affected robot; If the priority of the task performed by the other robot is lower than that of the affected robot, controlling the other robot to decelerate to a first speed threshold or enter a waiting state; If the priority of the task executed by the other robot is higher than that of the affected robot, the affected robot is controlled to suspend movement until the conflict is resolved.

6. The multi-robot collaborative construction operation method according to claim 1, characterized in that: The dynamic adjustment is achieved through a reinforcement learning model, the input of which includes: Conflict type, remaining battery power of the robot, task deadline, and output as task delay time or path replanning solution.

7. The multi-robot collaborative construction operation method according to claim 4, characterized in that: The types of path conflicts include: Intersection conflicts, conflicts of driving in opposite directions and conflicts of rear-end collisions in the same direction. The types of resource conflicts include tool occupancy conflicts and material allocation conflicts.

8. The multi-robot collaborative construction operation method according to claim 1, characterized in that: When the robot communication is interrupted, it autonomously performs obstacle avoidance and task continuation operations based on the locally stored three-dimensional building model and historical task allocation instructions.

9. A multi-robot collaborative construction operation system, used to execute the multi-robot collaborative construction operation method according to any one of claims 1 to 8, characterized in that: include: Task decomposition module, used to parse construction work tasks and generate subtask sets; Conflict detection module, used to build a dynamic conflict detection matrix and identify conflict types; Dynamic scheduling module, used to generate task allocation instructions based on priority rules; The communication control module is used to send instructions to each robot and receive status feedback.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

Images