A Distributed Robot Cooperative Control Method and System Based on Edge Computing

Abstract

This application provides a distributed robot cooperative control method and system based on edge computing. The method involves collecting environmental spatial data and dynamic obstacle motion parameters from the robot, and performing cross-robot joint compression to obtain a distributed perception dataset. Subsequently, based on the spatial density differences of this dataset, hierarchical communication links and their signal transmission path priorities within the group are dynamically constructed and adjusted in real time. Next, the adjusted priorities are converted into communication weight parameters, combined with obstacle motion parameters to generate a cooperative control instruction set, and synchronized across nodes with the obstacle information perceived by each node. Finally, the synchronized instruction set is used to drive the robot to perform edge computing resource interactions and generate a cooperative motion path. The technical solution provided in this application can improve the communication efficiency and adaptability of distributed robot cooperative control.

Description

Technical Field

[0001] This application relates to the field of distributed robot cooperative control technology, and in particular to a distributed robot cooperative control method and system based on edge computing. Background Technology

[0002] In scenarios such as industrial automation, intelligent logistics, and disaster relief, multi-robot systems need to collaboratively complete tasks in dynamic environments while avoiding moving obstacles. These scenarios demand high real-time performance, strong robustness, and low communication dependency from the systems. Robots must respond quickly to changes in obstacle movement through local perception and distributed decision-making, and achieve path planning and conflict avoidance without centralized control. Furthermore, the systems must adapt to complex and ever-changing environmental uncertainties, such as unknown obstacle movement patterns and sensor noise interference, which poses greater challenges to the flexibility and reliability of collaborative control algorithms.

[0003] Currently, a typical approach to dynamic obstacle avoidance is a distributed cooperative control algorithm based on multi-agent systems. This method enables the robot to dynamically adjust its trajectory based on the states of neighboring nodes through local information exchange and consensus protocols. Specifically, the robot acquires the positional information of surrounding obstacles and its companions using visual sensors, predicts the future trajectories of obstacles using a time prediction model, and then generates an obstacle avoidance path using an optimized control strategy. Simultaneously, the system transmits key state data through a distributed communication network, ensuring consistency in global behavior and independence of local decisions, thereby achieving efficient collaboration without a central controller.

[0004] While the aforementioned approach demonstrates certain advantages in dynamic obstacle avoidance, it still suffers from significant limitations. First, information exchange relies on local communication networks, which are susceptible to signal interference or latency in complex environments, leading to path planning delays or even conflicts. Second, this method demands high accuracy in the obstacle motion model prediction; if the actual obstacle behavior deviates from the prediction model, avoidance may fail. Furthermore, distributed consensus protocols are prone to reduced response speed in high-density robot swarms due to competition for computing resources, making it difficult to meet the real-time requirements of large-scale systems. These issues limit the practical application effectiveness of this approach in highly dynamic and complex scenarios. Summary of the Invention

[0005] This application provides a distributed robot cooperative control method and system based on edge computing to solve the problems of low communication efficiency and poor adaptability in the existing distributed robot cooperative control technology.

[0006] In a first aspect, this application provides a distributed robot cooperative control method based on edge computing, comprising:

[0007] Collect environmental space data and motion parameters of dynamic obstacles from the distributed robot, and perform cross-robot joint compression processing on the environmental space data to obtain a distributed perception dataset;

[0008] Based on the spatial density differences of the distributed sensing dataset, a hierarchical communication link is dynamically constructed in the group of distributed robots, and the priority of the signal transmission path in the hierarchical communication link is adjusted according to the real-time changes of the spatial density differences to obtain the adjusted signal transmission path priority.

[0009] The adjusted signal transmission path priority is converted into communication path weight parameters, and combined with the motion parameters to generate a cooperative control instruction set;

[0010] The collaborative control instruction set is synchronized across nodes with the dynamic obstacles perceived by the distributed robot to obtain the synchronized collaborative control instruction set;

[0011] The synchronized collaborative control instruction set drives the distributed robot to perform edge computing resource interaction behavior, thereby obtaining the collaborative movement path of the distributed robot.

[0012] Optionally, the step of converting the adjusted signal transmission path priority into communication path weight parameters and generating a cooperative control instruction set in combination with the motion parameters includes:

[0013] The hierarchical communication link is divided into intervals based on the adjusted signal transmission path priority level value, and the level value is converted within the interval to obtain the weight parameter corresponding to the hierarchical communication link.

[0014] Extract the direction change and velocity change from the motion parameters, and then superimpose and merge the direction change and velocity change into a dynamic threat coefficient;

[0015] The weight parameters and the dynamic threat coefficient are jointly constrained to obtain the weight parameter distribution result. Based on the weight parameter distribution result, an instruction generation template corresponding to each level in the hierarchical communication link is constructed.

[0016] The dynamic threat coefficient is input into the instruction generation template to generate a cooperative control instruction set containing distributed robot motion constraints.

[0017] Optionally, the step of jointly constraining the weight parameters and the dynamic threat coefficient to obtain the weight parameter distribution result includes:

[0018] The range of change of the dynamic threat coefficient is divided into continuous sub-intervals, and each sub-interval corresponds to a transformation function;

[0019] Based on the hierarchical position of the weight parameter in the hierarchical communication link, obtain the initial conversion factor associated with the hierarchical position;

[0020] Within the sub-interval, the dynamic threat coefficient and the initial conversion factor are correlated according to the conversion function to obtain the correlated sub-interval parameters;

[0021] The parameters of the associated sub-intervals are fused by interval boundaries to obtain fused parameters. Inter-level balance constraints are then applied to the fused parameters to obtain the weight parameter distribution results.

[0022] Optionally, within the sub-interval, the step of performing a correlation calculation between the dynamic threat coefficient and the initial conversion factor according to the conversion function to obtain the correlated sub-interval parameters includes:

[0023] The slope adjustment factor of the transformation function is determined based on the span range of the sub-interval, and the dynamic threat coefficient is scaled according to the slope adjustment factor to generate a dynamic threat value;

[0024] Within the sub-interval, a hierarchical adjustment factor is obtained based on the hierarchical position corresponding to the weight parameter, and the dynamic threat value is cross-correlated with the hierarchical adjustment factor to obtain an initial conversion factor;

[0025] The initial conversion factor is subjected to interval accumulation processing, and the accumulation processing result is superimposed with the dynamic threat value to generate the associated sub-interval parameters.

[0026] Optionally, the process involves collecting environmental space data from the distributed robot and motion parameters of dynamic obstacles, and performing cross-robot joint compression processing on the environmental space data to obtain a distributed perception dataset, including:

[0027] The spatial perception range of the distributed robot is divided into overlapping regions, and the environmental spatial data is segmented into discrete block datasets containing redundant identifiers.

[0028] Within the discrete block dataset, duplicate feature matching is performed on the overlapping area data of adjacent distributed robots, and redundant data points exceeding the set duplicate threshold are deleted to generate a deredundant block dataset.

[0029] High-density feature points and static environment feature points are extracted from the motion parameters of dynamic obstacles in the deredundancy block dataset, and the high-density feature points are separated from the static environment feature points to generate dynamic feature datasets and static feature datasets.

[0030] The dynamic feature dataset is used to predict the trajectory extension of the dynamic obstacle according to its movement direction to obtain an extended dynamic feature sequence. The extended dynamic feature sequence is then spatially concatenated with the static feature dataset to generate a distributed sensing dataset.

[0031] Optionally, the step of predicting the trajectory extension of the dynamic feature dataset according to the motion direction of the dynamic obstacle to obtain an extended dynamic feature sequence includes:

[0032] Extract obstacle position coordinates for continuous time periods from the dynamic feature dataset, and calculate the motion vector of the dynamic obstacle based on the change in the obstacle position coordinates;

[0033] Using the obstacle's position coordinates in the current time period as a reference point, along the direction of the motion vector, and extending the speed distance of the dynamic obstacle by a preset multiple, a predicted trajectory point sequence is generated.

[0034] Spatial credibility verification is performed on the predicted trajectory point sequence. The verified predicted trajectory point sequence is then timestamped and merged with the dynamic feature dataset to obtain an extended dynamic feature sequence.

[0035] Optionally, the step of driving the distributed robot to perform edge computing resource interaction behavior through the synchronized cooperative control instruction set to obtain the cooperative movement path of the distributed robot includes:

[0036] The synchronized collaborative control instruction set is decomposed into action sequences corresponding to the distributed robot, and the action sequences include target displacement and execution priority labels;

[0037] Based on the execution priority label, the distributed robot is allocated a resource occupancy ratio for edge computing, and the calculation coverage range of dynamic obstacles is determined according to the resource occupancy ratio.

[0038] The boundary conditions of the calculation coverage area are dynamically adjusted according to the target displacement and the resource occupancy ratio, and the overlapping buffer area of ​​the distributed robot in the direction of movement is calculated based on the adjusted calculation coverage area.

[0039] The motion parameters of the dynamic obstacle and the displacement state of the distributed robot are aligned and then input into the overlapping buffer area to generate the cooperative motion path of the distributed robot.

[0040] Secondly, this application provides a distributed robot collaborative control system based on edge computing, comprising:

[0041] The acquisition module collects environmental space data and motion parameters of dynamic obstacles from the distributed robot, and performs cross-robot joint compression processing on the environmental space data to obtain a distributed perception dataset.

[0042] The adjustment module dynamically constructs hierarchical communication links in the group of distributed robots based on the spatial density differences of the distributed sensing dataset, and adjusts the priority of the signal transmission paths in the hierarchical communication links according to the real-time changes of the spatial density differences, so as to obtain the adjusted signal transmission path priority.

[0043] The generation module converts the adjusted signal transmission path priority into communication path weight parameters, and combines the motion parameters to generate a cooperative control instruction set.

[0044] The synchronization module synchronizes the cooperative control instruction set with the dynamic obstacles sensed by the distributed robot across nodes to obtain a synchronized cooperative control instruction set.

[0045] The driving module drives the distributed robot to perform edge computing resource interaction behavior through the synchronized collaborative control instruction set, so as to obtain the collaborative motion path of the distributed robot.

[0046] Thirdly, embodiments of this application provide a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a distributed robot cooperative control method based on edge computing as described in the first aspect above.

[0047] Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which, when executed by a computer, implements a distributed robot collaborative control method based on edge computing as described in the first aspect.

[0048] In this embodiment, environmental spatial data of distributed robots and motion parameters of dynamic obstacles are collected, and the environmental spatial data is subjected to cross-robot joint compression processing to obtain a distributed perception dataset. Based on the spatial density differences of the distributed perception dataset, hierarchical communication links are dynamically constructed in the group of distributed robots, and the priority of signal transmission paths in the hierarchical communication links is adjusted according to the real-time changes of the spatial density differences to obtain adjusted signal transmission path priorities. The adjusted signal transmission path priorities are converted into communication path weight parameters and combined with the motion parameters to generate a cooperative control instruction set. The cooperative control instruction set is synchronized across nodes with the dynamic obstacles perceived by the distributed robots to obtain a synchronized cooperative control instruction set. The synchronized cooperative control instruction set drives the distributed robots to perform edge computing resource interaction behavior to obtain the cooperative motion path of the distributed robots.

[0049] The technical solution of this application has the following beneficial effects:

[0050] This application utilizes cross-robot joint compression technology to eliminate redundancy in multi-source environmental data, significantly reducing communication bandwidth consumption and improving the transmission efficiency and real-time performance of perceived data. Based on spatial density differences, a hierarchical communication network is adaptively constructed, and signal transmission path priorities are dynamically adjusted to ensure centralized allocation of communication resources in high-density areas, enhancing data transmission reliability in critical areas. Priorities are quantified into weight parameters and combined with dynamic obstacle motion parameters to generate constraints, achieving deep coupling between communication resource allocation and motion safety control. A distributed timestamp alignment mechanism eliminates differences in instruction execution latency across multiple nodes, ensuring spatiotemporal consistency of multi-robot actions. Edge computing resources are dynamically allocated based on instructions, optimizing the collaborative efficiency of local path planning and global obstacle avoidance, ultimately generating highly secure collaborative motion paths.

[0051] Furthermore, the adjusted signal transmission path priority is divided into intervals and converted into weight parameters. The directional and velocity changes of dynamic obstacles are extracted to generate dynamic threat coefficients. Through joint constraint processing of the weight parameters and dynamic threat coefficients, a command generation template corresponding to the hierarchical communication link is constructed. The dynamic threat coefficient is then input into the template to generate a collaborative control command set containing motion constraints. By quantifying the correlation between communication priority and dynamic threat factors, deep integration of communication resource allocation and obstacle avoidance strategies is achieved. The joint constraint processing of the weight parameter distribution results and dynamic threat coefficients enables the command generation template to dynamically adapt to environmental changes, improving the communication efficiency and obstacle avoidance response capability of multi-robot systems in complex scenarios. Simultaneously, the motion constraints in the command generation template ensure that robot path planning considers both communication priority and safety boundaries, significantly reducing conflict risks and optimizing overall task execution efficiency.

[0052] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart of a distributed robot cooperative control method based on edge computing provided in this application is shown;

[0055] Figure 2 A schematic diagram of the structure of a distributed robot collaborative control system based on edge computing provided in this application is shown;

[0056] Figure 3 A schematic diagram of the structure of a computing device provided in this application is shown. Detailed Implementation

[0057] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0058] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0059] In scenarios such as industrial automation, intelligent logistics, and disaster relief, multi-robot systems need to collaboratively complete tasks in dynamic environments while avoiding moving obstacles. Existing distributed cooperative control algorithms based on multi-agent systems, while achieving path planning through local perception and distributed decision-making, rely on local communication networks to transmit obstacle motion states and neighboring node information. This makes them susceptible to signal interference or latency, leading to path planning delays or even conflicts. Furthermore, these methods require high accuracy in obstacle motion model predictions; if the actual obstacle behavior deviates from the prediction model—for example, sudden acceleration or directional changes—avoidance failures may occur. In high-density robot swarms, distributed consensus protocols suffer from reduced response speed due to competition for computing resources, making it difficult to meet the real-time requirements of large-scale systems. These issues limit the practical application of existing solutions in highly dynamic and complex scenarios, necessitating a cooperative control method that balances real-time performance, robustness, and low communication dependency.

[0060] To address the aforementioned issues, this application proposes a distributed robot cooperative control method based on edge computing. This method first generates a distributed perception dataset by jointly compressing environmental spatial data and dynamic obstacle motion parameters across robots, reducing communication bandwidth consumption and improving data transmission efficiency. Subsequently, it dynamically constructs hierarchical communication links based on spatial density differences and adjusts signal transmission path priorities in real time, optimizing communication resource allocation to adapt to signal interference and latency issues in complex environments. Furthermore, it transforms communication path priorities into weight parameters, combines them with dynamic obstacle motion parameters to generate a cooperative control instruction set, and eliminates data latency differences through cross-node synchronization, ensuring consistency between the instruction set and the real-time state of obstacles. Finally, through edge computing-driven resource interaction, it dynamically adjusts the overlapping buffer area and computational coverage in distributed robot path planning, achieving high-precision obstacle avoidance and efficient collaboration. This solution significantly improves the system's response speed and robustness to dynamic obstacles through the localized processing capabilities of edge computing, dynamic communication link optimization, and cross-node synchronization mechanisms. It solves the core defects of existing technologies, such as high communication dependence, sensitivity to prediction errors, and slow response in large-scale clusters, providing reliable technical support for multi-robot cooperative control in highly dynamic scenarios.

[0061] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] Figure 1 A flowchart illustrating a distributed robot cooperative control method based on edge computing, as provided in this application embodiment, is shown below. Figure 1 As shown, the method includes:

[0063] 101. Collect environmental space data and motion parameters of dynamic obstacles from the distributed robot, and perform cross-robot joint compression processing on the environmental space data to obtain a distributed perception dataset;

[0064] In this step, a distributed robot refers to a system composed of multiple robots with independent perception, decision-making, and execution capabilities, which achieve task objectives through collaborative communication and data sharing.

[0065] Environmental spatial data includes three-dimensional spatial information of the static and dynamic environment within the perception range of distributed robots, such as obstacle locations and terrain features.

[0066] The motion parameters of dynamic obstacles are used to describe the real-time motion state of dynamic obstacles, including position coordinates, velocity, change in direction, and change in rate.

[0067] Cross-robot joint compression processing refers to the use of multi-robot collaborative data redundancy removal and feature extraction techniques to compress raw environmental spatial data into a structured distributed sensing dataset.

[0068] The distributed sensing dataset integrates redundant environmental data with dynamic obstacle features to form a unified dataset containing static environmental feature points and dynamic obstacle trajectory prediction information.

[0069] In this embodiment, firstly, each distributed robot collects environmental spatial data using sensors such as LiDAR and cameras, and identifies dynamic obstacles and their motion parameters using target detection algorithms. Then, all robots upload the environmental spatial data and the motion parameters of the dynamic obstacles to a distributed data exchange layer, where a consensus algorithm verifies data consistency. Next, a joint compression algorithm is used to reduce the dimensionality of the environmental spatial data, removing duplicate static obstacle information, and parameterizing the trajectories of the dynamic obstacles. Finally, a distributed perception dataset containing global environmental features and predicted trajectories of dynamic obstacles is generated.

[0070] In the warehousing and logistics process, multiple mobile robots work together to transport goods. The robots use LiDAR to collect the position data of shelves and dynamic obstacles (such as pedestrians). After cross-robot joint compression processing, a distributed perception dataset containing shelf layout and pedestrian movement trajectory prediction is generated, providing accurate environmental information for subsequent communication and collaborative control.

[0071] 102. Based on the spatial density difference of the distributed sensing dataset, a hierarchical communication link is dynamically constructed in the group of distributed robots, and the priority of the signal transmission path in the hierarchical communication link is adjusted according to the real-time change of the spatial density difference to obtain the adjusted signal transmission path priority.

[0072] In this step, spatial density difference refers to the data density distribution of each region in the distributed sensing dataset, reflecting the environmental complexity and the density of robot distribution.

[0073] Layered communication links are communication hierarchies divided according to differences in spatial density. High-density areas use high-frequency communication links, while low-density areas use low-frequency communication links.

[0074] In a distributed robot system, the signal transmission path refers to the sequence of communication links through which data or control commands travel from the sending end to the receiving end.

[0075] The adjusted signal transmission path priority is a communication link priority parameter that is dynamically optimized based on differences in environmental spatial density and task requirements. It is used to guide the efficient transmission of data in a distributed robot system.

[0076] In this embodiment, firstly, based on the spatial density differences of the distributed sensing dataset, the system identifies high-density and low-density regions using a density clustering algorithm, and accordingly divides the hierarchical communication links into levels. Then, the bandwidth and transmission rate of the hierarchical communication links are dynamically adjusted, and the priority of signal transmission paths is assigned to different levels based on current task requirements and the threat coefficient of dynamic obstacles. For example, high-frequency communication links are used and transmission priority is increased in high-density regions, while communication resource consumption is reduced in low-density regions. Finally, the adjusted signal transmission path priorities are generated, providing a basis for the transformation of communication weight parameters.

[0077] During warehousing and logistics operations, when multiple robots enter narrow passages (high-density areas), the system dynamically prioritizes the communication links in that area to ensure rapid transmission of obstacle avoidance commands between robots. In open areas (low-density areas), the priority of communication links is reduced to conserve resources.

[0078] 103. The adjusted signal transmission path priority is converted into communication path weight parameters, and a cooperative control instruction set is generated in combination with the motion parameters;

[0079] In this step, the communication path weight parameter maps the adjusted signal transmission path priority to numerical weights, which are used to quantify the importance of the communication link.

[0080] The dynamic threat coefficient is a threat assessment index generated by comprehensively considering the changes in the direction and rate of dynamic obstacles, and is used to constrain collaborative control commands.

[0081] The collaborative control instruction set contains a set of instructions for the motion constraints of distributed robots, guiding distributed robots in collaborative obstacle avoidance and path planning.

[0082] In this embodiment, firstly, a weighted graph model is used to convert the signal transmission path priority into communication path weight parameters, where higher priority paths correspond to higher values ​​and lower priority paths correspond to lower values. Then, a distributed optimization algorithm is used to combine the communication path weight parameters with the motion parameters of dynamic obstacles. The weighted graph model is used to calculate the obstacle avoidance potential field between each distributed robot and the dynamic obstacles. The model predicts and controls the optimization of the robot's trajectory, ensuring that robots with higher communication path weights execute critical tasks first, ultimately generating a cooperative control instruction set.

[0083] In warehousing and logistics processes, when multiple robots are transporting goods in narrow passages, the system generates instructions to require the robots to decelerate and maintain a safe distance based on the weight parameters of high-priority communication links and the high threat coefficient of dynamic obstacles. In open areas, low weight parameters and low threat coefficients allow the robots to accelerate.

[0084] 104. Synchronize the cooperative control instruction set with the dynamic obstacles perceived by the distributed robot across nodes to obtain a synchronized cooperative control instruction set;

[0085] In this step, cross-node synchronization refers to ensuring that the collaborative control instruction set received by the distributed robot is consistent in time and content through timestamp alignment and consensus algorithms.

[0086] The synchronized collaborative control instruction set refers to the set of instructions that have been synchronized across nodes, enabling the robot to perform collaborative actions.

[0087] In this embodiment, a unified time base is allocated to the collaborative control instruction set using timestamp alignment technology, and a distributed consensus algorithm is employed to verify instruction consistency, eliminating deviations caused by communication delays. The system distributes the verified instruction set to all robot nodes, ensuring that the instructions received by each node are completely synchronized in content and time, ultimately generating a synchronized collaborative control instruction set. For example, during warehousing, multiple robots simultaneously receive obstacle avoidance instructions and adjust their paths, avoiding the risk of collisions caused by asynchronous instructions.

[0088] In the warehousing and logistics process, when multiple robots receive obstacle avoidance instructions, cross-node synchronization ensures that all robots adjust their paths at the same time to avoid collisions caused by instruction delays.

[0089] 105. Drive the distributed robot to perform edge computing resource interaction behavior through the synchronized collaborative control instruction set to obtain the collaborative motion path of the distributed robot.

[0090] In this step, the resource interaction behavior of edge computing refers to the distributed robot allocating computing resources on local edge nodes and dynamically adjusting the coverage of obstacles and path planning parameters.

[0091] Cooperative motion paths refer to robot motion paths generated based on resource interaction behavior, taking into account both task objectives and obstacle avoidance requirements.

[0092] In this embodiment, firstly, each distributed robot decomposes the synchronized cooperative control instruction set into action sequences containing target displacement and execution priority labels, and generates a preliminary path by combining the predicted trajectory of dynamic obstacles. Subsequently, resource allocation among multiple robots is coordinated through the preliminary path and a distributed resource scheduling algorithm to avoid computational resource contention. For example, robots with high-priority tasks preferentially occupy the computational resources of the edge server. Finally, the paths generated by all robots are verified by a conflict detection algorithm, and the final cooperative motion path is output.

[0093] During warehousing and logistics, robots dynamically adjust the allocation range of resources based on resource interaction behavior, generating overlapping buffer areas in narrow passages to ensure that multiple robots can still efficiently complete cargo handling tasks while slowing down and avoiding obstacles.

[0094] In summary, steps 101 to 105, by collecting environmental data and dynamic obstacle parameters, combined with cross-robot joint compression and hierarchical communication link construction, enable the system to achieve dynamic optimization allocation of communication resources. The generation and synchronization of the collaborative control instruction set ensures efficient obstacle avoidance and collaborative path planning for multiple robots in complex environments. Edge computing resource interaction further enhances the real-time performance and security of path generation, ultimately significantly improving the operational efficiency and task completion rate of robot swarms in scenarios such as warehousing and logistics.

[0095] To address the challenges of optimizing communication link priority and handling dynamic motion threats in distributed robot systems, this solution divides communication link intervals into hierarchical segments and converts priority levels into weight parameters. It then extracts direction and velocity changes from motion parameters to generate dynamic threat coefficients. Finally, it constructs an instruction generation template by combining constraint weight parameters and dynamic threat coefficients, ultimately generating a cooperative control instruction set containing motion constraints to achieve efficient multi-robot collaboration. In some embodiments, step 103, which involves converting the adjusted signal transmission path priority into communication path weight parameters and combining them with the motion parameters to generate a cooperative control instruction set, includes:

[0096] 201. Divide the hierarchical communication link into intervals according to the adjusted signal transmission path priority level value, and convert the level value within the interval to obtain the weight parameter corresponding to the hierarchical communication link.

[0097] In step 201, the adjusted signal transmission path priority level value refers to the communication demand priority of each layer in the hierarchical communication link. This level value is composed of data such as the stability of the communication link, bandwidth utilization, and task urgency. The hierarchical communication link refers to the physical layer, data link layer, network layer, transport layer, and application layer, based on the modern communication network architecture. Each layer corresponds to different communication functions and data transmission characteristics. Interval division refers to dividing the communication link into several sub-intervals according to the distribution range of the level values. Each sub-interval corresponds to a different communication resource allocation strategy. The weight parameter is a parameter obtained after numerical transformation, used to quantify the importance of each hierarchical communication link in the overall system. For example, high-priority communication links may receive higher weight values ​​to ensure priority transmission of critical data.

[0098] In this embodiment, the stability indicators, bandwidth utilization, and task urgency data of the communication link are first collected in real time using sensors and a communication module. Then, a dynamic programming algorithm is used to sort and normalize the adjusted signal transmission path priority levels, mapping the original data to a zero-to-one interval. Next, a mean-clustering algorithm is used to divide the normalized level values ​​into intervals, with each sub-interval corresponding to a specific communication resource allocation strategy. Finally, a weighted average algorithm is used to convert the level values ​​of each sub-interval into corresponding weight parameters, ensuring that the weight parameters reflect the dynamic priority requirements of each layer of communication links.

[0099] 202. Extract the direction change and velocity change from the motion parameters, and combine the direction change and velocity change to form a dynamic threat coefficient;

[0100] In step 202, the change in direction refers to the magnitude of the change in the robot's direction of motion per unit time, used to describe the instability of the robot's path. The change in velocity refers to the range of fluctuation in the robot's velocity per unit time, used to characterize the dynamics of the robot's motion. The dynamic threat coefficient is a comprehensive index obtained by weighting and superimposing the changes in direction and velocity, used to assess the potential collision risk or path conflict between robots. The higher the coefficient, the greater the uncertainty of the robot's motion, requiring a more stringent obstacle avoidance strategy.

[0101] In this embodiment, motion parameters, including changes in direction and changes in velocity, are first extracted from robot sensor data. The change in direction is calculated using a direction difference algorithm to obtain the angle difference between adjacent time points. The change in velocity is calculated using a velocity change rate algorithm to determine the absolute change in velocity. Next, a weighted superposition and merging technique is applied to linearly combine the changes in direction and velocity according to a preset proportional coefficient, forming a dynamic threat coefficient. A higher dynamic threat coefficient indicates a higher motion risk.

[0102] 203. Perform joint constraint processing on the weight parameters and the dynamic threat coefficient to obtain the weight parameter distribution result, and construct the instruction generation template corresponding to each level in the hierarchical communication link based on the weight parameter distribution result;

[0103] In step 203, the weight parameter distribution result refers to the parameter distribution diagram formed after joint constraint processing of the weight parameters and dynamic threat coefficients. Joint constraint processing refers to simultaneously handling multiple interrelated constraints in optimization or decision-making problems, ensuring that these constraints can be satisfied in uncertain or dynamically changing environments. The instruction generation template refers to a structured framework built based on template engine technology, used to map dynamic threat coefficients and weight parameters into specific collaborative control instruction sets. This template includes modules such as communication resource allocation rules, obstacle avoidance strategies, and path optimization algorithms, and can generate adapted control instructions based on input parameters.

[0104] In this embodiment, firstly, the weight parameters and dynamic threat coefficients are jointly constrained using a constraint optimization algorithm. The dynamic threat coefficient is used as the boundary condition, and the weight parameters are adjusted using linear programming techniques to ensure that the weight parameters maintain an optimal distribution under the constraint of the dynamic threat coefficient. The processing result forms the weight parameter distribution result, which reflects the adjusted importance of each level of the communication link. Then, based on the weight parameter distribution result, a template construction technique is used to design instruction generation templates for each communication link level. Finally, the generated instruction generation templates are passed to the next step for actual control instruction generation.

[0105] 204. Input the dynamic threat coefficient into the instruction generation template to generate a cooperative control instruction set containing distributed robot motion constraints.

[0106] In step 204, the cooperative control instruction set refers to the specific set of control commands generated based on the instruction generation template and the dynamic threat coefficient. This instruction set includes motion constraints for the distributed robots, such as speed limits, path offsets, and obstacle avoidance strategies, used to guide the cooperative work of the multi-robot system. The dynamic threat coefficient, as an input parameter, is used to adjust the constraints in the instruction generation template, ensuring that the instruction set can respond to environmental changes in real time.

[0107] In this embodiment, the dynamic threat coefficient is first injected as an input parameter into the instruction generation template. Next, an instruction filling algorithm is applied to replace the dynamic threat coefficient variable in the instruction generation template, and specific motion constraints are generated according to the preset logic of the instruction generation template. Then, through distributed collaborative processing technology, the outputs of instruction generation templates from all levels are integrated to form a complete collaborative control instruction set. Finally, the collaborative control instruction set is sent to the robot execution unit to achieve real-time control. The entire process ensures that the dynamic threat coefficient is effectively integrated into the control instructions, improving the system's responsiveness.

[0108] Here is a specific example:

[0109] In warehousing and logistics, when multiple mobile robots collaboratively handle goods, assuming three robots are moving boxes in a warehouse, the communication link is divided into three layers: control, coordination, and execution. First, the system uses a pre-defined communication priority level (e.g., high priority for the control layer, medium priority for the coordination layer, and low priority for the execution layer) to divide the priority level into three intervals using an equal-width interval partitioning algorithm. Within each interval, weight parameters are obtained through linear normalization: control layer weights, coordination layer weights, and execution layer weights. Second, data is extracted from the robot motion sensors, including the robot's directional and velocity changes (in meters). These are then weighted and combined using a weighted superposition and merging technique, resulting in a dynamic threat coefficient. Next, the weight parameters and the dynamic threat coefficient are jointly constrained. Linear programming is used to optimize the weight distribution, ensuring that the control layer weights are adjusted under the threat coefficient constraints. Then, an instruction generation template is constructed; for example, the control layer template includes "collision avoidance" logic. Finally, the dynamic threat coefficient is input into the template to generate a collaborative control instruction set, such as "robot decelerate and turn." All robots then coordinate to handle goods, avoiding collisions and improving efficiency.

[0110] In summary, steps 201 to 204, by prioritizing communication links and dynamically assessing motion risks, achieve high efficiency and safety in robot collaborative control, improve the system's adaptability in complex environments, and reduce conflicts and delays during task execution.

[0111] To address the dynamic adaptation problem of dynamic threat coefficients and communication link weight parameters in multi-autonomous mobile robot systems within a warehousing and logistics environment, the solution divides the dynamic threat coefficient into sub-intervals and designs a transformation function. It then obtains an initial transformation factor based on the communication link hierarchical position, associates the dynamic threat coefficient with the transformation factor within each sub-interval to generate parameters, and obtains an optimized weight parameter distribution after interval fusion and hierarchical balance constraints, thereby improving the dynamic adaptability of the communication weights. In some embodiments, step 203, which involves jointly constraining the weight parameters with the dynamic threat coefficient to obtain the weight parameter distribution, includes:

[0112] 301. Divide the range of change of the dynamic threat coefficient into continuous sub-intervals, with each sub-interval corresponding to a transformation function;

[0113] In step 301, the dynamic threat coefficient refers to a quantitative indicator representing the risk of robot movement. The range of variation refers to the upper and lower bounds of the coefficient's possible values. A sub-interval refers to a continuous small range segment after the range of variation is divided. The transformation function refers to the rules used to perform mathematical transformations on the values ​​within this sub-interval.

[0114] In this embodiment, firstly, the minimum and maximum values ​​of the dynamic threat coefficient are determined through historical data analysis or real-time monitoring. Then, an equal-width binning algorithm is used to divide the value range into several continuous and non-overlapping sub-intervals, each corresponding to a different threat level, such as low, medium, and high. Next, a corresponding transformation function is designed for each sub-interval. Finally, the dynamic threat coefficient value is input into the corresponding sub-interval and the transformation function to complete the initial mapping.

[0115] 302. Based on the hierarchical position of the weight parameter in the hierarchical communication link, obtain the initial conversion factor associated with the hierarchical position;

[0116] In step 302, the hierarchical position of the layered communication link refers to the specific layer of the communication link within a layered architecture, such as the physical layer, data link layer, network layer, transport layer, and application layer. The initial conversion factor is a preset parameter associated with the hierarchical position, used to quantify the basic weight of that layer in communication resource allocation. For example, the initial conversion factor for the physical layer may be lower, while the initial conversion factor for the application layer may be higher, to reflect the impact of different layers on system performance.

[0117] In this embodiment, firstly, the layered architecture of the communication link is identified, such as the physical layer, data link layer, network layer, transport layer, and application layer, and the specific layer position of each weight parameter is determined. Then, an initial conversion factor is assigned to each layer based on expert experience or historical data analysis. The setting of the initial conversion factor needs to comprehensively consider the contribution of each layer to communication stability and task urgency. For example, the application layer, because it is directly related to task execution, has a higher initial conversion factor, while the physical layer, because its basic transmission requirements are lower, has a smaller initial conversion factor.

[0118] 303. Within the sub-interval, the dynamic threat coefficient and the initial conversion factor are correlated according to the conversion function to obtain the correlated sub-interval parameters;

[0119] In step 303, the correlation calculation refers to combining the dynamic threat coefficient with the initial transformation factor and generating parameters corresponding to the sub-intervals through a transformation function. The correlated sub-interval parameters are intermediate results after mapping and are used for subsequent interval boundary fusion and inter-level balance constraints.

[0120] In this embodiment, firstly, the value of the dynamic threat coefficient is substituted into the sub-interval and matched with the corresponding transformation function. For example, if the dynamic threat coefficient belongs to the medium threat interval, a piecewise function is used. Subsequently, the output value of the transformation function is combined with the initial transformation factor to calculate the associated sub-interval parameters. For example, the output value of the transformation function is multiplied by the initial transformation factor to generate intermediate parameters reflecting the association between threat level and communication layer. This process, through function matching and calculation, quantitatively associates the dynamic threat coefficient with the communication link layer weight, forming operable intermediate parameters.

[0121] 304. Merge the parameters of the associated sub-intervals by interval boundaries to obtain the merged parameters, and apply inter-level balance constraints to the merged parameters to obtain the weight parameter distribution results.

[0122] In step 304, interval boundary fusion refers to eliminating parameter abrupt changes caused by differences in transformation functions between sub-intervals, ensuring a smooth transition of parameters within adjacent intervals. Inter-level balance constraints coordinate the weight parameters of different levels through a multi-objective optimization algorithm, ensuring that the allocation of communication resources meets the overall system requirements.

[0123] In this embodiment, all associated sub-interval parameters output in step 303 are first received. Next, a boundary smoothing fusion algorithm is used to perform interpolation at the sub-interval boundaries, merging the parameter values ​​into a fused parameter. Then, constraint optimization techniques are applied, and a linear programming algorithm is used to adjust the parameter values ​​to ensure that the weight parameter distribution at each level satisfies the balance condition. Finally, the weight parameter distribution result is output.

[0124] Here is a specific example:

[0125] In warehousing and logistics, when multiple mobile robots collaboratively handle goods, consider three robots working together to move boxes in a warehouse. The hierarchical communication link includes a control layer, a coordination layer, and an execution layer. First, the dynamic threat coefficient, ranging from its minimum to its maximum value, is divided into three continuous sub-intervals using an equal-width interval partitioning algorithm. A transformation function is assigned to each sub-interval; for example, a linear transformation function is used for the first sub-interval. Second, initial transformation factors are obtained based on the hierarchical position of the weight parameters. For example, a larger initial transformation factor corresponds to the control layer, a medium one to the coordination layer, and a smaller one to the execution layer. Next, the sub-interval corresponding to the current dynamic threat coefficient value is determined, and the transformation function for that sub-interval is applied. The dynamic threat coefficient is then weighted and multiplied by the initial transformation factor to obtain the associated sub-interval parameters. Then, all sub-interval parameters are fused at the boundaries, for example, using interpolation algorithms for smooth merging. Finally, inter-level balance constraint optimization ensures a balanced distribution of weight parameters across the control, coordination, and execution layers. The final weight parameter distribution result is output for subsequent collaborative control.

[0126] In summary, steps 301 to 304, by dividing the dynamic threat coefficient into sub-intervals and designing a transformation function, combined with the initial transformation factor of the hierarchical communication link and the balance constraints between levels, achieve dynamic adaptation of communication resource allocation to robot motion threats. This method can improve the priority response capability of communication links in warehousing and logistics environments, reduce the collision risk caused by communication delays, and ensure reasonable weight allocation at each level through a multi-objective optimization algorithm, thereby improving the collaborative efficiency and safety of multi-robot systems.

[0127] To address the dynamic adaptation issue of dynamic threat coefficients and communication link weight parameters in multi-autonomous mobile robot systems within a warehousing and logistics environment, this solution optimizes sub-interval parameters through cross-correlation of dynamic threat coefficients and weight parameters. The slope of the transformation function is adjusted based on the sub-interval span, the dynamic threat coefficient is scaled to generate a dynamic threat value, and an adjustment factor is obtained based on hierarchical position for cross-correlation. Finally, the correlated sub-interval parameters are generated through interval accumulation and superposition processing, enhancing the parameter distribution's responsiveness to dynamic threats. In some embodiments, step 303, within the sub-interval, involves calculating the correlation between the dynamic threat coefficient and the initial transformation factor based on the transformation function to obtain the correlated sub-interval parameters, including:

[0128] 401. Determine the slope adjustment factor of the conversion function based on the span range of the sub-interval, and scale the dynamic threat coefficient according to the slope adjustment factor to generate a dynamic threat value;

[0129] In step 401, the span of the sub-interval refers to the difference between the upper and lower limits of the dynamic threat coefficient for that sub-interval. The slope adjustment factor is a proportional coefficient calculated based on the span of the sub-interval, used to adjust the sensitivity of the transformation function. Scaling refers to numerically adjusting the dynamic threat coefficient according to this proportional coefficient, and the dynamic threat value refers to the standardized threat index obtained after scaling.

[0130] In this embodiment, firstly, the span range of the sub-intervals of the dynamic threat coefficient is determined through historical data analysis. Then, based on the size of the sub-interval span (e.g., a larger span results in a smaller slope adjustment factor), a linear interpolation algorithm is used to calculate the slope adjustment factor of the transformation function. Next, the dynamic threat coefficient is input into the scaling process, and a multiplication operation is used to multiply it by the slope adjustment factor to generate the dynamic threat value.

[0131] 402. Within the sub-interval, obtain the hierarchical adjustment factor according to the hierarchical position corresponding to the weight parameter, and cross-correlate the dynamic threat value with the hierarchical adjustment factor to obtain the initial conversion factor;

[0132] In step 402, the hierarchical position refers to the identifier of the communication link level to which the weight parameter belongs in the hierarchical structure. The hierarchical adjustment factor is a preset weight coefficient bound to the hierarchical position, used to reflect the importance differences between different levels. Cross-association refers to performing a mathematical correlation operation between the dynamic threat value and the hierarchical adjustment factor. The initial conversion factor is an intermediate adjustment parameter generated after the correlation operation.

[0133] In this embodiment, the hierarchical position of the weight parameters is first determined based on the layered architecture of the communication link, such as the physical layer and the application layer. Then, a hierarchical adjustment factor is assigned to each layer through expert experience or historical data analysis. Next, the dynamic threat value is multiplied by the hierarchical adjustment factor using a multiplicative correlation algorithm to obtain the initial conversion factor value.

[0134] 403. Perform interval accumulation processing on the initial conversion factor, and superimpose the accumulation processing result with the dynamic threat value to generate the associated sub-interval parameters.

[0135] In step 403, the interval accumulation processing refers to the process of summing all initial conversion factors within the current sub-interval. The accumulation processing result refers to the summed value. The superposition processing refers to adding the accumulation processing result to the dynamic threat value. The associated sub-interval parameter refers to the parameter generated after superposition, representing the final adjustment effect of the sub-interval.

[0136] In this embodiment, the initial conversion factors are first summed using an accumulator technique to obtain a cumulative processing result. Then, the dynamic threat value is added to the cumulative processing result to generate the associated sub-interval parameter value. This parameter is ultimately output, representing the comprehensive adjustment amount of the sub-interval under a specific threat level and level of importance.

[0137] Here is a specific example:

[0138] In warehousing and logistics, when multiple mobile robots collaboratively transport goods, suppose an execution-layer robot detects an obstacle during transport, and its dynamic threat coefficient falls into a preset sub-interval. First, the span of this sub-interval is calculated, and a slope adjustment factor is determined to a specific value based on an inverse proportional relationship. The dynamic threat coefficient is then multiplied by this factor to obtain the dynamic threat value. Second, based on the execution layer position to which the robot's weight parameters belong, a mapping table is consulted to obtain a smaller layer adjustment factor. The dynamic threat value is then multiplied by this factor to obtain an initial conversion factor. Next, the initial conversion factors of all robots within this sub-interval are summed, and the summation result is added to the dynamic threat value to finally generate the associated sub-interval parameters. These parameters are used to adjust the movement priority of the execution-layer robot, allowing it to decelerate appropriately during obstacle avoidance, while ensuring that the global scheduling commands of the control-layer robots are still efficiently transmitted, maintaining the overall coordination of the transport task.

[0139] In summary, steps 401 to 403 achieve dynamic adaptation between communication resource allocation and robot motion threats through scaling the slope adjustment factor of the dynamic threat coefficient, cross-correlation of the hierarchical adjustment factor, and interval cumulative superposition processing. This method can improve the priority response capability of communication links in warehousing and logistics environments, reduce the collision risk caused by communication delays, and eliminate local fluctuations through cumulative processing to ensure the stability of parameter distribution, thereby improving the collaborative efficiency and safety of multi-robot systems.

[0140] To address the low accuracy of environmental modeling in multi-distributed robot systems within warehousing and logistics environments due to redundant perception data and confusion between dynamic and static features, this solution improves environmental data perception efficiency through cross-robot joint compression processing. It divides overlapping perception ranges into discrete blocks, removes redundant data points to generate de-redundant blocks, separates dynamic obstacles from static environmental features, and combines trajectory prediction to expand dynamic features and concatenate them with static features to generate a distributed perception dataset, providing accurate data support for collaborative control. In some embodiments, step 101 involves collecting environmental spatial data from distributed robots and motion parameters of dynamic obstacles, and performing cross-robot joint compression processing on the environmental spatial data to obtain a distributed perception dataset, including:

[0141] 501. Divide the spatial perception range of the distributed robot into overlapping areas, and segment the environmental spatial data into discrete block datasets containing redundant identifiers;

[0142] In step 501, the spatial perception range refers to the three-dimensional spatial area that a single robot sensor can detect. Overlapping region segmentation refers to the process of marking the boundaries of the overlapping portions of multiple robot perception ranges. Environmental spatial data refers to the raw environmental information collected by all robots. Redundancy markers are labels that mark the repetition of data. Discrete block datasets refer to labeled subsets of data formed after the environmental space is cut according to overlapping regions.

[0143] In this embodiment, firstly, environmental spatial data is collected using distributed robot sensors to obtain the coverage area of ​​each robot over its surroundings. Then, a rasterization algorithm is used to divide the environmental space into regular discrete blocks, each recording spatial coordinates and perception data. Next, a spatial intersection detection algorithm is used to identify overlapping areas in the perception ranges of adjacent robots, and redundant markers are added to the discrete blocks within these overlapping areas, indicating that multiple robots are sharing the same perception data in that block. Finally, all discrete blocks are classified according to their redundancy markers, generating a discrete block dataset containing these markers. This process, through spatial partitioning and redundancy marking, provides a structured data foundation for subsequent data deduplication.

[0144] 502. Within the discrete block dataset, perform duplicate feature matching on the overlapping area data of adjacent distributed robots, delete redundant data points that exceed the set duplicate threshold, and generate a deredundant block dataset.

[0145] In step 502, overlapping region data refers to block data that is simultaneously perceived by multiple robots. Duplicate feature matching processing refers to the process of identifying and comparing identical block data collected by different robots. The duplication threshold refers to the upper limit on the number of times data redundancy is determined. The deredundant block dataset is a simplified dataset after removing redundant data.

[0146] In this embodiment, firstly, overlapping region data with redundant markers are extracted from the discrete block dataset. Then, a feature hashing algorithm is used to compare the similarity of perception data from adjacent distributed robots in the overlapping regions, identifying duplicate feature points, such as obstacle edges at the same location. Next, the number of duplicate feature points within each discrete block dataset is counted; if it exceeds a preset duplicate threshold, redundant data points are deleted, retaining only the unique valid data. Finally, the processed discrete block datasets are integrated to generate a redundant-free de-redundant block dataset.

[0147] 503. Extract high-density feature points and static environment feature points from the motion parameters of the dynamic obstacles in the deredundancy block dataset, and separate the high-density feature points from the static environment feature points to generate dynamic feature datasets and static feature datasets.

[0148] In step 503, dynamic obstacles refer to moving people or objects. Motion parameters include speed and direction. High-density feature points refer to key points densely distributed along the trajectory of the dynamic obstacle. Static environment feature points refer to the contour feature points of stationary objects. The dynamic feature dataset and the static feature dataset are two separate sets of features.

[0149] In this embodiment, firstly, motion parameters of dynamic obstacles, such as velocity and rate of change of direction, are extracted from the deredundant block dataset. Then, a density clustering algorithm is used to identify high-density feature points in the dynamic obstacle trajectory, such as densely distributed movement path points. Simultaneously, time series analysis is used to distinguish static environmental feature points, such as fixed shelves and walls. Next, the high-density feature points and static environmental feature points are stored in the dynamic feature dataset and static feature dataset, respectively. This process provides data support for trajectory prediction and environmental modeling.

[0150] 504. The dynamic feature dataset is extended and predicted according to the movement direction of the dynamic obstacle to obtain an extended dynamic feature sequence, and the extended dynamic feature sequence is spatially concatenated with the static feature dataset to generate a distributed perception dataset.

[0151] In step 504, trajectory extension prediction refers to predicting the possible future trajectories of dynamic obstacles based on their historical motion parameters using an algorithm. Extended dynamic feature sequence refers to a set of dynamic feature data containing the predicted trajectory points. Spatial stitching refers to merging the extended dynamic feature sequence with the static feature dataset in the spatial dimension to generate a distributed sensing dataset.

[0152] In this embodiment, firstly, the Kalman filter algorithm is used to predict the trajectory extension of high-density feature points in the dynamic feature dataset, generating future trajectory points of dynamic obstacles and forming an extended dynamic feature sequence. Subsequently, the extended dynamic feature sequence is concatenated with the static feature dataset using a spatial coordinate alignment algorithm, integrating the dynamic obstacle trajectory and static environmental information to generate a distributed perception dataset. This process, through trajectory prediction and spatial fusion, achieves dynamic updating and global consistency of the environmental model, providing complete perception data for multi-robot collaborative decision-making.

[0153] Here is a specific example:

[0154] In warehousing and logistics, three mobile robots collaboratively handle large shelving units, scanning the environment with their LiDAR. First, the system detects an overlap between the perception areas of Robot 1 and Robot 2 at the shelving aisle, dividing this area into discrete blocks and marking them as redundant. Second, for this overlapping block, the system compares the shelving point cloud data collected by the two robots, removing duplicate shelving support point clouds to generate a de-redundant dataset. Next, it identifies high-density feature points of the mobile forklift and static shelving feature points from the dataset, separating them into two datasets. Finally, based on the forklift's movement direction, it predicts its future trajectory, generating an extended feature sequence, which is then combined with the static shelving features to form a fused environment model. Each robot adjusts its path according to this model to avoid collisions with the forklift and improve collaborative efficiency.

[0155] In summary, steps 501 to 504 achieve efficient modeling and dynamic updating of environmental data through redundant identification of the distributed robot's perception range, deduplication of duplicate data, separation of dynamic and static features, and trajectory prediction and stitching. This method can reduce the storage and computational pressure caused by data redundancy in multi-robot systems, improve the accuracy of dynamic obstacle trajectory prediction, and ensure the consistency of global environmental perception through the fusion of static features and dynamic trajectories, thereby improving the safety and efficiency of multi-robot collaborative operations in warehousing and logistics.

[0156] To address the path conflict problem caused by inaccurate dynamic obstacle trajectory prediction in multi-autonomous mobile robot systems within a warehousing and logistics environment, the solution focuses on the prediction and verification of dynamic obstacle trajectories. By extracting obstacle position coordinates over continuous time periods to calculate motion vectors, and extending these vectors to generate a sequence of predicted trajectory points, this sequence is aligned and merged with the dynamic feature dataset after spatial reliability verification to generate an extended dynamic feature sequence. This provides future state information of dynamic obstacles for path planning. In some embodiments, step 504, which involves extending and predicting the trajectory of the dynamic feature dataset according to the motion direction of the dynamic obstacle to obtain the extended dynamic feature sequence, includes:

[0157] 601. Extract obstacle position coordinates for continuous time periods from the dynamic feature dataset, and calculate the motion vector of the dynamic obstacle based on the change in the obstacle position coordinates;

[0158] In step 601, the dynamic feature dataset refers to a high-density set of feature points of the obstacle, containing the obstacle position coordinates for a continuous time period. A continuous time period refers to a sequence of obstacle positions recorded at fixed time intervals. The change in obstacle position coordinates is a vector representation of the difference between coordinates in adjacent time periods, reflecting the displacement direction and distance of the obstacle. The motion vector is a composite vector of the obstacle's velocity and direction within the continuous time period, used to characterize its motion state.

[0159] In this embodiment, firstly, the obstacle's position coordinates over multiple consecutive time periods are obtained from the dynamic feature dataset. Then, the change in the obstacle's position coordinates between adjacent time periods is calculated to generate a displacement vector. Next, a Kalman filter algorithm is used to smooth the displacement vector, eliminating noise interference, and finally obtaining the motion vector of the dynamic obstacle. This process achieves accurate modeling of the obstacle's motion state.

[0160] 602. Using the obstacle's position coordinates in the current time period as a reference point, along the direction of the motion vector, and extending the speed distance of the dynamic obstacle by a preset multiple, generate a sequence of predicted trajectory points;

[0161] In step 602, the reference point refers to the obstacle's position coordinates in the current time period, used as the starting point of the predicted trajectory. The preset multiplier refers to a parameter dynamically adjusted based on the obstacle's historical movement trends and the environment, used to control the length of the trajectory extension. The rate distance characterizes the relationship between the speed of an object or signal's movement and its coverage area under specific conditions. The predicted trajectory point sequence refers to multiple future position points generated along the direction of the motion vector, reflecting the obstacle's possible movement path.

[0162] In this embodiment, firstly, the obstacle's position coordinates in the current time period are used as the starting point of the predicted trajectory. Then, its direction angle and velocity values ​​are extracted along the motion vector direction, and combined with the speed distance of the obstacle extended by a preset multiple, to generate predicted trajectory points for future times. Next, a linear interpolation algorithm is used to generate a continuous sequence of predicted trajectory points. This process achieves dynamic prediction of the obstacle's trajectory through direction extension and speed adjustment.

[0163] 603. Perform spatial credibility verification processing on the predicted trajectory point sequence, and merge the verified predicted trajectory point sequence with the dynamic feature dataset by timestamp alignment to obtain the extended dynamic feature sequence.

[0164] In step 603, spatial credibility verification processing refers to determining the rationality of the trajectory points by comparing the probability of conflict between the predicted trajectory points and surrounding environmental data, such as static obstacles and other dynamic obstacles. Timestamp alignment refers to synchronizing the predicted trajectory point sequence with the dynamic feature dataset according to a time period to ensure data consistency. Extended dynamic feature sequence refers to the fusion result of the verified predicted trajectory points and historical dynamic feature points.

[0165] In this embodiment, firstly, an environmental collision detection algorithm is used to compare the predicted trajectory point sequence with a static map, eliminating invalid trajectory points that overlap with obstacles. Simultaneously, kinematic feasibility analysis is applied to filter trajectory points that exceed robot dynamics constraints. If the verification passes, the trajectory point is retained. Subsequently, the verified predicted trajectory point sequence is synchronized with the dynamic feature dataset according to a time period. Finally, the two are merged into an extended dynamic feature sequence. This process ensures the reliability of the predicted trajectory and data consistency.

[0166] Here is a specific example:

[0167] During warehousing and logistics, when two mobile robots are collaboratively moving boxes of goods, a LiDAR scanner detects a moving forklift. First, the forklift's position coordinates from the most recent few frames are extracted, and the difference between adjacent frame coordinates is calculated to obtain a motion vector, oriented at a certain angle east-northeast, with a constant speed. Second, using the latest coordinates as a reference, several predicted trajectory points are generated along this direction at preset time units, spaced at a certain distance. Next, it is verified whether the predicted points overlap with the shelving area. It was found that a predicted point at a certain time point was located in a prohibited area of ​​the shelving aisle, so this point and its subsequent points were deleted. The remaining predicted points are then timestamped and merged with the actual forklift trajectory points to form an extended sequence. The robot adjusts its path accordingly, avoiding the forklift's predicted path in advance.

[0168] In summary, steps 601 to 603, through dynamic obstacle position coordinate extraction and motion vector calculation, trajectory extension prediction, and spatial reliability verification, achieve accurate prediction of obstacle trajectories and fusion update of dynamic feature data. This method can improve the perception and obstacle avoidance capabilities of multi-robot systems in warehousing and logistics environments, reduce the risk of path conflicts caused by trajectory prediction deviations, and ensure data consistency through timestamp alignment, thereby improving the safety and efficiency of collaborative handling.

[0169] To address path conflicts and resource allocation issues when multiple mobile robots collaboratively handle goods in warehousing and logistics processes, this solution optimizes the collaborative motion path through resource interaction behavior. It decomposes the collaborative control instruction set into action sequences and allocates edge computing resources accordingly. Combined with dynamic adjustment of the calculation coverage boundary based on displacement, it generates an overlapping buffer zone in the movement direction. Dynamic obstacle parameters are aligned with the robot's displacement state and input into the buffer zone to generate a collaborative motion path that balances resource allocation and obstacle avoidance. In some embodiments, step 105, which involves driving the distributed robots to perform edge computing resource interaction behavior through the synchronized collaborative control instruction set to obtain the collaborative motion path of the distributed robots, includes:

[0170] 701. Decompose the synchronized cooperative control instruction set into action sequences corresponding to the distributed robot, wherein the action sequence includes target displacement and execution priority label;

[0171] In step 701, the synchronized cooperative control instruction set is a set of target actions for multiple robots. An action sequence refers to breaking down instructions into specific execution steps. The target displacement represents the distance and direction the robot needs to move. The execution priority label indicates the urgency of the action.

[0172] In this embodiment, firstly, a collaborative control instruction set is received from the central control system, containing the target positions and cooperation rules of all robots involved in the handling. Then, a task decomposition algorithm is used to break down the collaborative control instruction set into a sequence of actions for each robot. Each action sequence includes a target displacement generated by the path planning module based on task requirements, and an execution priority label dynamically allocated based on task urgency and resource consumption needs.

[0173] 702. Allocate the edge computing resource occupancy ratio to the distributed robot based on the execution priority label, and determine the calculation coverage range of dynamic obstacles according to the resource occupancy ratio;

[0174] In step 702, edge computing refers to distributed computing node resources. Resource occupancy ratio refers to the percentage of computing resources allocated to each robot. Computational coverage refers to the maximum spatial area that a single robot can process obstacle data in real time.

[0175] In this embodiment, firstly, an edge computing resource allocation ratio is assigned to each distributed robot using a weighted resource allocation algorithm based on the execution priority label. High-priority robots receive a higher resource allocation, while low-priority robots receive a lower allocation. Subsequently, by combining real-time motion data of dynamic obstacles, such as the position and speed of a forklift, a load balancing model is used to determine the computational coverage area of ​​the dynamic obstacles. For example, when a high-priority robot needs to avoid a high-speed moving forklift, its corresponding computational coverage area will be expanded to include more predicted trajectory points, ensuring that the edge computing nodes can process obstacle avoidance decisions in a timely manner.

[0176] 703. Dynamically adjust the boundary conditions of the calculation coverage area according to the target displacement and the resource occupancy ratio, and calculate the overlapping buffer area of ​​the distributed robot in the direction of movement based on the adjusted calculation coverage area;

[0177] In step 703, boundary conditions refer to the spatial constraint parameters of the calculated coverage area. The overlapping buffer region is a common obstacle avoidance area formed by the intersection of the calculated coverage areas of multiple robots.

[0178] In this embodiment, firstly, a directional weighting algorithm is used to expand the coverage boundary conditions in the direction of the target displacement and shrink the boundary in the opposite direction, based on the target displacement and resource occupancy ratio. Next, based on the adjusted calculated coverage area, spatial overlap detection technology is executed to calculate the intersection of coverage areas in the direction of distributed robot movement. Finally, a safety boundary expansion algorithm is used to extend the intersection area by a preset buffer distance, generating an overlapping buffer area.

[0179] 704. Align the motion parameters of the dynamic obstacle and the displacement state of the distributed robot and input them into the overlapping buffer area to generate the cooperative motion path of the distributed robot.

[0180] In step 704, the motion parameters of the dynamic obstacle include velocity and direction. The displacement state of the distributed robot refers to the real-time position and velocity of multiple robots with independent perception, decision-making, and execution capabilities. Alignment processing is the operation of synchronizing the two types of data in time. The cooperative motion path is the final generated collision-avoidance motion trajectory.

[0181] In this embodiment, firstly, the motion parameters of the dynamic obstacles and the real-time displacement states of the distributed robots are synchronized using timestamps. The aligned data is then input into an overlapping buffer region. A multi-agent potential field obstacle avoidance algorithm is used to calculate the obstacle repulsion trajectory for each robot within the overlapping buffer region. Simultaneously, distributed cooperative optimization technology is employed to balance path conflicts and ensure global motion consistency. Finally, a collision-free cooperative motion path is output, and each robot performs its displacement action according to this path.

[0182] Here is a specific example:

[0183] During warehousing and logistics, when three mobile robots are collaboratively transporting goods, a LiDAR detector detects a high-speed forklift entering the work area. The central control system first breaks down the synchronized collaborative control command set into action sequences for the three robots. Robot 1, which needs to urgently avoid the forklift, is marked as high priority, while Robots 2 and 3 are marked as medium priority. Then, based on the execution priority labels, a weighted resource allocation algorithm is used to allocate the resource occupancy ratio of edge computing nodes to the distributed robots. Combined with real-time motion data of dynamic obstacles, the computational coverage of Robot 1 is dynamically expanded to include the predicted trajectory of the forklift over a certain period. Next, based on the target displacement and resource occupancy ratio, a spatial topology analysis algorithm is used to adjust the boundary conditions of the computational coverage, generating an overlapping buffer zone in the path intersection area of ​​Robots 1 and 2, reserving a safe distance. Finally, the motion parameters of the dynamic obstacles are timestamped and input into the overlapping buffer zone after being aligned with the displacement states of the distributed robots, generating a collaborative motion path. Robot 1 moves along the outside of the buffer zone to avoid the forklift, while Robots 2 and 3 pass through the buffer zone in staggered shifts, ensuring efficient collaboration among multiple robots without collision risk.

[0184] In summary, steps 701 to 704, by breaking down collaborative instructions, dynamically allocating resources, adjusting coverage boundaries, and generating buffer zones, achieve efficient collaboration among multiple robots in complex dynamic environments. In warehousing and logistics scenarios, this method significantly reduces the risk of collisions between robots and dynamic obstacles, improves the real-time performance of path planning and resource utilization, while ensuring the priority execution of high-priority tasks, thereby improving overall handling efficiency and system stability.

[0185] Figure 2 This application provides a schematic diagram of the structure of a distributed robot collaborative control system based on edge computing, as shown in the embodiment of the present application. Figure 2 As shown, the system includes:

[0186] The acquisition module 21 acquires environmental space data of the distributed robot and motion parameters of dynamic obstacles, and performs cross-robot joint compression processing on the environmental space data to obtain a distributed perception dataset.

[0187] The adjustment module 22 dynamically constructs hierarchical communication links in the group of distributed robots based on the spatial density differences of the distributed sensing dataset, and adjusts the priority of the signal transmission path in the hierarchical communication link according to the real-time changes of the spatial density differences to obtain the adjusted signal transmission path priority.

[0188] The generation module 23 converts the adjusted signal transmission path priority into communication path weight parameters, and combines the motion parameters to generate a cooperative control instruction set;

[0189] Synchronization module 24 synchronizes the collaborative control instruction set with the dynamic obstacles perceived by the distributed robot across nodes to obtain a synchronized collaborative control instruction set;

[0190] The drive module 25 drives the distributed robot to perform edge computing resource interaction behavior through the synchronized collaborative control instruction set, so as to obtain the collaborative motion path of the distributed robot.

[0191] Figure 2 The aforementioned edge computing-based distributed robot collaborative control system can execute... Figure 1 The implementation principle and technical effects of the edge computing-based distributed robot cooperative control method described in the illustrated embodiment will not be repeated here. The specific methods by which each module and unit performs operations in the edge computing-based distributed robot cooperative control system described in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0192] In one possible design, Figure 2 The edge computing-based distributed robot collaborative control system of the illustrated embodiment can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0193] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0194] The processing component 32 is used for the above Figure 1 The embodiment describes a distributed robot collaborative control method based on edge computing.

[0195] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0196] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0197] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0198] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0199] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0200] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0201] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The illustrated embodiment presents a distributed robot collaborative control method based on edge computing.

[0202] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0203] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0204] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0205] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A distributed robot cooperative control method based on edge computing, characterized in that, include: Collect environmental space data and motion parameters of dynamic obstacles from distributed robots, and perform cross-robot joint compression processing on the environmental space data to obtain a distributed perception dataset. The cross-robot joint compression processing refers to compressing the original environmental space data into a structured distributed perception dataset through multi-robot collaborative data redundancy removal and feature extraction techniques. Based on the spatial density differences of the distributed sensing dataset, a hierarchical communication link is dynamically constructed in the group of distributed robots, and the priority of the signal transmission path in the hierarchical communication link is adjusted according to the real-time changes of the spatial density differences to obtain the adjusted signal transmission path priority. The adjusted signal transmission path priority is converted into communication path weight parameters, and combined with the motion parameters to generate a cooperative control instruction set; The collaborative control instruction set is synchronized across nodes with the dynamic obstacles perceived by the distributed robot to obtain the synchronized collaborative control instruction set; The synchronized collaborative control instruction set drives the distributed robot to perform resource interaction behavior of edge computing in order to obtain the collaborative movement path of the distributed robot; The process involves collecting environmental space data from the distributed robot and motion parameters of dynamic obstacles, and performing cross-robot joint compression processing on the environmental space data to obtain a distributed perception dataset, including: The spatial perception range of the distributed robot is divided into overlapping regions, and the environmental spatial data is segmented into discrete block datasets containing redundant identifiers. Within the discrete block dataset, duplicate feature matching is performed on the overlapping area data of adjacent distributed robots, and redundant data points exceeding the set duplicate threshold are deleted to generate a deredundant block dataset. High-density feature points and static environment feature points are extracted from the motion parameters of dynamic obstacles in the deredundancy block dataset, and the high-density feature points are separated from the static environment feature points to generate dynamic feature datasets and static feature datasets. The dynamic feature dataset is used to predict the trajectory extension of the dynamic obstacle according to its movement direction to obtain an extended dynamic feature sequence. The extended dynamic feature sequence is then spatially concatenated with the static feature dataset to generate a distributed sensing dataset.

2. The method according to claim 1, characterized in that, The step of converting the adjusted signal transmission path priority into communication path weight parameters and generating a cooperative control instruction set in combination with the motion parameters includes: The hierarchical communication link is divided into intervals based on the adjusted signal transmission path priority level value, and the level value is converted within the interval to obtain the weight parameter corresponding to the hierarchical communication link. Extract the direction change and velocity change from the motion parameters, and then superimpose and merge the direction change and velocity change into a dynamic threat coefficient; The weight parameters and the dynamic threat coefficient are jointly constrained to obtain the weight parameter distribution result. Based on the weight parameter distribution result, an instruction generation template corresponding to each level in the hierarchical communication link is constructed. The dynamic threat coefficient is input into the instruction generation template to generate a cooperative control instruction set containing distributed robot motion constraints.

3. The method according to claim 2, characterized in that, The step of jointly constraining the weight parameters and the dynamic threat coefficient to obtain the weight parameter distribution result includes: The range of change of the dynamic threat coefficient is divided into continuous sub-intervals, and each sub-interval corresponds to a transformation function; Based on the hierarchical position of the weight parameter in the hierarchical communication link, obtain the initial conversion factor associated with the hierarchical position; Within the sub-interval, the dynamic threat coefficient and the initial conversion factor are correlated according to the conversion function to obtain the correlated sub-interval parameters; The parameters of the associated sub-intervals are fused by interval boundaries to obtain fused parameters. Inter-level balance constraints are then applied to the fused parameters to obtain the weight parameter distribution results.

4. The method according to claim 3, characterized in that, Within the sub-interval, the dynamic threat coefficient and the initial conversion factor are correlated according to the conversion function to obtain the correlated sub-interval parameters, including: The slope adjustment factor of the transformation function is determined based on the span range of the sub-interval, and the dynamic threat coefficient is scaled according to the slope adjustment factor to generate a dynamic threat value; Within the sub-interval, a hierarchical adjustment factor is obtained based on the hierarchical position corresponding to the weight parameter, and the dynamic threat value is cross-correlated with the hierarchical adjustment factor to obtain an initial conversion factor; The initial conversion factor is subjected to interval accumulation processing, and the accumulation processing result is superimposed with the dynamic threat value to generate the associated sub-interval parameters.

5. The method according to claim 1, characterized in that, The step of extending and predicting the trajectory of the dynamic feature dataset according to the motion direction of the dynamic obstacle to obtain an extended dynamic feature sequence includes: Extract obstacle position coordinates for continuous time periods from the dynamic feature dataset, and calculate the motion vector of the dynamic obstacle based on the change in the obstacle position coordinates; Using the obstacle's position coordinates in the current time period as a reference point, along the direction of the motion vector, and extending the speed distance of the dynamic obstacle by a preset multiple, a predicted trajectory point sequence is generated. Spatial credibility verification is performed on the predicted trajectory point sequence. The verified predicted trajectory point sequence is then timestamped and merged with the dynamic feature dataset to obtain an extended dynamic feature sequence.

6. The method according to claim 1, characterized in that, The step of driving the distributed robot to perform edge computing resource interaction behavior through the synchronized cooperative control instruction set to obtain the cooperative movement path of the distributed robot includes: The synchronized collaborative control instruction set is decomposed into action sequences corresponding to the distributed robot, and the action sequences include target displacement and execution priority labels; Based on the execution priority label, the distributed robot is allocated a resource occupancy ratio for edge computing, and the calculation coverage range of dynamic obstacles is determined according to the resource occupancy ratio. The boundary conditions of the calculation coverage are dynamically adjusted according to the target displacement and the resource occupancy ratio, and the overlapping buffer area of ​​the distributed robot in the direction of movement is calculated based on the adjusted calculation coverage. The motion parameters of the dynamic obstacle and the displacement state of the distributed robot are aligned and then input into the overlapping buffer area to generate the cooperative motion path of the distributed robot.

7. A distributed robot cooperative control system based on edge computing, applied to the distributed robot cooperative control method based on edge computing as described in any one of claims 1-6, characterized in that, include: The acquisition module collects environmental space data and motion parameters of dynamic obstacles from the distributed robot, and performs cross-robot joint compression processing on the environmental space data to obtain a distributed perception dataset. The adjustment module dynamically constructs hierarchical communication links in the group of distributed robots based on the spatial density differences of the distributed sensing dataset, and adjusts the priority of the signal transmission paths in the hierarchical communication links according to the real-time changes of the spatial density differences, so as to obtain the adjusted signal transmission path priority. The generation module converts the adjusted signal transmission path priority into communication path weight parameters, and combines the motion parameters to generate a cooperative control instruction set. The synchronization module synchronizes the collaborative control instruction set with the dynamic obstacles perceived by the distributed robot across nodes to obtain a synchronized collaborative control instruction set. The driving module drives the distributed robot to perform edge computing resource interaction behavior through the synchronized collaborative control instruction set, so as to obtain the collaborative motion path of the distributed robot.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a distributed robot cooperative control method based on edge computing as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The system contains a computer program that, when executed by a computer, implements a distributed robot cooperative control method based on edge computing as described in any one of claims 1 to 6.

Images