A robot cluster optical network communication method and system based on optical communication data

Abstract

The application discloses a kind of robot cluster optical network communication method and system based on optical communication data, it is related to robot communication field, this method includes: based on the partition instruction of coupling communication link, central control module, robot is logically partitioned;Robot establishes the optical communication network between each logical partition inside robot according to logical partition and self attitude adjustment positioning mechanism;Based on the local ordering of each logical partition by partition bidirectional bubble merge sorting algorithm;According to fuzzy immune clone selection algorithm, determine the minimum robot and maximum robot of each logical partition, boundary information;Data merging and boundary ordering are carried out between each logical partition;According to distributed negotiation model, global broadcast is carried out.The application solves the problems of low efficiency of existing distributed sorting, wireless communication bottleneck and consistency by integrated process of robot partition efficient sorting, boundary collaborative merging, distributed broadcast negotiation and the like.

Description

Technical Field

[0001] The present invention relates to the field of robot communication, and in particular to a robot cluster optical network communication method and system based on optical communication data. Background Art

[0002] Utilizing high-speed lasers, visible light, or other optical carriers, physical-layer optical communication channels enable data exchange, coordinated control, and distributed information sharing among multiple robots equipped with optical communication modules. This communication method relies on specialized optical communication antennas, photoelectric receivers, and modems, connecting the robot nodes into a high-bandwidth, low-latency, and highly real-time distributed network. This significantly surpasses the link capacity and security of traditional radio frequency communications.

[0003] In this architecture, a central control module or any node can broadcast control instructions, synchronize status, or forward task information to specific robots or all robots in the swarm via optical channels. This type of optical network communication in robot clusters provides a strong communication foundation for large-scale multi-robot collaborative tasks (such as intelligent sorting, automated warehousing, and unmanned inspection), achieving efficient, stable, and scalable data interconnection.

[0004] However, existing robot cluster communication methods still have the following shortcomings:

[0005] (1) Existing robot clusters mostly adopt simple distributed sorting or centralized scheduling, which easily leads to efficiency bottlenecks when facing high concurrency, multi-partition member merging and dynamic identification of partition boundaries. There is a lack of an efficient multi-partition sorting and merging mechanism that combines parallelism and local self-coordination.

[0006] (2) When current distributed robot systems conduct collaborative broadcasting, the lack of refined scheduling and confirmation mechanisms often leads to disordered message transmission, partial message loss or duplication, which in turn affects the consistency of distributed task decomposition and global coordination.

[0007] Currently, no effective solutions have been proposed for the problems in related technologies. Summary of the Invention

[0008] In response to the problems in the related art, the present invention proposes a robot cluster optical network communication method and system based on optical communication data to overcome the above-mentioned technical problems existing in the existing related art.

[0009] To this end, the specific technical solutions adopted in the present invention are as follows:

[0010] A robot cluster optical network communication method and system based on optical communication data, comprising:

[0011] Establishing a coupled communication link between the robot integrated with the optical communication module and the central control module; logically partitioning the robot based on the coupled communication link and according to the partitioning instructions of the central control module;

[0012] The robots adjust their positioning mechanisms based on the logical partitions and autonomous postures, and establish an optical communication network between robots within each logical partition for point-to-point data exchange.

[0013] Based on the partitioned bidirectional bubble merge sort algorithm, each logical partition is locally ordered; based on the fuzzy immune clonal selection algorithm, the minimum and maximum robots of each logical partition are determined, and boundary information for merging between logical partitions is generated; data is merged and boundaries are ordered between each logical partition;

[0014] Through the optical communication network, the boundary ordering completion signals and boundary status information of all logical partitions are broadcast globally according to the distributed negotiation model.

[0015] Furthermore, a coupled communication link is established between the robot integrated with the optical communication module and the central control module; based on the coupled communication link and according to the partition instruction of the central control module, the robot is logically partitioned, including:

[0016] Integrate an optical communication module into each robot and establish a coupled communication link with the central control module;

[0017] Using the central control module and the integrated optical communication module, the identity is sent to the robot in accordance with the coupling link communication protocol. The robot confirms the receipt after completing the communication.

[0018] The robot determines its own logical partition number, actively broadcasts the local partition identity through the optical communication self-discovery protocol, and listens to the broadcasts of other robots in the same partition to complete the two-way mapping of member identities within the logical partition and collaborative state synchronization.

[0019] Furthermore, the robot adjusts its positioning mechanism based on the logical partitions and autonomous posture, and establishes an optical communication network between the robots within each logical partition to conduct point-to-point data exchange, including:

[0020] The robot obtains its current position data from the local inertial measurement unit and formats its current position data;

[0021] The robot broadcasts its current position data to other robot nodes in the same logical partition through the optical communication module, and receives the spatial position information of all members in the same logical partition in parallel, and aggregates them to obtain a set of spatial distribution data;

[0022] Based on the spatially distributed data set, point-to-point link requirements, and the physical constraints of the optical communication module, the robot uses its autonomous posture adjustment optimization model to adjust its posture to the optimal physical orientation for the point-to-point link facing the partition members.

[0023] All robots in the partition complete the establishment of point-to-point optical links and link parameter configuration between all members based on negotiated parameters and physical postures, forming a complete point-to-point optical communication network within the logical partition.

[0024] Furthermore, based on the partitioned bidirectional bubble merge sort algorithm, each logical partition is locally sorted, including:

[0025] Each robot sends its own sort key value to all members in the same partition through a point-to-point optical communication network, and simultaneously receives and organizes the sort key values ​​of all members in the partition, forming the original data set required for partition ordering locally;

[0026] Based on the number of partition members and the distribution of sorting key values, each robot divides the partitioned original data set into several continuous segments through segment division locally and specifies the sorting direction;

[0027] Each robot member performs forward bubbling and reverse bubbling operations simultaneously according to the segment allocation: the forward segment realizes adjacent comparison and exchange from low to high, and the reverse segment realizes adjacent comparison and exchange from high to low;

[0028] Each robot member simultaneously advances the sorting process through multiple rounds of iterations until all local disordered pairs in all segments are eliminated;

[0029] The robots at the boundaries of each segment execute the merging process and gradually compare and exchange the data at the segment boundaries to achieve overall fusion and sorting between logical segments. At the same time, the segment boundaries are dynamically corrected until the local data of all robots in the entire area are arranged in order according to the target.

[0030] Furthermore, according to the fuzzy immune clonal selection algorithm, the minimum robot and the maximum robot of each logical partition are determined, and the boundary information for merging between logical partitions is generated, including:

[0031] The identifiers and values ​​of the first and last members in the broadcast area are sorted through the point-to-point optical communication network to form the candidate sets of robots with the minimum value and the maximum value in the partition; after all the first and last identifiers and key values ​​of the partition are collected in parallel by each logical partition member, the local boundary candidates are confirmed;

[0032] Initialize the fitness evaluation function, activity threshold and cloning factor parameters according to the fuzzy immune clonal selection algorithm;

[0033] The fitness of each candidate robot is evaluated based on the sort key value and the local order of the partition, and the candidate robot is assigned the corresponding fuzzy membership and activity intensity. Cloning, amplification and gene mutation operations are then performed to determine the minimum and maximum value robots.

[0034] Each logical partition will use the identities and sorting key values ​​of the final minimum and maximum value robots as partition boundary information for grouping and sorting;

[0035] Confirmation is broadcast to all robots in the partition through the point-to-point optical communication network to achieve consistent synchronization within the partition.

[0036] Furthermore, the fitness of each candidate robot is evaluated based on the sort key value and the local order of the partition, and the corresponding fuzzy membership and activity strength are assigned to the candidate robot. Cloning, amplification and gene mutation operations are performed to determine the minimum value robot and the maximum value robot. The following are included:

[0037] According to the distance between the sort key value and the lower limit of the interval and the order degree of the left neighbor of the candidate robot, the first fuzzy membership corresponding to the minimum robot candidate is constructed; based on the opposite principle, the second fuzzy membership corresponding to the maximum robot candidate is constructed;

[0038] Calculate the fuzzy fitness score of each candidate robot according to the first fuzzy membership and the second fuzzy membership; calculate the activity strength of each candidate robot according to the key value of the candidate robot;

[0039] Based on the fuzzy fitness score and activity intensity, immune cloning operations, gene mutation and activity correction are performed on each candidate robot;

[0040] Adopt the survival of the fittest strategy, sort by fitness score, and select the members with the highest fitness as the final minimum value robot and maximum value robot.

[0041] Furthermore, data merging and boundary ordering between each logical partition include:

[0042] Get the structured partition boundary robot nodes, including the boundary robot identifiers and sorted key value lists of each logical partition;

[0043] Compare the sort key values ​​of the boundary robots of two adjacent logical partitions one by one according to the partition number sequence, and construct a globally ordered merge and ordered instruction queue based on the comparison results;

[0044] Based on the globally ordered merge and ordered instruction queue, local ordering and global merging processing of the logical partition boundaries are performed.

[0045] Furthermore, through the optical communication network, the boundary ordering completion signal and boundary status information of all logical partitions are broadcasted globally according to the distributed negotiation model, including:

[0046] Integrate and encode the logic partition boundary ordering completion signal and boundary status information to form a message packet to be broadcast;

[0047] Schedule and arrange the sending sequence and broadcast path of the message packets to be broadcast according to the distributed negotiation model;

[0048] By utilizing the optical communication network, the message packets to be broadcasted in each logical partition are sent to all other partitions, thus achieving global data broadcast and message convergence at the physical channel level.

[0049] Furthermore, scheduling and arranging the sending sequence and broadcast path of the message packets to be broadcast according to the distributed negotiation model includes:

[0050] Determine the message broadcast order of each logical partition based on the distributed negotiation model, and assign the starting node, forwarding node and final destination node of each message packet to be broadcast to form a broadcast path allocation table;

[0051] According to the two-way confirmation mechanism set by the distributed negotiation model, an independent broadcast time slot is allocated to each message packet to be broadcast to ensure that there is no overlap and blockage in message forwarding on the same broadcast path.

[0052] According to another aspect of the present invention, a robot cluster optical network communication system based on optical communication data is also provided, including: a logical partitioning module for establishing a coupled communication link between a robot with an integrated optical communication module and a central control module; based on the coupled communication link and according to the partitioning instructions of the central control module, the robot is logically partitioned; a communication network establishment module for the robot to establish an optical communication network between robots within each logical partition according to the logical partitioning and autonomous posture adjustment positioning mechanism to perform point-to-point data exchange; a logical partition ordering and boundary merging module for locally ordering each logical partition based on a partitioned bidirectional bubble merge sorting algorithm; according to a fuzzy immune clonal selection algorithm, the minimum robot and the maximum robot of each logical partition are determined, and boundary information for merging between logical partitions is generated; data merging and boundary ordering are performed between each logical partition; a global broadcast module for globally broadcasting the boundary ordering completion signal and boundary status information of all logical partitions according to a distributed negotiation model through the optical communication network.

[0053] The beneficial effects of the present invention are:

[0054] (1) This invention combines a partitioned bidirectional bubble merge sort algorithm with a fuzzy immune clonal selection algorithm to achieve local ordering of robot clusters within logical partitions, automatically coordinate data merging and boundary information between partitions, and dynamically optimize the data ordering and coordination efficiency of the entire system in real time. This achieves efficient self-organizing sorting and boundary coordinated merging of robot clusters across multiple partitions.

[0055] (2) A point-to-point, self-adaptive distributed optical communication network built using integrated optical communication modules can effectively improve the data exchange rate and link reliability of high-density collaborative tasks between robots, and provide low-latency and high-bandwidth support for complex clusters that far exceeds traditional wireless radio frequency communications.

[0056] (3) Based on the distributed negotiation model (i.e., with a two-way confirmation mechanism), the ordered information of partition boundaries is efficiently broadcast across the entire network to achieve result synchronization and consistency confirmation between logical partitions, thereby eliminating signal conflicts and data redundancy in distributed sorting communication, improving reliability and data convergence efficiency, and supporting large-scale, decentralized multi-partition autonomy and intelligent merging. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 is a flow chart of a robot cluster optical network communication method based on optical communication data according to an embodiment of the present invention;

[0059] Figure 2 This is a principle block diagram of a robot cluster optical network communication system based on optical communication data according to an embodiment of the present invention.

[0060] In the picture:

[0061] 1. Logical partitioning module; 2. Communication network establishment module; 3. Logical partition ordering and boundary merging module; 4. Global broadcast module. DETAILED DESCRIPTION

[0062] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. By referring to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0063] According to an embodiment of the present invention, a robot cluster optical network communication method and system based on optical communication data are provided.

[0064] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, a robot cluster optical network communication method based on optical communication data is provided, comprising:

[0065] S1. Establish a coupled communication link between the robot integrated with the optical communication module and the central control module; based on the coupled communication link and according to the partitioning instruction of the central control module, logically partition the robot.

[0066] In a further embodiment, establishing a coupled communication link between the robot integrated with the optical communication module and the central control module; and logically partitioning the robot based on the coupled communication link and according to the partitioning instruction of the central control module includes:

[0067] An optical communication module is integrated inside each robot, and a coupled communication link is established with the central control module. The optical communication module includes a micro laser transmitter and receiver, an LED array, and a high-speed light receiving component. The central control module is used to send an identity to the robot through the integrated optical communication module in accordance with the coupled link communication protocol. The robot confirms the receipt after completing the communication. The robot determines its own logical partition number and actively broadcasts the local partition identity through the optical communication self-discovery protocol, and listens to the broadcasts of other robots in the same partition to complete the bidirectional mapping of member identities within the logical partition and collaborative state synchronization. Unique intra-partition communication parameters are allocated to each logical partition, point-to-point links within the partition are initialized, and identification and links are established between members of each logical partition.

[0068] Robots utilize an optical communication self-discovery protocol to proactively broadcast their own partition identities within their respective partitions, while simultaneously monitoring the identities broadcast by other members of the same partition. This enables bidirectional mapping of identities and collaborative state synchronization among robots within a partition, ensuring that each member is aware of the identities of all robots within the same logical partition. Coupled communication and partitioning mechanisms can significantly enhance the flexibility and collaborative efficiency of large-scale robot cluster management. For example, in unmanned warehousing, robots equipped with optical communication modules communicate with a central dispatch system via coupled links. Upon receiving partitioning instructions, they automatically complete grouping and link configuration, enabling each partition to form an independent and orderly information network. If robots A, B, and C belong to the same logical partition, after broadcasting and synchronizing their identities, they can maintain high-speed, low-interference data interaction through a dedicated optical communication channel, effectively executing tasks such as collaborative handling and path planning within the partition.

[0069] S2. The robot adjusts its positioning mechanism according to the logical partition and autonomous posture, and establishes an optical communication network between the robots within each logical partition to perform point-to-point data exchange.

[0070] In a further embodiment, the robots adjust their positioning mechanisms based on the logical partitions and autonomous postures, and establish an optical communication network between the robots within each logical partition for point-to-point data exchange, including:

[0071] The robot obtains its current position data from the local inertial measurement unit and formats its current position data; the robot broadcasts its current position data to other robot nodes in the same logical partition through the optical communication module, and in parallel receives the spatial position information of all members in the same logical partition, and summarizes it to obtain a spatial distribution data set; based on the spatial distribution data set, point-to-point link requirements and physical constraints of the optical communication module, the robot uses the robot's autonomous posture adjustment optimization model to adjust its posture to the optimal physical orientation facing the point-to-point link of the partition members; all robots in the partition complete the establishment of point-to-point optical links and link parameter configuration between all members based on the negotiated parameters and physical posture, and form a complete point-to-point optical communication network in the logical partition.

[0072] The inertial measurement unit (IMU) acquires its own position and attitude data in real time and formats this data in a standardized format. Based on this, the robot utilizes an integrated optical communication module to actively broadcast its own position information and simultaneously monitor the position information of other robots within the same logical partition. A link establishment method based on spatially distributed holographic perception and autonomous attitude optimization significantly improves the bandwidth efficiency and link reliability of point-to-point communication within the cluster. Compared to traditional wireless radio frequency communication methods, optical communication networks are less susceptible to interference and offer enhanced privacy. In dense multi-robot scenarios, they can flexibly avoid link conflicts and achieve high-quality simultaneous transmission of multiple signals. The autonomous attitude adjustment optimization model is an algorithm and decision-making mechanism for real-time calculation and adjustment of the robot's spatial orientation and attitude. It comprises a spatial perception unit (such as a positioning and inertial measurement module), a data exchange and neighborhood information extraction module, an attitude optimization decision algorithm (such as a heuristic optimization, graph theory algorithm, or reinforcement learning module), and a control interface for mechanical actuators (such as servos and gear trains). This completes a closed perception-decision-execution loop, enabling efficient communication and attitude coordination within a distributed robot cluster.

[0073] For example, in an automated rack handling system, handling robots belonging to a specific logical partition are distributed across an area of ​​several square meters. While operating, each robot relies on IMU positioning to determine its own spatial coordinates and exchange coordinate information with each other. Based on the real-time collected overall position distribution data, each robot automatically optimizes its orientation, accurately pointing its optical communication module toward the nearest other robot in the same partition, achieving multiple efficient point-to-point optical links. When a new member joins a partition or a robot moves, the network senses the node changes, and all members automatically replan their postures and links, continuously maintaining high-speed information interoperability and full coverage within the entire partition.

[0074] S3. Based on the partitioned bidirectional bubble merge sort algorithm, each logical partition is locally ordered; according to the fuzzy immune clonal selection algorithm, the minimum robot and maximum robot of each logical partition are determined, and the boundary information for merging between logical partitions is generated; data is merged and the boundaries are ordered between each logical partition.

[0075] In a further embodiment, locally sorting each logical partition based on a partitioned bidirectional bubble merge sort algorithm includes:

[0076] Each robot sends its own sorting key value to all members in the same partition through a point-to-point optical communication network, and receives and organizes the sorting key values ​​of all members in the partition in parallel, forming the original data set required for partition ordering locally; based on the number of partition members and the distribution of sorting key values, each robot divides the partition original data set into several continuous segments through segment division locally, and specifies the sorting direction; each robot member simultaneously performs forward bubbling and reverse bubbling operations based on segment allocation: the forward segment realizes adjacent comparison and exchange from low to high, and the reverse segment realizes adjacent comparison and exchange from high to low; each robot member synchronously advances the sorting process through multiple rounds of iteration until all local disordered pairs in all segments are completely eliminated; the robots at the boundaries of each segment execute the merging process, and gradually compare and exchange the data at the boundary of the segment to achieve overall fusion and sorting between logical segments, and dynamically correct the segment boundaries until the local data of all robots in the entire area are arranged in order according to the target.

[0077] The partitioned bidirectional bubble merge sort algorithm is a distributed sorting method performed collaboratively by all robots within each logical partition. First, each robot broadcasts its local sort key (such as number, task priority, and location information) to all members in the same partition via a point-to-point optical communication network, and simultaneously receives the key values ​​from other members. Ultimately, the original data set containing all sort keys is locally aggregated. Subsequently, each robot locally divides this data set into several contiguous segments based on the number of members and the distribution of key values, and specifies a sorting direction for each segment (for example, forward bubbling for odd-numbered segments and reverse bubbling for even-numbered segments). The partitioned bidirectional bubble merge sort algorithm leverages the advantages of distributed parallel processing to improve sorting speed and accuracy. By implementing synchronous bubbling within segments and iterative merging between segments, it avoids data silos, conflicts, and the extreme time consumption inherent in traditional collective sorting. Furthermore, point-to-point optical communication enables fast and low-latency data exchange during the sorting process, allowing each robot to promptly detect changes in the sorting of all members, significantly accelerating the convergence of local ordering across the entire partition. The dynamic boundary adjustment in the merging stage can adapt to the changes in the number of members in the partition or the emergence of outliers, and improve the robustness of the algorithm to abnormal distribution and dynamic changes under complex actual working conditions.

[0078] For example, within a logistics warehouse zone, eight robots are sorted by task priority. The robots first broadcast their priorities to each other, receiving the priority data of all members. Each robot then locally divides its task into four contiguous segments based on the eight values, designating segments 1 and 3 as forward bubbling and segments 2 and 4 as reverse bubbling. Within each segment, robots exchange task orders with adjacent members within their assigned segments. After multiple rounds of local bubbling, order is essentially achieved within each segment. Next, boundary robots perform inter-segment merging and comparison, adjusting their data based on the comparison results. This gradually promotes global ordering of the entire zone's data. Ultimately, all robots' task priorities are arranged in the target order, allowing scheduling instructions to be issued directly and sequentially, improving cluster responsiveness and collaboration efficiency.

[0079] In a further embodiment, determining the minimum robot and the maximum robot of each logical partition according to the fuzzy immune clonal selection algorithm and generating boundary information for merging between logical partitions includes:

[0080] The identifiers and values ​​of the first and last members of the sorting area are broadcasted through the point-to-point optical communication network to form a candidate set of the minimum robot and the maximum robot in the partition; after each logical partition member collects the head and tail identifiers and key values ​​of all the partitions in parallel, the local boundary candidates are confirmed; according to the fuzzy immune clone selection algorithm, the fitness evaluation function, activity threshold and cloning factor parameters are initialized; the fitness of each candidate robot is evaluated according to the sorting key value and the local order of the partition, and the candidate robot is assigned the corresponding fuzzy membership and activity strength, and clone amplification and gene mutation operations are performed to determine the minimum robot and the maximum robot; each logical partition will use the identity and sorting key value of the final minimum robot and maximum robot as the partition boundary information for packaging and sorting; the confirmation is broadcast to all robots in the partition through the point-to-point optical communication network to achieve consistency synchronization within the partition.

[0081] In a further embodiment, the fitness of each candidate robot is evaluated based on the sort key value and the partition local order, the candidate robot is assigned a corresponding fuzzy membership and activity strength, and cloning, amplification and gene mutation operations are performed to determine the minimum value robot and the maximum value robot. The following examples are included:

[0082] According to the distance of the sorting key value from the lower limit of the interval and the order degree of the left neighbor of the candidate robot, the first fuzzy membership corresponding to the minimum robot candidate is constructed; based on the opposite principle, the second fuzzy membership corresponding to the maximum robot candidate is constructed; according to the first fuzzy membership and the second fuzzy membership, the fuzzy fitness score of each candidate robot is calculated; according to the key value of the candidate robot, the activity intensity of each candidate robot is calculated; based on the fuzzy fitness score and activity intensity, immune cloning operation, gene mutation and activity correction are performed on each candidate robot; adopt the survival of the fittest strategy, sort by fitness score, and select the members with the highest fitness as the final minimum robot and maximum robot.

[0083] Among them, the candidate for the minimum value robot should be as close to the lower limit of the partition as possible and have a high degree of order in the left neighbor:

[0084] ;

[0085] The candidate for the maximum value robot should be as close to the partition upper limit as possible and have a high right neighbor order:

[0086]

[0087] Where, u min,i Indicates the i The first fuzzy membership of the robot as the candidate for the minimum value of the partition, u max,i Indicates the iThe second fuzzy membership of the robot as the partition maximum candidate; x i Indicates the i The sort key value of the robot; x min Indicates the lower bound of all robot sort key values ​​in the current partition. x max Indicates the upper limit of all robot sort key values ​​in the current partition; a min represents the parameter that controls the sensitivity of the first fuzzy membership, a max Represents the parameters that control the sensitivity of the second fuzzy membership. These two parameters are used to adjust the penalty intensity for distance and are set empirically. S l,i Indicates the i The left neighbor order of the robot, S r,i Indicates the i The order degree of the right neighbor of the robot. When the order degree is equal to 1, it is completely ordered.

[0088] In a further embodiment, performing data merging and boundary ordering between each logical partition includes:

[0089] Obtain structured partition boundary robot nodes, including the boundary robot identifiers and sorting key value lists of each logical partition (i.e., boundary information used for merging between logical partitions); compare the sorting key values ​​of the boundary robots of two adjacent logical partitions one by one in the order of partition numbers to determine whether there is overlap, discreteness, or misordering of the boundaries, and construct a globally ordered merging and ordering instruction queue based on the comparison results; perform local ordering and global merging processing of the logical partition boundaries based on the globally ordered merging and ordering instruction queues.

[0090] The fuzzy immune clonal selection algorithm improves on the immune clonal selection algorithm by applying fuzzy mathematical theory. Its structure involves integrating member identifiers and key values ​​to form a candidate population. Fuzzy logic is used to design an evaluation function, defining membership based on parameters such as the robot's sorting key and local order, thereby determining quality. Excellent candidates are replicated and amplified at a certain ratio, with the number of clones related to fitness. To avoid falling into local optima, the clones undergo moderate perturbations of key attributes, such as key values. Based on fitness, the optimal candidate is selected from the parent and child sets, and the optimal boundary member (minimum / maximum value robot) is partitioned.

[0091] The fuzzy immune cloning mechanism comprehensively measures the boundary value of each candidate robot and is resistant to abnormal fluctuations or intermediate distributions, robustly selecting the most appropriate boundary nodes. This improves the accuracy of partition boundary identification and segmentation, avoiding limitations of traditional algorithms such as misjudgment of local extreme values ​​or boundary jumps. For example, for five robots with inventory numbers {3, 5, 7, 8, 10}, when evaluating boundary robots, the fuzzy membership function is prioritized to determine the minimum value for numbers 3 and 8 (lower bound 3, no left neighbor) and the maximum value for numbers 10 and 7 (upper bound 10, right neighbor 8). Combined with the local order and activity strength of previous partitions, cloning and mutation simulations ultimately select robots 3 and 10 as the minimum / maximum value. When merging multiple partitions, the maximum value of the adjacent partition is automatically compared with the minimum value of the next partition. If a misordering is detected (maximum value 10 in partition A, minimum value 9 in partition B), the merge queue is adjusted to restore the data to a globally ascending distribution, achieving seamless fusion and ordering of logically partitioned data.

[0092] S4. Through the optical communication network, the boundary ordering completion signals and boundary status information of all logical partitions are broadcast globally according to the distributed negotiation model.

[0093] In a further embodiment, broadcasting the boundary ordering completion signal and boundary status information of all logical partitions globally according to the distributed negotiation model through the optical communication network includes:

[0094] The logical partition boundary order completion signal and boundary status information are integrated and encoded to form a structured and holistically parsable broadcast message packet. According to the distributed negotiation model, the sending timing and broadcast path of the broadcast message packet are scheduled and arranged. Using the optical communication network, the broadcast message packet of each logical partition is sent to all other partitions, realizing global data broadcast and message convergence at the physical channel level.

[0095] In a further embodiment, scheduling and arranging the sending timing and broadcast path of the message packets to be broadcast according to the distributed negotiation model includes:

[0096] The message broadcast order of each logical partition is determined according to the distributed negotiation model, and the starting node, forwarding node and final destination node of each message packet to be broadcast are allocated to form a broadcast path allocation table; according to the two-way confirmation mechanism set by the distributed negotiation model, an independent broadcast time slot is allocated for each message packet to be broadcast to ensure that there is no overlap and blockage in message forwarding on the same broadcast path.

[0097] Among them, the distributed negotiation model (such as the two-way confirmation mechanism in the distributed consensus algorithm) determines the timing and path of the broadcast, assigns the initial transmission, forwarding, and destination nodes of message packets in each partition, and forms a broadcast path that avoids conflicts and duplications. The two-way confirmation mechanism in the distributed consensus algorithm requires not only a request / push (forward message) when data or instructions are transmitted from the source node to the target node, but also a confirmation / response (reverse confirmation) after the other party successfully receives and processes it. This includes: the source node initiates the data message that needs to be broadcast or synchronized based on the consistency requirements and pushes it to the target node; after receiving the message, the target node performs preprocessing tasks such as content verification, persistence, or local status update; after completing processing, the target node actively generates a confirmation message and sends it back to the source node to notify it that the message has been safely received and adopted; if the forward message is not confirmed, the source node automatically resends it after a set timeout period, ensuring that the message is not lost even in the event of communication anomalies.

[0098] The broadcast mechanism, combining distributed negotiation with optical communications, effectively overcomes the congestion, disorder, and slow convergence issues inherent in traditional cluster broadcasts. Structured message packets ensure automatic decoding and rapid distribution of status information; distributed scheduling avoids message duplication, deadlock, and packet loss, improving the real-time and reliability of message consistency; and a two-way confirmation mechanism ensures feedback confirmation of message delivery on each link, further strengthening global synchronization.

[0099] For example, in a robot cluster with four logical partitions, after each completes ordering its boundary state, partition A encodes its completion signal and the current boundary state into a message packet. Through a negotiation model, partition A is designated as the first broadcast node. According to the table, partition A sends the message to the designated forwarding nodes of B and C. After receiving and confirming the message, partition B forwards it to D, which then sends feedback back to partition A for bidirectional confirmation. Throughout this process, messages on each optical communication link between partitions are transmitted in independent time slots, without overlap or blocking, ensuring that all partitions receive global boundary state information synchronously.

[0100] In addition, the present invention builds an optical communication mesh network with self-organizing capabilities based on the optical links between robots. It combines the robot's motion trajectory with task scheduling information to dynamically adjust the communication topology in real time. It utilizes the large bandwidth characteristics of optical communication to achieve high-speed transmission of task information, perception data, and synchronization control instructions between robots. By integrating optical communication modules into humanoid robots, an efficient optical link mesh network is constructed, achieving stable, high-speed, low-latency data exchange and collaborative control between robot clusters, significantly improving the cluster's response speed and collaborative efficiency. Optical network system equipment replaces current 5G, broadband, WIFI, etc., and data transmission is more stable, faster, and has higher bandwidth.

[0101] like Figure 2According to another embodiment of the present invention, a robot cluster optical network communication system based on optical communication data is provided, comprising:

[0102] The logical partitioning module 1 is used to establish a coupled communication link between the robot with integrated optical communication module and the central control module; based on the coupled communication link and according to the partitioning instructions of the central control module, the robot is logically partitioned.

[0103] The communication network establishment module 2 is used for the robot to adjust the positioning mechanism according to the logical partition and autonomous posture, and to establish an optical communication network between the robots within each logical partition for point-to-point data exchange.

[0104] The logical partition ordering and boundary merging module 3 is used to locally order each logical partition based on the partition bidirectional bubble merge sort algorithm; determine the minimum robot and maximum robot of each logical partition according to the fuzzy immune clone selection algorithm, and generate boundary information for merging between logical partitions; perform data merging and boundary ordering between each logical partition.

[0105] The global broadcast module 4 is used to broadcast the boundary ordering completion signals and boundary status information of all logical partitions globally according to the distributed negotiation model through the optical communication network.

[0106] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A robot cluster optical network communication method based on optical communication data, characterized in that: include: Establishing a coupled communication link between the robot integrated with the optical communication module and the central control module; logically partitioning the robot based on the coupled communication link and according to the partitioning instructions of the central control module; The robots adjust their positioning mechanisms based on the logical partitions and autonomous postures, and establish an optical communication network between robots within each logical partition for point-to-point data exchange. Based on the partitioned bidirectional bubble merge sort algorithm, each logical partition is locally ordered; based on the fuzzy immune clonal selection algorithm, the minimum and maximum robots of each logical partition are determined, and boundary information for merging between logical partitions is generated; data is merged and boundaries are ordered between each logical partition; Through the optical communication network, the boundary ordering completion signals and boundary status information of all logical partitions are broadcast globally according to the distributed negotiation model.

2. The method for robot cluster optical network communication based on optical communication data according to claim 1, characterized in that: The step of establishing a coupled communication link between the robot integrated with the optical communication module and the central control module; and logically partitioning the robot based on the coupled communication link and according to the partitioning instruction of the central control module includes: Integrate an optical communication module into each robot and establish a coupled communication link with the central control module; Using the central control module and the integrated optical communication module, the identity is sent to the robot in accordance with the coupling link communication protocol. The robot confirms the receipt after completing the communication. The robot determines its own logical partition number, actively broadcasts the local partition identity through the optical communication self-discovery protocol, and listens to the broadcasts of other robots in the same partition to complete the two-way mapping of member identities within the logical partition and collaborative state synchronization.

3. The method for robot cluster optical network communication based on optical communication data according to claim 1, characterized in that: The robots adjust their positioning mechanisms according to the logical partitions and autonomous postures, and establish an optical communication network between the robots within each logical partition to perform point-to-point data exchange, including: The robot obtains its current position data from the local inertial measurement unit and formats its current position data; The robot broadcasts its current position data to other robot nodes in the same logical partition through the optical communication module, and receives the spatial position information of all members in the same logical partition in parallel, and aggregates them to obtain a set of spatial distribution data; Based on the spatially distributed data set, point-to-point link requirements, and the physical constraints of the optical communication module, the robot uses its autonomous posture adjustment optimization model to adjust its posture to the optimal physical orientation for the point-to-point link facing the partition members. All robots in the partition complete the establishment of point-to-point optical links and link parameter configuration between all members based on negotiated parameters and physical postures, forming a complete point-to-point optical communication network within the logical partition.

4. The method for robot cluster optical network communication based on optical communication data according to claim 1, characterized in that: The partition-based bidirectional bubble merge sort algorithm locally sorts each logical partition, including: Each robot sends its own sort key value to all members in the same partition through a point-to-point optical communication network, and simultaneously receives and organizes the sort key values ​​of all members in the partition, forming the original data set required for partition ordering locally; Based on the number of partition members and the distribution of sorting key values, each robot divides the partitioned original data set into several continuous segments through segment division locally and specifies the sorting direction; Each robot member performs forward bubbling and reverse bubbling operations simultaneously according to the segment allocation: the forward segment realizes adjacent comparison and exchange from low to high, and the reverse segment realizes adjacent comparison and exchange from high to low; Each robot member simultaneously advances the sorting process through multiple rounds of iterations until all local disordered pairs in all segments are eliminated; The robots at the boundaries of each segment execute the merging process and gradually compare and exchange the data at the segment boundaries to achieve overall fusion and sorting between logical segments. At the same time, the segment boundaries are dynamically corrected until the local data of all robots in the entire area are arranged in order according to the target.

5. The method for robot cluster optical network communication based on optical communication data according to claim 1, characterized in that: The method of determining the minimum robot and the maximum robot of each logical partition according to the fuzzy immune clonal selection algorithm and generating boundary information for merging between logical partitions includes: The identifiers and values ​​of the first and last members in the broadcast area are sorted through the point-to-point optical communication network to form the candidate sets of robots with the minimum value and the maximum value in the partition; after all the first and last identifiers and key values ​​of the partition are collected in parallel by each logical partition member, the local boundary candidates are confirmed; Initialize the fitness evaluation function, activity threshold and cloning factor parameters according to the fuzzy immune clonal selection algorithm; The fitness of each candidate robot is evaluated based on the sort key value and the local order of the partition, and the candidate robot is assigned the corresponding fuzzy membership and activity intensity. Cloning, amplification and gene mutation operations are then performed to determine the minimum and maximum value robots. Each logical partition will use the identities and sorting key values ​​of the final minimum and maximum value robots as partition boundary information for grouping and sorting; Confirmation is broadcast to all robots in the partition through the point-to-point optical communication network to achieve consistent synchronization within the partition.

6. The method for robot cluster optical network communication based on optical communication data according to claim 5, characterized in that: The fitness evaluation of each candidate robot is performed based on the sort key value and the partition local order, the corresponding fuzzy membership and activity intensity are assigned to the candidate robot, and cloning, amplification and gene mutation operations are performed to determine the minimum value robot and the maximum value robot. The following are included: According to the distance between the sort key value and the lower limit of the interval and the order degree of the left neighbor of the candidate robot, the first fuzzy membership corresponding to the minimum robot candidate is constructed; based on the opposite principle, the second fuzzy membership corresponding to the maximum robot candidate is constructed; Calculate the fuzzy fitness score of each candidate robot according to the first fuzzy membership and the second fuzzy membership; calculate the activity strength of each candidate robot according to the key value of the candidate robot; Based on the fuzzy fitness score and activity intensity, immune cloning operations, gene mutation and activity correction are performed on each candidate robot; Adopt the survival of the fittest strategy, sort by fitness score, and select the members with the highest fitness as the final minimum value robot and maximum value robot.

7. The method for robot cluster optical network communication based on optical communication data according to claim 1, characterized in that: The data merging and boundary ordering between each logical partition includes: Get the structured partition boundary robot nodes, including the boundary robot identifiers and sorted key value lists of each logical partition; Compare the sort key values ​​of the boundary robots of two adjacent logical partitions one by one according to the partition number sequence, and construct a globally ordered merge and ordered instruction queue based on the comparison results; Based on the globally ordered merge and ordered instruction queue, local ordering and global merging processing of the logical partition boundaries are performed.

8. The method for robot cluster optical network communication based on optical communication data according to claim 1, characterized in that: The method of broadcasting the boundary ordering completion signal and boundary status information of all logical partitions to the entire area according to the distributed negotiation model through the optical communication network includes: Integrate and encode the logic partition boundary ordering completion signal and boundary status information to form a message packet to be broadcast; Schedule and arrange the sending sequence and broadcast path of the message packets to be broadcast according to the distributed negotiation model; By utilizing the optical communication network, the message packets to be broadcasted in each logical partition are sent to all other partitions, thus achieving global data broadcast and message convergence at the physical channel level.

9. The method for robot cluster optical network communication based on optical communication data according to claim 8, characterized in that: The scheduling and arranging of the sending timing and broadcast path of the message packets to be broadcast according to the distributed negotiation model includes: Determine the message broadcast order of each logical partition based on the distributed negotiation model, and assign the starting node, forwarding node and final destination node of each message packet to be broadcast to form a broadcast path allocation table; According to the two-way confirmation mechanism set by the distributed negotiation model, an independent broadcast time slot is allocated to each message packet to be broadcast to ensure that there is no overlap and blockage in message forwarding on the same broadcast path.

10. A robot cluster optical network communication system based on optical communication data, used to implement the robot cluster optical network communication method based on optical communication data according to any one of claims 1 to 9, characterized in that: It includes a logical partitioning module, a communication network establishment module, a logical partition ordering and boundary merging module, and a global broadcast module: The logic partitioning module is used to establish a coupled communication link between the robot integrated with the optical communication module and the central control module; based on the coupled communication link and according to the partitioning instructions of the central control module, the robot is logically partitioned; The communication network establishment module is used for the robot to adjust the positioning mechanism according to the logical partition and autonomous posture, and establish an optical communication network between the robots within each logical partition to perform point-to-point data exchange; The logic partition ordering and boundary merging module is used to locally order each logic partition based on the partition bidirectional bubble merge sort algorithm; determine the minimum robot and maximum robot of each logic partition based on the fuzzy immune clonal selection algorithm, and generate boundary information for merging between logic partitions; and perform data merging and boundary ordering between each logic partition; The global broadcast module is used to broadcast the boundary ordering completion signals and boundary status information of all logical partitions globally according to the distributed negotiation model through the optical communication network.

Images