A cross-regional intelligent transfer system based on multi-node cooperation
The intelligent transfer system with multi-node collaboration solves the problems of cross-regional control command mismatch and collaboration failure by using continuous dual-mode verification and hierarchical conflict resolution methods, thereby improving the collaboration accuracy and stability in cross-regional dynamic environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- QINGDAO PORT INT CO LTD
- Filing Date
- 2026-01-29
- Publication Date
- 2026-06-19
Smart Images
Figure CN122243005A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of multi-agent collaborative scheduling technology, and in particular to a cross-regional intelligent transfer system based on multi-node collaboration. Background Technology
[0002] In recent years, with the rapid development of intelligent logistics, unmanned warehousing, and cross-regional collaborative scheduling, multi-node collaborative transfer control architecture has gradually become a path to improve logistics efficiency and resource utilization. Currently, centralized task allocation and static path planning are commonly used, combined with sensor networks and communication infrastructure, to achieve preliminary scheduling and execution of transfer tasks. This approach has already achieved application results in e-commerce logistics, intelligent manufacturing, and port scheduling scenarios.
[0003] However, significant limitations remain in dynamic task coupling, multi-source information fusion, and ensuring the consistency of real-time control commands. In cross-regional, multi-node, and high-concurrency task environments, initial control commands are susceptible to fluctuations in operating status, communication delays, and resource conflicts, making it difficult to achieve high-precision synchronization and closed-loop adjustment, resulting in a coordinated failure of overall robustness and response efficiency. Summary of the Invention
[0004] In view of the aforementioned existing problems, the present invention is proposed.
[0005] Therefore, this invention provides a cross-regional intelligent transfer system based on multi-node collaboration to solve the problems of control command mismatch and collaboration failure across regions.
[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a cross-regional intelligent transportation system based on multi-node collaboration, comprising: an information acquisition module, which acquires multi-source task demand inputs and, combined with the operating status of each execution node and task-resource coupling information, generates an initial parameter set for multi-node collaborative transportation control; a perception and monitoring module, which uses the operating status of each execution node and the initial parameter set as control inputs to form preliminary control commands; a verification and correction module, which uses a continuous dual-mode verification method to dynamically verify and correct the preliminary control commands to generate real-time control commands; an optimization and distribution module, which performs multi-source information fusion on the real-time control commands to suppress noise and deviations, and distributes them to each execution node through a communication link to obtain synchronous control signals; a coordination and feedback module, which, based on the real-time control commands and the operating status of each execution node, uses a hierarchical conflict resolution method to perform self-negotiation and arbitration scheduling to generate feedback signals; and a closed-loop update module, which performs closed-loop adjustment of the real-time control commands and the initial parameter set based on the feedback signals.
[0007] As a preferred embodiment of the cross-regional intelligent transfer system based on multi-node collaboration described in this invention, the multi-source task requirements refer to a set of task instructions from different task input sources that include transfer targets, timeliness requirements, and resource constraints.
[0008] As a preferred embodiment of the cross-regional intelligent transfer system based on multi-node collaboration described in this invention, the specific steps for generating the initial parameter set for multi-node collaborative transfer control are as follows: Based on multi-source task requirements input and combined with the real-time running status of each execution node, a quantitative assessment of the relationship between control objectives and node execution capabilities is conducted to generate resource allocation mapping data. Based on the resource allocation mapping data, a multi-node collaborative scheduling method is adopted to collaboratively plan the initial control path, execution timing and load allocation parameters to obtain the initial parameter set for multi-node collaborative control.
[0009] As a preferred embodiment of the cross-regional intelligent transfer system based on multi-node collaboration described in this invention, the specific steps for generating preliminary control commands are as follows: The real-time running status of each execution node is aligned and fused with the task execution intent of the initial parameter set to generate a collaborative control reference signal; The control action mapping is performed on the collaborative control reference signal to generate preliminary control commands.
[0010] As a preferred embodiment of the cross-regional intelligent transfer system based on multi-node collaboration described in this invention, the specific steps for generating real-time control commands are as follows: Based on the initial control commands and the operating status of each execution node, a continuous dual-mode verification method is used to dynamically verify the consistency of the initial control commands and obtain consistency verification information. Based on the consistency verification information, the timing deviations, resource allocation conflicts, and transfer path information in the preliminary control instructions are jointly corrected to generate real-time control instructions.
[0011] As a preferred embodiment of the cross-regional intelligent transfer system based on multi-node collaboration described in this invention, the communication link refers to the communication channel through which real-time control commands, after multi-source information fusion processing, are transmitted to each execution node.
[0012] As a preferred embodiment of the cross-regional intelligent transfer system based on multi-node collaboration described in this invention, the specific steps for obtaining the synchronization control signal are as follows: Based on the real-time control commands generated from multi-source task requirements input and the running status of each execution node, multi-source information fusion and joint suppression are performed on the noise and deviation components in the real-time control commands to obtain preliminarily purified control commands. Based on the initial purification control instructions, synchronous control signals are distributed and pushed to each execution node through the communication link.
[0013] As a preferred embodiment of the cross-regional intelligent transportation system based on multi-node collaboration described in this invention, the specific steps for generating the feedback signal are as follows: Based on real-time control commands and the running status of each execution node, a hierarchical conflict resolution method is adopted to self-negotiate resource occupation conflicts between adjacent execution nodes and obtain local resource allocation adjustment information. Based on local resource allocation adjustment information, a hierarchical conflict resolution method is used to arbitrate and schedule task timing and path conflicts between multiple nodes across regions, generating feedback signals.
[0014] As a preferred embodiment of the cross-regional intelligent transportation system based on multi-node collaboration described in this invention, the initial parameter set undergoes closed-loop adjustment, and the specific steps are as follows. Based on feedback signals, a dynamic deviation mapping method is used to quantify the deviation between real-time control commands and the actual operating status of each execution node, and to obtain control correction quantities. Based on the control correction amount and the current task requirements, the scheduling parameters of the initial parameter centralized transfer path, task timing and load distribution are corrected online to generate updated collaborative scheduling parameters.
[0015] As a preferred embodiment of the cross-regional intelligent transfer system based on multi-node collaboration described in this invention, the specific steps for performing closed-loop adjustment of real-time control commands and initial parameter sets based on feedback signals are as follows: Based on the updated collaborative scheduling parameters and the current running status of each execution node, the transfer path sequence, task execution sequence and load distribution parameters in the initial parameter set are jointly replanned to generate an updated initial parameter set. Based on the updated initial parameter set, the initial parameter set is adjusted in a closed loop according to the coupling relationship between the initial parameter set and the current task requirements to generate an updated initial parameter set. The real-time control command is then recalibrated using the updated initial parameter set to obtain the coordinated control command after closed-loop adjustment.
[0016] The beneficial effects of this invention are as follows: by adopting a continuous dual-mode verification method to dynamically verify and correct the initial control commands, the joint correction of timing deviations, resource conflicts and path information is achieved, ensuring the semantic consistency of multi-node control; by performing closed-loop adjustment of real-time control commands and initial parameter sets, online collaborative replanning of control parameters and task requirements is achieved, improving the collaborative accuracy and operational stability in cross-regional dynamic environments. Attached Figure Description
[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is a schematic diagram of a cross-regional intelligent transportation system based on multi-node collaboration.
[0019] Figure 2 A flowchart for multi-node collaborative scheduling generation and instruction mapping.
[0020] Figure 3 Flowchart for generating real-time control commands for continuous dual-mode verification Figure 4 Flowchart for hierarchical conflict resolution and closed-loop update Detailed Implementation To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0021] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.
[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.
[0023] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a cross-regional intelligent transfer system based on multi-node collaboration, comprising the following modules: The information acquisition module acquires multi-source task requirements inputs and, combined with the running status of each execution node and the coupling degree between tasks and resources, generates an initial parameter set for multi-node collaborative transfer control.
[0024] Multi-source task requirements refer to a set of task instructions from different task input sources, which include information on transfer targets, timeliness requirements, and resource constraints.
[0025] Based on multi-source task requirements input and combined with the real-time operating status of each execution node, a quantitative assessment of the relationship between control objectives and node execution capabilities is performed to generate resource allocation mapping data.
[0026] Furthermore, the transfer targets, timeliness requirements, and resource constraints included in the multi-source task requirements are matched and analyzed with the real-time operating status of each execution node. Based on the multi-node collaborative scheduling method, the execution capabilities of the control target are compared with the current execution capabilities of each execution node. By quantitatively evaluating the adaptability of each execution node in three dimensions—transfer path coverage capability, task timing capability, and load allocation parameter carrying capacity—a matching degree value between each execution node and the control target is formed. Based on the matching degree value, the resource constraints in the multi-source task requirements are decomposed and mapped to generate resource allocation mapping data.
[0027] The formula for calculating the matching degree is: ; in, It is the matching degree value between each execution node and the control target. It refers to the coverage capability of each execution node along the transit path, for the nodes. The range of currently executable paths, in meters (m). It is the path length required for the transit target included in the multi-source mission requirements, in meters (m). It is a node The task timing capability is the currently schedulable time window (such as the length of an idle period), in seconds (s). It is a task The time requirement is the upper limit of the time for task completion, in seconds (s). It is a node The load distribution parameter carrying capacity is the amount of load that can be carried at present, in kilograms (kg). It is a task The load requirement in the resource constraint information is expressed in kilograms (kg). It is the index of the execution node. It is an index of the task control target.
[0028] It should be noted that each item is a ratio of the same physical quantity (m / m, s / s, kg / kg), which is dimensionless, so the formula for calculating the matching degree has a unified dimension.
[0029] The matching degree value refers to the numerical information obtained by quantifying the degree of fit between the control target indicators and the execution capabilities of the nodes when comparing the control targets in the multi-source task requirements with the real-time running status of each execution node.
[0030] Node execution capability refers to the ability of each execution node, based on its current operating state, to execute the transfer path, task timing, and load allocation parameters required by the control objective under the multi-node collaborative scheduling method. Node execution capability is the actual executability of the quantitative evaluation information in the resource allocation mapping data.
[0031] A control objective is a set of control objectives corresponding to a single transfer task, consisting of transfer path, task timing, and load allocation parameters. A set of control objectives is a set of all control objectives in multi-source task requirements.
[0032] The multi-node collaborative scheduling method aims to generate an initial parameter set that satisfies global collaborative constraints. Under the joint constraints of multi-source task requirements input and the real-time running status of each execution node, it collaboratively plans the initial control path, execution sequence, and load distribution parameters.
[0033] During the planning process, the allocated transfer paths do not exceed the current transfer path coverage capacity of each execution node, the task execution sequence meets the timeliness requirements of the tasks and does not misalign with the current task arrangement of the nodes, the load allocation does not exceed the current load allocation parameter carrying capacity of each execution node, and during scheduling, tasks are preferentially allocated to nodes with higher matching degrees based on the matching degree values between each execution node and the control target. When multiple tasks have resource allocation conflicts with competing relationships, the task execution sequence is adjusted in combination with the timeliness requirements of the tasks and resource constraint information, and the occupation relationship of shared resources is redistributed to ensure that each execution node maintains consistency and coordination in terms of path, sequence and load dimensions.
[0034] By quantitatively assessing the matching degree between control objectives and node execution capabilities, resource allocation mapping data is formed. Furthermore, the task coupling, resource consumption, and timing dependencies among multiple execution nodes across regions are comprehensively considered to generate an initial parameter set that satisfies global collaborative constraints.
[0035] A superior method is the multi-node collaborative scheduling method, which runs through the initial parameter set generation, real-time control command correction, feedback signal arbitration and closed-loop update process. During task execution, each execution node maintains consistency and coordination in terms of path, timing and load dimensions, supporting the synchronous control and dynamic optimization of multiple nodes in cross-regional intelligent transfer.
[0036] Based on the resource allocation mapping data, a multi-node collaborative scheduling method is adopted to collaboratively plan the initial control path, execution timing and load allocation parameters to obtain the initial parameter set for multi-node collaborative control.
[0037] Furthermore, based on the quantitative matching relationship between the execution capabilities of each execution node and the control objectives reflected in the resource allocation mapping data, the multi-node collaborative scheduling method is used to plan the initial control path for the spatial path sequence of cross-regional transfer tasks, and the execution timing of the task execution start and end times of each execution node is arranged in sequence. The task load is allocated according to the load carrying capacity of each node in the resource allocation mapping data. Under the unified scheduling of the multi-node collaborative scheduling method, collaborative planning is carried out, and the initial parameter set of multi-node collaborative control, including the initial control path, execution timing and load allocation parameters, is output.
[0038] Specifically, quantitative matching relationship refers to the correspondence between the control target and the execution capability of the nodes by quantitatively evaluating the input of multi-source task requirements and the real-time running status of each execution node, and obtaining the corresponding data of resource allocation mapping.
[0039] Initial control path planning refers to the process of assigning an initial spatial trajectory with time and load constraints to each task in the feasible path set using a multi-node collaborative scheduling method, based on the matching information between each execution node and task in the resource allocation mapping data, combined with the starting and ending points of the transfer target, the environmental topology, and the node path coverage capability, and using this trajectory as the benchmark path sequence for collaborative control.
[0040] Load parameter allocation refers to allocating the load required by a task to one or more execution nodes based on the matching degree of each execution node to the task and its current load capacity in the resource allocation mapping data. If the task cannot be split, the node with the highest load capacity that meets the task requirements and the matching degree is selected to bear the entire load. If the task can be split, the load is distributed according to the proportion of the current remaining load of each matching node. The total load after allocation to any node does not exceed its physical limit. At the same time, the allocation information is coordinated with the planned path and timing to form a load allocation scheme that meets resource constraints, capacity matching and is executable.
[0041] The physical upper limit refers to the maximum load that the execution node can safely bear under the current operating state, which is determined by the structural strength, power performance, safety margin and operating conditions of the equipment.
[0042] Collaborative planning refers to optimizing the initial control path, execution sequence, and load allocation parameters in multi-node collaborative scheduling, and coordinating the task arrangements of each execution node in terms of space, time, and resources.
[0043] The perception and monitoring module takes the running status and initial parameter set of each execution node as control input to form preliminary control commands.
[0044] The real-time running status of each execution node is aligned and fused with the task execution intent of the initial parameter set to generate a collaborative control reference signal.
[0045] Furthermore, the real-time running status (location, speed, load status, and task execution progress information) of each execution node is extracted, and the task execution intent in the initial parameter set is obtained. The task execution intent is the initial control path, execution sequence, and load allocation parameters planned by the multi-node collaborative scheduling method. The status is aligned and fused, and the real-time running status of each execution node is aligned with the task execution intent in the initial parameter set in both time and space dimensions. The current state of each execution node is kept consistent with the corresponding task execution intent in the initial parameter set, and a collaborative control benchmark signal is generated to show the matching status between the task execution intent and the current execution capability.
[0046] Specifically, the task execution intent refers to the set of control objectives planned by the multi-node collaborative scheduling method, which includes the initial control path, execution sequence and load allocation parameters. The set of control objectives originates from the quantitative evaluation information of multi-source task demand input and the real-time running status of each execution node, and guides the behavioral intent of each execution node to complete the transfer task.
[0047] The task execution intent refers to the path tracing, timing compliance, and load behavior objectives that each execution node needs to complete, as expressed by the initial parameter set, including the initial control path, execution timing, and load allocation parameters.
[0048] Alignment and fusion refers to aligning the real-time running status of each execution node with the task execution intent in the initial parameter set in both time and space dimensions, ensuring that the current state of each execution node is consistent with the control objective in the task execution intent, and generating a collaborative control reference signal.
[0049] The control action mapping is performed on the collaborative control reference signal to generate preliminary control commands.
[0050] Furthermore, the multi-node task execution intention contained in the collaborative control reference signal is matched with the real-time operating status of each execution node. Based on the control path, execution timing and load allocation parameters of the multi-node collaborative scheduling method, the control objective in the collaborative control reference signal is decomposed into a sequence of action instructions that each execution node can execute. Based on the node execution capability quantitatively evaluated in the resource allocation mapping data, the action instruction sequence is mapped to control actions to generate preliminary control instructions that match the capabilities of each execution node.
[0051] Specifically, control action mapping refers to the process of decomposing the control objective in the collaborative control reference signal into a sequence of action instructions that can be executed by each execution node, based on the control path, execution timing and load allocation parameters of the multi-node collaborative scheduling method.
[0052] Decomposing the control objective in the collaborative control baseline signal into a sequence of executable action instructions for each execution node means, based on the quantitative evaluation of each execution node's execution capabilities in three dimensions—transfer path coverage capability, task timing capability, and load allocation parameter carrying capacity—in the resource allocation mapping data, breaking down the initial control path, execution timing, and load allocation parameters in the global collaborative intent into control commands specific to each node according to the matching relationship between tasks and nodes. These commands are ordered in time (e.g., start and stop times), clear in path (e.g., waypoint sequence), and feasible in resources (e.g., load does not exceed the carrying capacity limit). This generates a sequence of executable instructions that drives each execution node to complete movement, start and stop, loading and unloading, and handover actions.
[0053] An executable sequence of action instructions refers to the control objectives abstracted from the collaborative control reference signal, which are transformed into control commands that are ordered in time, feasible in resources, and clear in path, according to the current operating status of each execution node and the node execution capability of the resource allocation mapping data. These commands drive the execution nodes to complete the movement, start-stop, loading and unloading, and handover actions in the transfer task.
[0054] The verification and correction module uses a continuous dual-mode verification method to dynamically verify and correct the consistency of the initial control commands, and generate real-time control commands.
[0055] It should be noted that the continuous dual-mode verification method refers to simultaneously performing logical consistency verification and spatial consistency verification during the generation and execution of control commands. Logical consistency verification verifies that the task timing arrangement and resource allocation conform to the node idle state and task constraints, while spatial consistency verification verifies that the transfer path is physically reachable under the node's mobility and environmental constraints. Joint comparison is performed at the verification information level. If a conflict exists, a correction is triggered. By iteratively executing dual-channel verification and joint judgment in the time dimension, the control commands can simultaneously meet the requirements of cross-regional multi-node collaborative transfer in terms of logical rationality and spatial executability.
[0056] Logical consistency verification generates logical consistency verification information by aligning the task execution intent in the initial control command with the running status of each execution node in terms of timing and comparing the resource occupancy status. Spatial consistency verification generates spatial consistency verification information by verifying the feasibility of the transfer path information in the initial control command with the current spatial position and movement status of each execution node. The continuous dual-mode verification method continuously iterates and performs verification in the time dimension, and jointly compares the logical consistency verification information obtained from each verification with the spatial consistency verification information to generate consistency verification information for the corrected control command.
[0057] Based on the initial control commands and the operating status of each execution node, a continuous dual-mode verification method is used to dynamically verify the consistency of the initial control commands and obtain consistency verification information.
[0058] Furthermore, the task execution intent in the preliminary control command is time-aligned with the running status of each execution node and compared with the resource occupancy status to generate logical consistency verification information. The transfer path information in the preliminary control command is used to perform feasibility verification with the current spatial position and movement status of each execution node to generate spatial consistency verification information. The logical consistency verification information and the spatial consistency verification information are jointly compared. Through the dual-channel consistency determination of the continuous dual-mode verification method, the logical and spatial dual consistency verification information of the preliminary control command is output as consistency verification information.
[0059] Specifically, logical consistency verification information and spatial consistency verification information are jointly compared. The process is as follows: within the same task time window, logical consistency verification information (including conflict between task timing and node idle time periods, and violation of exclusive and shared resource allocation constraints) and spatial consistency verification information (including instruction path exceeding the current movement capability of the node, path point reachability, and spatial conflict between trajectory and environmental obstacles and node paths) are checked simultaneously. When a task is logically allocated an effective time slot and resources, but is spatially infeasible (such as insufficient turning radius and excessive distance), or the path is reachable but the execution time overlaps with an existing task, it is determined to be jointly inconsistent, and verification feedback containing timing offset, resource release instructions, and path fine-tuning suggestions is generated.
[0060] The dual-channel consistency determination of the continuous dual-mode verification method is as follows: the first channel (logical consistency verification) verifies the task execution intention in the preliminary control instruction, and checks whether the execution sequence of the task is compatible with the current task arrangement and idle state of each execution node, and whether the resource allocation violates the resource constraints in the multi-source task requirements (such as multiple tasks occupying the same exclusive resource at the same time).
[0061] The second channel (spatial consistency verification) verifies that the transfer path in the instruction is executable in physical space, reachable under the current position, movement capability and environmental constraints of each execution node, and runs synchronously and iterates continuously in the time dimension in cases where the path crosses obstacles, exceeds the path coverage capability and the node trajectory interferes with space. It also compares the verification information generated by each node together. When both logic and space are consistent, the initial control instruction is confirmed to be valid; otherwise, a joint correction is triggered.
[0062] Based on the consistency verification information, the timing deviations, resource allocation conflicts, and transfer path information in the preliminary control instructions are jointly corrected to generate real-time control instructions.
[0063] Furthermore, based on the consistency verification information output by the continuous dual-mode verification method, timing deviation items that are inconsistent with the running status of each execution node and resource allocation conflicts that compete for task requirements are identified in the preliminary control instructions. According to the deviation direction and conflict type indicated by the consistency verification information, the task execution timing sequence in the preliminary control instructions is adjusted synchronously, competing resource occupancy relationships are reallocated, and the transfer path information is locally replanned. The adjusted parameters are kept consistent with the consistency verification information to form real-time control instructions.
[0064] Specifically, the timing deviation refers to the inconsistency in the time dimension between the task execution timing contained in the preliminary control instructions and the current running status of each execution node. This manifests as a misalignment between the execution time and timing sequence specified by the instructions and the actual executable capability of the nodes and the current task arrangement.
[0065] Conflict types refer to timing discrepancy conflicts (misalignment between task execution time and actual node state), resource allocation conflicts (multiple tasks competing for the same and mutually exclusive resources), and path space conflicts (infeasible transit paths and overlapping node paths).
[0066] Resource allocation conflicts arising from competing task requirements refer to situations where the same and mutually exclusive resources allocated to multiple tasks in the initial control instructions overlap and compete with the resource constraint information input from multiple source task requirements, making it impossible to simultaneously satisfy the exclusive and shared resource requirements of all tasks.
[0067] The distribution module is optimized to perform multi-source information fusion on real-time control commands to suppress noise and deviation, and distributes them to each execution node through the communication link to obtain synchronous control signals.
[0068] A communication link refers to the communication channel through which real-time control commands, after being processed by multi-source information fusion, are transmitted to each execution node.
[0069] Based on the real-time control commands generated from multi-source task requirements and the running status of each execution node, multi-source information fusion and joint suppression are performed on the noise and deviation components in the real-time control commands to obtain preliminarily purified control commands.
[0070] Furthermore, based on the multi-source task requirement input and the real-time control commands generated from the running status of each execution node, the noise and deviation components introduced by communication interference, sensing errors and execution delays in the real-time control commands are jointly suppressed through multi-source information fusion, retaining control information that is logically consistent with and state-aligned with the collaborative control reference signal, thus obtaining a preliminarily purified control command.
[0071] It should be noted that the process of multi-source information fusion is to fuse the data from the information sources, including the task path coordinates, execution time and load allocation specified in the real-time control command, the current position, actual action timestamp and current load status fed back by each execution node through its own sensors, and the command reception confirmation and task execution progress feedback returned by the communication link.
[0072] Multi-source information fusion refers to cross-comparing the control intent in real-time control commands with the control reference signals of collaborative control. By consistently mapping the transfer path, task timing and resource allocation parameters in the real-time control commands with the current operating status of each execution node, it identifies and retains mutually supporting control elements from multiple information sources to form a unified control representation.
[0073] Joint suppression refers to the process of simultaneously eliminating inconsistent control information in the real-time control command that conflicts with the cooperative control reference signal in terms of transmission path, task timing, and resource allocation during the multi-source information fusion process, by combining the timing deviation, resource allocation conflict, and path deviation of the consistency verification information.
[0074] Based on the initial purification control instructions, synchronous control signals are distributed and pushed to each execution node through the communication link.
[0075] Furthermore, the control commands for preliminary purification are synchronously distributed and pushed according to the transmission protocol supported by the communication link. A unified timestamp is added to the distributed and pushed commands so that each execution node can align its execution time according to the timestamp when it receives the command. The control commands for preliminary purification carrying the unified timestamp are pushed to all execution nodes in parallel through the communication link. After receiving the commands, each execution node aligns its start time in the local scheduler according to the timestamp, so that all execution nodes can execute the corresponding actions under the same control timing and obtain a synchronous control signal.
[0076] The coordination and feedback module, based on real-time control commands and the operating status of each execution node, adopts a hierarchical conflict resolution method to perform self-negotiation and arbitration scheduling, and generates feedback signals.
[0077] Based on real-time control commands and the running status of each execution node, a hierarchical conflict resolution method is adopted to self-negotiate resource occupation conflicts between adjacent execution nodes and obtain local resource allocation adjustment information.
[0078] Furthermore, based on the resource allocation information contained in the real-time control instructions and the current resource occupancy status of each execution node in its running status, the overlapping and competing resource occupancy of adjacent execution nodes in terms of transfer path, task timing, and load allocation parameters are identified. A hierarchical conflict resolution method is applied at the local level. Based on task priority, timeliness requirements, and resource constraint information, conflicting resource requests are negotiated and sorted. By adjusting the task execution timing of each adjacent execution node, the right to use shared resources is reallocated, and local resource allocation adjustment information is output.
[0079] It should be noted that the hierarchical conflict resolution method refers to a graded processing method used to handle resource and task conflicts during multi-node collaborative transfer. At the local level, resource occupation conflicts between adjacent execution nodes are self-negotiated. By comparing the real-time operating status of each adjacent execution node with the resource allocation intention in the real-time control instructions, local conflicts in transfer paths, task timing, and load allocation are identified. Based on task timeliness requirements and resource constraint information, the resource occupation plans of adjacent execution nodes are dynamically adjusted to generate local resource allocation adjustment information. At the global level, task timing and path conflicts between multiple nodes across regions are arbitrated and scheduled. By integrating the resource allocation adjustment information of each locality and the overall task requirements, the execution rhythm and path planning between different regions are coordinated to reduce task blocking and resource deadlock on a global scale and generate a closed-loop adjustment with feedback signals.
[0080] Negotiating and prioritizing conflicting resource requests means that when multiple adjacent execution nodes simultaneously request the use of the same mutually exclusive resource (such as the same transfer channel, the same loading and unloading station, and the same communication bandwidth during the same period), the coordination feedback at the local level prioritizes the conflicting resource requests according to their execution urgency based on the timeliness requirements of each task (such as the urgency of the deadline) and resource constraint information (such as task priority and load type).
[0081] For example, if task A requires completion within 10 seconds while task B allows completion within 30 seconds, resources will be allocated to task A first. At the same time, based on the current task schedule of the node, the execution sequence of task B will be adjusted to postpone the use of resources, generating local resource allocation adjustment information.
[0082] A local collaborative region refers to a logical region composed of adjacent execution nodes that handles resource conflicts.
[0083] Based on local resource allocation adjustment information, a hierarchical conflict resolution method is used to arbitrate and schedule task timing and path conflicts between multiple nodes across regions, generating feedback signals.
[0084] Furthermore, at the global level of the hierarchical conflict resolution method, local resource allocation adjustment information is used as input. Combined with the transfer targets, timeliness requirements, and resource constraints in the multi-source task requirements, consistency analysis is performed on the task timing and transfer path planning between multiple nodes across regions. By comparing the real-time running status of each execution node in different regions with the task execution intent of real-time control commands, cross-regional task timing misalignment and path intersection conflicts caused by local adjustments are identified. Based on the arbitration scheduling of the hierarchical conflict resolution method, the execution rhythm between different regions is coordinated at the global level, the transfer paths of conflicts are replanned, and the timing of conflicting tasks is staggered and rearranged. Under the premise of task requirements, each execution node reduces task blocking and resource deadlock, and outputs a global coordination feedback signal.
[0085] The closed-loop update module performs closed-loop adjustment of real-time control commands and initial parameter sets based on feedback signals.
[0086] Based on feedback signals, a dynamic deviation mapping method is used to quantify the deviation between real-time control commands and the actual operating status of each execution node, thereby obtaining control correction values.
[0087] Furthermore, the transfer path, task timing, and load allocation parameters in the real-time control command are compared item by item with the corresponding transfer path execution position, task execution time node, and load allocation parameter occupancy in the actual running state of each execution node. The dynamic deviation mapping method is used to quantify the deviation of each dimension parameter in the timing, space, and resource dimensions, and the deviation is mapped into the correction amount for the transfer path, task timing, and load allocation to generate control correction amount.
[0088] The formula for calculating the control correction is: ; in, It is a task The global control correction amount. It is the first A local collaborative region for the task The resulting local deviation correction amount It is the total number of local regions participating in collaborative feedback (e.g., the number of regions, the number of adjacent node pairs), a dimensionless positive integer. It is a local area index.
[0089] It should be noted that all local deviation terms in the control correction calculation formula... With global correction They all describe the same physical dimension (path, timing, and load) and use the same units, so their dimensions are consistent.
[0090] It should be noted that the dynamic deviation mapping method refers to the method of establishing a dynamic correlation mapping relationship between control command parameters and node operating status parameters based on the difference information between the actual operating status of each execution node and the real-time control command contained in the feedback signal. This method quantifies the deviations in three dimensions—transfer path, task timing, and load allocation—in real time. Based on the quantitative evaluation between the control objective and the execution capability of the node in the multi-node collaborative scheduling method, the deviations in each dimension are converted into corresponding executable correction magnitudes, thereby generating control correction amounts for adjusting control commands.
[0091] Quantitative analysis refers to comparing the task parameters of real-time control commands (such as target position, execution time, and load) with the actual operating status of each execution node (such as current position, actual action time, and current load) in three dimensions: path, timing, and load. This yields the deviation in physical units (meters, seconds, and kilograms) and direction, and the deviation is directly converted into an executable correction amount. This provides an adjustment basis for the closed-loop update module, enabling dynamic alignment and online collaborative optimization between control commands and actual execution status.
[0092] Based on the control correction amount and the current task requirements, the scheduling parameters of the initial parameter centralized transfer path, task timing and load distribution are corrected online to generate updated collaborative scheduling parameters.
[0093] Furthermore, based on the deviation between the real-time control command of the control correction quantity and the actual operating status of each execution node, and combined with the transfer target, timeliness requirements and resource constraint information contained in the current task requirements, the transfer path contained in the initial parameter set is corrected online, the task timing is compensated for timing offset, and the load distribution is rebalanced to obtain the corrected transfer path, task timing and load distribution that meet the current operating status constraints of each execution node, and the updated collaborative scheduling parameters are generated.
[0094] Based on the updated collaborative scheduling parameters and the current running status of each execution node, the transfer path sequence, task execution timing, and load allocation parameters in the initial parameter set are jointly replanned to generate an updated initial parameter set.
[0095] Furthermore, by utilizing a multi-node collaborative scheduling method, the updated collaborative scheduling parameters are matched with the current running status of each execution node. The transfer path sequence in the initial parameter set is replanned, the resource usage of the task execution sequence is replanned, and the load allocation parameters are replanned based on the node processing capacity and the current load status. Through the multi-node collaborative scheduling method, the node jump order of the transfer path sequence, the triggering time arrangement of the task execution sequence, and the distribution ratio of the load allocation parameters among the execution nodes are adjusted synchronously to generate an updated initial parameter set.
[0096] It should be noted that joint replanning refers to using a multi-node collaborative scheduling method to synchronously adjust the transit path sequence, task execution timing, and load allocation parameters in the initial parameter set, thereby generating an updated initial parameter set.
[0097] Based on the updated initial parameter set, the initial parameter set is adjusted in a closed loop according to the coupling relationship between the initial parameter set and the current task requirements to generate an updated initial parameter set. The real-time control command is then recalibrated using the updated initial parameter set to obtain the coordinated control command after closed-loop adjustment.
[0098] Furthermore, based on the control correction amount and combined with the transfer target, timeliness requirements, and resource constraints included in the current task requirements, the transfer path, task timing, and load allocation parameters included in the initial parameter set are corrected online to generate updated collaborative scheduling parameters. Based on the updated collaborative scheduling parameters and the current operating status of each execution node, the transfer path sequence, task execution timing, and load allocation parameters in the initial parameter set are recalibrated to form an updated initial parameter set, thus obtaining the collaborative control command after closed-loop adjustment.
[0099] It should be noted that the closed-loop update module is configured to quantify the control deviation based on the feedback signal using a dynamic deviation mapping method, and to jointly re-plan the initial parameter set and real-time control commands to achieve online collaborative optimization.
[0100] In summary, this invention achieves dynamic consistency verification and correction of initial control commands by employing a continuous dual-mode verification method, thereby jointly correcting timing deviations, resource conflicts, and path information to ensure semantic consistency of multi-node control. Furthermore, by performing closed-loop adjustment of real-time control commands and initial parameter sets, it enables online collaborative replanning of control parameters and task requirements, thus improving the collaborative accuracy and operational stability in cross-regional dynamic environments.
[0101] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A cross-regional intelligent transfer system based on multi-node collaboration, characterized in that: include, The information acquisition module acquires multi-source task requirements inputs, and combines the running status of each execution node and the coupling degree between tasks and resources to generate an initial parameter set for multi-node collaborative transfer control. The perception and monitoring module takes the running status and initial parameter set of each execution node as control input to form preliminary control commands; The verification and correction module uses a continuous dual-mode verification method to dynamically verify and correct the consistency of the initial control commands and generate real-time control commands. The distribution module is optimized to perform multi-source information fusion on real-time control commands to suppress noise and deviation, and distributes them to each execution node through the communication link to obtain synchronous control signals; The coordination and feedback module, based on real-time control commands and the running status of each execution node, adopts a hierarchical conflict resolution method to perform self-negotiation and arbitration scheduling, and generates feedback signals. The closed-loop update module performs closed-loop adjustment of real-time control commands and initial parameter sets based on feedback signals.
2. The cross-regional intelligent transfer system based on multi-node collaboration as described in claim 1, characterized in that: The multi-source task requirements refer to a set of task instructions from different task input sources, which include information on transfer targets, timeliness requirements, and resource constraints.
3. The cross-regional intelligent transfer system based on multi-node collaboration as described in claim 2, characterized in that: The specific steps for generating the initial parameter set for multi-node coordinated transport control are as follows: Based on multi-source task requirements input and combined with the real-time running status of each execution node, a quantitative assessment of the relationship between control objectives and node execution capabilities is conducted to generate resource allocation mapping data. Based on the resource allocation mapping data, a multi-node collaborative scheduling method is adopted to collaboratively plan the initial control path, execution timing and load allocation parameters to obtain the initial parameter set for multi-node collaborative control.
4. The cross-regional intelligent transfer system based on multi-node collaboration as described in claim 3, characterized in that: The specific steps for generating the initial control command are as follows: The real-time running status of each execution node is aligned and fused with the task execution intent of the initial parameter set to generate a collaborative control reference signal; The control action mapping is performed on the collaborative control reference signal to generate preliminary control commands.
5. The cross-regional intelligent transfer system based on multi-node collaboration as described in claim 4, characterized in that: The specific steps for generating real-time control commands are as follows: Based on the initial control commands and the operating status of each execution node, a continuous dual-mode verification method is used to dynamically verify the consistency of the initial control commands and obtain consistency verification information. Based on the consistency verification information, the timing deviations, resource allocation conflicts, and transfer path information in the preliminary control instructions are jointly corrected to generate real-time control instructions.
6. The cross-regional intelligent transfer system based on multi-node collaboration as described in claim 5, characterized in that: The communication link refers to the communication channel through which real-time control commands, after multi-source information fusion processing, are transmitted to each execution node.
7. The cross-regional intelligent transfer system based on multi-node collaboration as described in claim 6, characterized in that: The specific steps for obtaining the synchronization control signal are as follows: Based on the real-time control commands generated from multi-source task requirements input and the running status of each execution node, multi-source information fusion and joint suppression are performed on the noise and deviation components in the real-time control commands to obtain preliminarily purified control commands. Based on the initial purification control instructions, synchronous control signals are distributed and pushed to each execution node through the communication link.
8. The cross-regional intelligent transfer system based on multi-node collaboration as described in claim 7, characterized in that: The specific steps for generating the feedback signal are as follows: Based on real-time control commands and the running status of each execution node, a hierarchical conflict resolution method is adopted to self-negotiate resource occupation conflicts between adjacent execution nodes and obtain local resource allocation adjustment information. Based on local resource allocation adjustment information, a hierarchical conflict resolution method is used to arbitrate and schedule task timing and path conflicts between multiple nodes across regions, generating feedback signals.
9. The cross-regional intelligent transfer system based on multi-node collaboration as described in claim 8, characterized in that: The initial parameter set is then subjected to closed-loop adjustment, and the specific steps are as follows. Based on feedback signals, a dynamic deviation mapping method is used to quantify the deviation between real-time control commands and the actual operating status of each execution node, and to obtain control correction quantities. Based on the control correction amount and the current task requirements, the scheduling parameters of the initial parameter centralized transfer path, task timing and load distribution are corrected online to generate updated collaborative scheduling parameters.
10. The cross-regional intelligent transfer system based on multi-node collaboration as described in claim 9, characterized in that: The closed-loop adjustment of real-time control commands and initial parameter sets based on feedback signals involves the following steps. Based on the updated collaborative scheduling parameters and the current running status of each execution node, the transfer path sequence, task execution sequence and load distribution parameters in the initial parameter set are jointly replanned to generate an updated initial parameter set. Based on the updated initial parameter set, the initial parameter set is adjusted in a closed loop according to the coupling relationship between the initial parameter set and the current task requirements to generate an updated initial parameter set. The real-time control command is then recalibrated using the updated initial parameter set to obtain the coordinated control command after closed-loop adjustment.