Multi-RGV Internet of Things method and system based on optical communication

Abstract

The invention discloses a multi-RGV Internet of Things method and system based on optical communication, and relates to the technical field of multi-device cooperative control and optical communication in an industrial automation logistics system, and the method comprises the following steps: S1, collecting the position, motion state and environment obstacle information in real time through a multidirectional tunable optical transceiver module carried by each RGV, and generating a dynamic optical relay priority weight. According to the multi-RGV Internet-of-things method and system based on optical communication, through deep fusion of a dynamic optical relay network and an adaptive wavelength allocation technology, the communication stability and cooperation efficiency of multiple RGVs in a complex industrial scene are significantly improved; based on a node selection mechanism of a dynamic optical relay priority weight, in combination with a mixed integer programming model of an elastic channel, uninterrupted transmission and efficient channel multiplexing of an optical communication link in a mobile shielding scene are realized.

Description

Technical Field

[0001] The present invention relates to the field of multi-device collaborative control and optical communication technology in industrial automation logistics systems, and specifically to a multi-RGV Internet of Things method and system based on optical communication. Background Art

[0002] In industrial automation scenarios, the coordinated scheduling of multiple rail-guided vehicles (RGVs) is a core requirement for efficient logistics delivery. With the introduction of optical communication technology into RGV IoT systems due to its strong resistance to electromagnetic interference and high bandwidth, it replaces traditional radio frequency or infrared communications with optical signals, significantly improving data transmission speed and stability. However, existing RGV IoT solutions based on optical communication face significant challenges in dynamic operations. Because RGVs must maneuver, avoid obstacles, and adjust their paths in real time within complex environments, traditional optical communication, which relies on fixed nodes or pre-set linear transmission paths, struggles to adapt to dynamic topological changes, leading to frequent communication link interruptions. For example, when multiple RGVs intersect in narrow corridors, the vehicles or environmental obstacles can easily block the linear optical path, causing signal loss. Furthermore, the real-time changes in RGV positions create blind spots in the coverage of fixed optical nodes, preventing timely reconfiguration of the communication network and causing some devices to lose control of their dispatch instructions. Furthermore, existing technologies lack a flexible mechanism for allocating flexible channel resources. Concurrent data transmission by multiple RGVs can easily lead to channel contention conflicts, further exacerbating communication delays. These issues significantly reduce the system's real-time collaboration efficiency in dynamic scenarios, and may even lead to mission stalls or device collisions. Although existing patents have optimized optical node layout or single-device communication performance for static scenarios, none have overcome the bottlenecks of adaptive networking and anti-obstruction transmission brought about by mobility. Summary of the Invention

[0003] (1) Technical problems solved In response to the shortcomings of the existing technology, the present invention provides a method and system for interconnecting multiple RGVs based on optical communication, which solves the problem of how to ensure the continuous stability and efficient coordination of optical communication links in the dynamic movement scenario of multiple RGVs.

[0004] (2) Technical solution To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for connecting multiple RGVs based on optical communication, comprising the following steps: S1: The multi-directional tunable optical transceiver module on each RGV collects real-time location, motion status, and environmental obstacle information to generate dynamic optical relay priority weights; During implementation, the multi-directional tunable optical transceiver module installed on each RGV consists of at least three optical signal transceiver units evenly distributed in the horizontal plane. Each unit covers an angle of no less than 240 degrees. It uses laser ranging, inertial navigation, and environmental obstacle detection sensors to obtain its own position coordinates, movement speed, and direction data in real time, while also scanning the outline information of surrounding obstacles. S2: Based on the dynamic optical relay priority weight, adjacent RGVs are selected as relay nodes to build a distributed dynamic optical relay network, and link coverage is achieved through optical signal relay transmission; During implementation, when the signal attenuation of the primary communication link exceeds a preset threshold due to RGV movement or environmental obstacles, the neighboring RGV nodes are screened according to the dynamic optical relay priority weight list; S3: Divide the optical wavelength resources into basic channels and flexible channels, assign fixed wavelength transmission control instructions to the basic channels, and dynamically assign wavelength transmission data streams to the flexible channels based on mixed integer programming; During implementation, optical wavelength resources are divided into two parts: basic channels and flexible channels based on preset spectrum segmentation rules. The basic channels use a fixed wavelength allocation strategy, allocating a unique and non-overlapping fixed wavelength to each RGV for transmitting scheduling instructions and emergency shutdown signals. This low-latency, high-priority control information ensures the real-time and conflict-free transmission of critical instructions. The flexible channels retain a dynamic wavelength pool. Wavelength resources are not pre-bound to devices, but are dynamically allocated based on the real-time network topology and task requirements through a mixed integer programming model. S4: Identify potential channel contention through conflict prediction mechanisms and dynamically adjust wavelengths or switch transmission paths; During implementation, the conflict prediction mechanism is implemented through the following steps: First, based on the optical signal propagation delay calculation model, combined with the real-time position, speed and movement direction data of each RGV, the overlapping range of the communication areas of each RGV in the future time period is predicted; the Kalman filter algorithm is used to model the RGV movement trajectory and infer its expected position in the future time window. Combined with the elastic channel wavelength allocation table, it is determined whether there is a risk of multiple RGVs in the same area using the same wavelength to transmit data; if a potential conflict is detected, a preemption release instruction is generated based on the task urgency and the current channel load status; S5: Dual optical paths are used for asynchronous redundant transmission of critical data. The receiving end verifies and merges data packets based on timestamps and topology fingerprints. During the implementation process, key data packets are copied into two copies and transmitted through two independent relay paths. That is, the sender selects a non-overlapping sequence of relay nodes for each path based on the dynamic optical relay priority weight list to ensure that the two paths do not have common nodes or parallel optical paths in the physical topology. Each data packet is attached with a timestamp and path topology fingerprint information accurate to the microsecond level. That is, the timestamp is generated synchronously by the sender's atomic clock, and the topology fingerprint is encrypted and encoded by the ID sequence of the relay nodes passed through the route through a hash algorithm. After asynchronously receiving the data packets from the two paths within the preset time window, the receiver first extracts the timestamp and calculates the transmission delay difference. If the delay difference exceeds the fault tolerance threshold, it is determined to be a path anomaly and an alarm is triggered. The topology fingerprint is then decrypted and verified to match the preset relay path node sequence. If there is a missing node ID or the sequence is tampered with, it is marked as an abnormal data packet. The data packets that pass the verification enter the fusion stage, adopting a strategy based on timestamp priority. S6: Real-time monitoring of optical signal intensity fluctuation trends, determining the type of obstruction and triggering path switching or topology reconstruction; During the implementation process, the system collects optical signal strength data in real time at a sampling frequency of thousands of times per second through a multi-directional tunable optical transceiver module, and calculates the short-term fluctuation trend and long-term attenuation slope of the signal strength based on the sliding window algorithm; when it is detected that the signal strength fluctuates violently in the short-term window, for example, the intensity drop rate exceeds 50dB / ms, but the mean value in the long-term window is stable, it is determined to be a transient obstruction, and the path switching mechanism is immediately triggered, namely: the suboptimal node is selected as the backup path from the dynamic optical relay priority weight list, the transmission direction and wavelength of the multi-directional tunable optical transceiver module are adjusted, and the data stream is seamlessly switched to the new path, while the original path monitoring is retained until the signal is restored; if the signal strength is continuously lower than the preset threshold of 70dBm in the long-term window and the attenuation slope approaches zero, it is determined to be a permanent obstruction, and the local topology reconstruction is initiated. Reconstruction process: First, freeze the current relay link, recalculate the path redundancy model based on the real-time updated RGV position information, generate a new relay node candidate list, and allocate elastic channel wavelengths to the affected data streams based on the mixed integer programming model; a progressive switching strategy is adopted during the reconstruction process, first establishing a new path and verifying its stability, and then releasing the old path resources to avoid communication interruption during the switching; in addition, the system matches historical obstruction scenarios through the light intensity fluctuation pattern library. If a known instantaneous obstruction pattern is matched, the backup path parameters are preloaded in advance to shorten the switching response time to less than 5 milliseconds; after completing the switching or reconstruction, the system continuously monitors the signal strength stability of the new path. If the same type of obstruction alarm is triggered again within 10 seconds, the area is marked as a high-risk communication blind spot, and global topology optimization is triggered to avoid the area.

[0005] Preferably, generating the dynamic optical relay priority weight includes: Based on the real-time location, signal strength, path redundancy, and load balancing of RGVs, the priority ranking of relay nodes is determined through weighted calculation, and the optimal relay path is selected; During the implementation process, the generation of dynamic optical relay priority weights is achieved through the following steps: each RGV obtains the optical signal strength data of the neighboring nodes in real time through the multi-directional tunable optical transceiver module, and combines its own coordinates and motion vectors output by the laser ranging and inertial navigation module to build a local topology map centered on the current RGV; the calculation of path redundancy is based on the relay hop count and path redundancy model of the candidate nodes in the topology map, where the hop count is obtained by solving the shortest path through the Dijkstra algorithm, and the link stability is dynamically corrected according to the historical communication packet loss rate and delay fluctuation data; the load balancing is evaluated by monitoring the number of concurrent data streams and remaining bandwidth resources of each candidate node and processing delay, quantifying its available communication capacity and converting it into a standardized score of 0-1; in the weighted calculation stage, the signal strength, path redundancy and load balancing are linearly superimposed according to the preset coefficients to generate a priority score list, and sorted from high to low according to the score; when screening the optimal relay path, the system gives priority to the node with the highest score, and at the same time verifies whether its current load is lower than the capacity threshold and whether the number of path hops is less than the maximum allowed value. If the main selected node does not meet the constraints, the suboptimal node is automatically selected in descending order; in addition, when a change in network topology is detected, the weight recalculation process is immediately triggered, the priority list is updated and synchronized to all related nodes to ensure that the relay decision is adapted to the dynamic environment in real time.

[0006] Preferably, the dynamic allocation of the elastic channel includes: Based on the relative positions between RGVs, task urgency, and relay link load status, a mixed integer programming model is constructed to calculate the wavelength allocation plan in real time. The elastic channel wavelength pool is dynamically adjusted as the network topology changes to maximize channel utilization; During the implementation process, the dynamic allocation of elastic channels is achieved through the following steps: the system constructs the objective function and constraints of the mixed integer programming model based on the real-time collected relative position coordinates between RGVs, task type identification and relay link load rate data, namely: the objective function is set to maximize channel utilization, and the constraints include wavelength uniqueness constraints in the same communication area, the maximum number of concurrent data streams per node, and the task urgency weight factor; after the model input data is preprocessed, the branch and bound algorithm is used to solve it and output the optimal wavelength allocation plan, including the elastic channel wavelength value and effective duration corresponding to each data stream; during the allocation process, the system reserves a wavelength switching buffer for high-priority tasks to ensure that urgent tasks can quickly switch wavelengths without interrupting transmission when the topology changes; in the dynamic adjustment stage, the network topology status is scanned every 100 milliseconds. When it is detected that the RGV movement causes When the overlap of the communication area changes by more than 15% or a new data transmission request is added, the model recalculation is triggered and the available resource list of the wavelength pool is updated according to the latest topology map. At the same time, a 10nm spectrum protection interval is set between the elastic channel wavelength and the basic channel wavelength, and adjacent channel interference is eliminated through a bandpass filter. For sudden wavelength conflicts, the system monitors the occupancy status of each wavelength in real time and predicts conflicts for RGV data streams that are about to enter the same area: if it is detected that two data streams will use the same wavelength to enter the overlapping area within the next 30 milliseconds, the data stream with lower priority will be forced to perform wavelength switching and rerouted to the idle wavelength of the adjacent relay node. Before switching, the spectrum isolation between the new wavelength and the basic channel of the target receiving end will be verified. After the allocation is completed, the system records the usage history data of each wavelength to optimize the load balancing weight coefficient in the subsequent model, forming a closed-loop optimization mechanism.

[0007] Preferably, the conflict prediction mechanism includes: Predict potential channel contention based on optical signal propagation delay and RGV motion trajectory; When a conflict risk is detected, a preemption release instruction is triggered to switch wavelengths or release channel resources; During the implementation process, the realization of the conflict prediction mechanism includes the following steps: the system obtains the signal transmission rate corresponding to the current channel wavelength through the optical signal propagation delay calculation module, combines the real-time coordinates, velocity vector and movement direction of each RGV, and uses the extended Kalman filter algorithm to predict the communication area boundary of each RGV in the next 50 milliseconds; based on the elastic channel wavelength allocation table and the predicted position data, a conflict probability matrix is ​​constructed to identify the communication areas and corresponding time windows where wavelength overlap may occur, that is: if two or more RGV data streams will use the same wavelength in the same area and at the same time, they are marked as high-risk conflict events; for detected conflict events, the system generates differentiated processing strategies based on the task urgency label: high-priority tasks retain the original wavelength and activate the path switching mechanism, forwarding their data streams through the idle wavelength of the adjacent relay node, and adjusting the receiving filter parameters of the target node to match the new wavelength; low-priority tasks retain the original wavelength and activate the path switching mechanism, forwarding their data streams through the idle wavelength of the adjacent relay node, and adjusting the receiving filter parameters of the target node to match the new wavelength; The task triggers the preemption release instruction, forcibly releasing the current wavelength and reallocating it to an available wavelength outside the spectrum protection interval. Before allocation, the spectrum isolation between the new wavelength and the wavelength in the adjacent area is verified through the simulation module; during the path switching process, if a multi-hop relay link is involved, the system synchronously corrects the signal forwarding timing of the relay node. The process includes: calculating the transmission delay compensation value based on the number of newly added hops and injecting it into the timestamp field of the data packet header to ensure the timing synchronization of the receiving end; after completing the wavelength or path adjustment, the system tracks the conflict resolution status through the real-time monitoring module. If wavelength overlap is still detected within 20 milliseconds after the adjustment, the local topology reconstruction process is initiated, the elastic channel allocation plan for the affected area is recalculated, and the data packet retransmission mechanism is triggered to repair possible lost information; in addition, the system records the time, location and processing results of historical conflict events to optimize the motion trajectory prediction accuracy of the Kalman filter algorithm, forming a conflict prediction self-learning closed loop.

[0008] Preferably, the dual optical path asynchronous redundant transmission includes: Send the same data packet to the receiving end through two independent relay paths; The receiving end removes abnormal data packets based on timestamp differences and topology fingerprint information and completes data fusion; During implementation, the execution flow of dual-optical path asynchronous redundant transmission is as follows: the sending end copies the key data packet into two independent copies and assigns a different relay path to each copy. The process includes: the first path selects the main relay node sequence based on the dynamic optical relay priority weight list, and the second path selects a topologically non-intersecting backup relay node combination based on the path redundancy model to ensure that the two paths do not have a common optical transceiver unit or parallel optical path on the physical transmission link; each data packet header is embedded with a microsecond timestamp generated by a high-precision atomic clock, and the ID sequence of the relay node passing through is encrypted using the SHA256 hash algorithm to generate a unique topological fingerprint code; the receiving end listens to the data packets of the two paths within the preset asynchronous receiving window, first extracts the timestamp and calculates the transmission delay difference of the two paths. If the delay difference exceeds the preset threshold, it is determined that there is an abnormality in a certain path and triggers the alarm log; then the topological fingerprint is decrypted and verified, and the relay node ID is restored by reverse hash calculation The system then compares the data packet with the preset path node list. If a missing node ID, disordered order, or illegal node insertion is detected, the data packet is marked as abnormal and stored in the isolation buffer. The valid data packet that passes the verification enters the fusion stage and adopts a dynamic weight fusion strategy: if the data of the two paths are valid and the content is consistent, the data packet with the latest timestamp is selected first. If there is a difference in the content, the data is weightedly fused according to the path reliability weight, and the integrity of the fused data is verified by the cyclic redundancy check code. If the verification fails, a retransmission request is sent to the sender, and the dynamic optical relay priority weight of the abnormal path is reduced. At the same time, the system periodically counts the transmission indicators of each path. When a path is abnormal for three consecutive times, it is automatically removed from the relay candidate list, and the path weight recalculation process is triggered to dynamically optimize the dual path selection strategy. In addition, the receiving end maintains a topology fingerprint whitelist database and performs real-time blacklist filtering on unverified relay node IDs to prevent malicious node injection attacks.

[0009] Preferably, determining the occlusion type and triggering path switching or topology reconstruction includes: If the optical signal intensity fluctuation is a transient drop, it is determined to be a transient obstacle and the transmission path is dynamically switched; If the optical signal strength is continuously lower than the threshold, it is determined to be permanently blocked and triggers local topology reconstruction and updates the relay nodes; During the implementation process, the system monitors the optical signal strength data in real time at a sampling frequency of 2000 times per second through a multi-directional tunable optical transceiver module, and calculates the short-term and long-term intensity fluctuation characteristics based on the sliding window algorithm: when the signal strength in the short-term window drops sharply and exceeds the preset threshold value of 60dB attenuation rate, but the long-term window mean fluctuates within the threshold range, it is judged as a transient obstacle obstruction, such as a mobile forklift passing through briefly, and the path switching is immediately triggered, that is: the current suboptimal node information is extracted from the dynamic relay priority list, the transmission direction of the multi-directional tunable optical transceiver module is adjusted to the corresponding angle of the backup path, and the system is synchronously switched to the pre-assigned protection wavelength in the elastic channel. The data flow forwarding switching is completed within 5 milliseconds, while the original path monitoring thread is retained to continuously collect signal recovery data; if the signal strength in the long-term window is lower than 75dBm for 10 consecutive sampling cycles and the attenuation slope approaches zero, it is judged as permanent obstruction, and the local topology reconstruction process is initiated: the affected relay link communication is frozen, and the mixed integer programming model is called The optimal relay node combination under the current network topology is recalculated, with candidate nodes with path redundancy greater than 0.8 and load balancing scores in the top 20% being prioritized to generate a new relay path sequence. A two-stage verification mechanism is adopted during the reconstruction process: the signal strength stability of the new path is simulated through a simulation module, and then test data packets are sent in the actual link to verify the bit error rate. After verification, the data flow is gradually migrated to the new path, and the wavelength resources occupied by the old path are released after the migration is complete. For known obstruction patterns, the system matches the characteristic curve with a historical light intensity fluctuation database. If the match similarity exceeds 90%, the backup path parameters are pre-loaded and the protection wavelength is pre-allocated, shortening the switching response time to less than 3 milliseconds. After the switching or reconstruction is completed, the system initiates a continuous monitoring thread. If the new path triggers the same type of alarm again within 30 seconds, it is determined that there is a structural communication defect in the area, triggering the global topology optimization algorithm to replan the RGV route and mark it as a high-risk avoidance area on the electronic map.

[0010] A multi-RGV IoT system based on optical communication, including: Relay management unit, used to generate and update the relay node network in real time, including topology perception submodule and path correction submodule; Wavelength assignment controller, which performs mixed integer programming and conflict prediction for elastic channels; Redundant check engine, used to achieve asynchronous check and fusion of dual optical path data; The multiple RGV IoT system based on optical communication is linked with each unit through a multi-directional tunable optical transceiver module to execute the steps of the multiple RGV IoT method based on optical communication; During the implementation process, the multiple RGV IoT system based on optical communication realizes its functions through the coordinated operation of the relay management unit, wavelength allocation controller and redundancy check engine: the relay management unit has a built-in topology perception submodule, which receives the RGV coordinates, signal strength and obstacle data uploaded by the multi-directional tunable optical transceiver module in real time, updates the global topology map every 50 milliseconds using the Dijkstra algorithm, and generates a relay node candidate list based on the path redundancy model; the path correction submodule continuously analyzes the light intensity fluctuation data, and when it detects that the signal attenuation slope exceeds the threshold, calls the protection wavelength parameters in the elastic channel wavelength pool, generates a path switching instruction and sends it to the multi-directional tunable optical transceiver module of the corresponding RGV; the wavelength allocation controller runs a mixed integer programming model, combines the topology map with the task queue data, calculates the elastic channel wavelength allocation plan every 100 milliseconds, and monitors the usage status of each wavelength through the conflict prediction module. If it is detected that the RGV movement trajectory causes the risk of wavelength overlap within the next 20 milliseconds, it triggers a preemptive resource release instruction, and forces The redundancy check engine deploys a packet timestamp comparison unit at the receiving end, calibrating the timing differences of dual-path packets using a high-precision clock synchronization protocol. It also uses a topology fingerprint verification unit to perform hash inversion on the decrypted relay node ID sequence to match the preset path whitelist. The system implements data interaction between units through a bus architecture: the relay management unit pushes updated topology information to the wavelength allocation controller in real time. The wavelength allocation table generated by the latter is synchronized to the redundancy check engine for data fusion weight calculation. When the multi-directional tunable optical transceiver module dynamically adjusts the transmission direction and wavelength parameters according to instructions, it verifies the signal isolation after wavelength switching through closed-loop feedback and completes link reconstruction within 5 milliseconds. When local topology reconstruction is triggered, the system initiates a multi-module linkage protocol: the relay management unit freezes communications in the affected area, the wavelength allocation controller recalculates the wavelength mapping relationship, and the redundancy check engine suspends data fusion and enables the cache mechanism. After the new path passes the two-stage verification, data transmission is gradually resumed and the global status log is updated.

[0011] Preferably, the multi-directional tunable optical transceiver module includes: At least three optical signal transceiver units evenly distributed in the horizontal plane, with a coverage angle of not less than 240 degrees; The transceiver unit supports dynamic wavelength tuning to adapt to the wavelength allocation requirements of the basic channel and the flexible channel; During the implementation process, the hardware realization of the multi-directional tunable optical transceiver module includes: three independent optical signal transceiver units are evenly distributed in the horizontal plane on the top of the RGV body, each unit covers an azimuth angle of 80 degrees, and the total coverage range reaches 240 degrees. Each unit achieves adaptive alignment of the pitch angle of ±15 degrees through a universal adjustment bracket; each transceiver unit integrates a tunable laser and a photodetector, and the laser wavelength tuning range covers the C band, supporting rapid switching between the fixed wavelength of the basic channel and the dynamic wavelength of the elastic channel; dynamic wavelength tuning is achieved through a digital micromirror array, which dynamically adjusts the incident angle of the diffraction grating to select the target wavelength according to the instruction parameters issued by the wavelength allocation controller, and monitors the output wavelength deviation in real time and compensates for it through a closed-loop feedback circuit. The optical path design of the transceiver unit adopts a dual-ring structure, namely: the inner ring is used for fixed-wavelength communication of the basic channel, and the outer ring carries the dynamic wavelength data flow of the elastic channel. The two rings are isolated by a wavelength division multiplexer to ensure that the spectrum interval is greater than 10nm. When the transmission direction needs to be switched, the module drives the universal bracket of the specified transceiver unit to rotate to the target azimuth angle, and synchronously activates the laser wavelength tuning module in the corresponding direction, completing the optical path alignment and wavelength locking within 2 milliseconds. In addition, the module has a built-in self-test function module, which periodically scans the transmission power, receiving sensitivity and wavelength offset of each transceiver unit. Abnormal data is reported to the relay management unit in real time, triggering the module hot backup switching or wavelength recalibration process to ensure the physical layer reliability of the communication link.

[0012] Preferably, the topology awareness submodule includes: An algorithm module that calculates the relative position between RGVs and the light path occlusion status in real time; The path deviation correction submodule includes an instruction generation module for dynamically adjusting the transmission path according to the light intensity fluctuation trend; During the implementation process, the topology perception submodule collects the optical signal strength, azimuth and distance data of the neighboring RGV in real time through the multi-directional tunable optical transceiver module, and combines its own three-dimensional coordinates and motion vector output by the inertial navigation module to build a dynamic topology map centered on the current RGV; the improved Dijkstra algorithm is run every 50 milliseconds to calculate the shortest path to the target node and its redundancy index, where redundancy is defined as the product of the number of replaceable relay nodes in the path and the link stability, and the result is stored in the topology database; the determination of the optical path obstruction state is achieved by fusing the optical signal intensity attenuation model with the environmental obstacle contour data: when the signal intensity drop rate is detected to exceed 40dB / ms and overlaps with the obstacle position scanned by the lidar, the direction is marked as an obstruction area, and the optical path interruption risk level is generated; the path correction submodule makes a decision based on the topology perception results and the real-time light intensity fluctuation data: if the light intensity If the standard deviation of fluctuations within a 100-millisecond window exceeds a preset threshold of 10dB, the pre-stored protection wavelength parameters in the elastic channel wavelength pool are called to generate a path switching instruction. This path switching instruction includes the target relay node ID, the transmission azimuth adjustment value, and the new wavelength allocation scheme. The switching instruction is then sent to the corresponding RGV's multi-directional tunable optical transceiver module via a low-latency bus. The module drives the universal bracket to rotate to the target angle and synchronously tunes the laser wavelength. At the same time, a closed-loop calibration process is activated: the signal strength of the new path is monitored in real time by a photodetector. If the intensity does not reach 65dBm for three consecutive sampling times, the original path is rolled back and the topology perception submodule is triggered to recalculate candidate nodes. In the event of permanent obstruction, the path correction submodule uses a mixed integer programming model to complete elastic channel reallocation and relay path reconstruction in the affected area within 200 milliseconds. The coordinates of high-risk areas are marked on the electronic map for use by the global path planning algorithm.

[0013] Preferably, the redundancy check engine includes: A packet timestamp comparison unit, used to identify differences in packet timing between two independent paths; A topology fingerprint verification unit, used to match the relay path topology information of the data packet source; During the implementation process, the redundancy check engine works in conjunction with the topology fingerprint verification unit through the data packet timestamp comparison unit: the timestamp comparison unit calibrates the receiving timestamps of the dual-path data packets based on the high-precision clock synchronization protocol, and calculates the transmission delay difference between the two paths. If the delay difference exceeds the 15 millisecond threshold, the abnormal path alarm is triggered and the data packet buffer isolation mechanism is started; the topology fingerprint verification unit performs a SHA256 hash inversion operation on the header encrypted field of the received data packet, decrypts the relay node ID sequence, and matches it with the preset legal path node whitelist database, that is: if the node ID is missing, the order is disordered, or there is an unauthorized node, it is determined to be a path tampering attack, the data packet is discarded, and a security log is generated; for the data packets that pass the verification, the engine executes a dynamic fusion strategy: when the dual-path data content is consistent, the data packet with the latest timestamp is selected; if there is a difference in the content, then Weighted fusion is performed based on the historical path reliability score, and the integrity of the fused data is verified by the CRC32 check code. If the check fails, a retransmission request is initiated to the sender, and the dynamic optical relay priority weight of the abnormal path is reduced by 40%. At the same time, the relay node of the path is marked as temporarily disabled. The engine periodically analyzes the historical verification data, counts the packet loss rate, number of tampering events and delay distribution of each path, and dynamically updates the path reliability score library. That is, when a path fails to be verified for three consecutive times, it is automatically removed from the relay candidate list, and the relay management unit is triggered to regenerate the dynamic optical relay priority weight. In addition, the topology fingerprint verification unit maintains a dynamic blacklist mechanism, implements real-time interception of paths with abnormal node IDs after decryption, and sends an alarm signal to the wavelength allocation controller through the bus, forcing the switching of the wavelength and transmission path of the affected data stream to form a closed-loop security protection.

[0014] (3) Beneficial effects The present invention provides a method and system for connecting multiple RGVs based on optical communication. It has the following beneficial effects: (1) This multi-RGV IoT method and system based on optical communication significantly improves the communication stability and collaborative efficiency of multiple RGVs in complex industrial scenarios through the deep integration of dynamic optical relay network and adaptive wavelength allocation technology; the node selection mechanism based on dynamic optical relay priority weight, combined with the mixed integer programming model of elastic channel, realizes uninterrupted transmission and efficient channel multiplexing of optical communication links in mobile obstruction scenarios; the coordinated application of conflict prediction mechanism and dual optical path redundancy check technology effectively avoids the risk of channel competition and ensures the integrity and real-time performance of key data, solving the problems of communication interruption, delay and data loss caused by signal obstruction or topology rigidity in traditional fixed node networking.

[0015] (2) This multi-RGV IoT method and system based on optical communication greatly enhances the robustness and adaptability of the system in a dynamic environment through a closed-loop optimization mechanism and multi-module linkage design; the dual-stage verification mechanism of path switching and topology reconstruction reduces the risk of network reconstruction, while the self-test function of the optical transceiver module and the dynamic weight adjustment strategy of the redundant check engine further enhance the physical layer reliability and anti-interference capability; its technical effect is directly reflected in the improvement of the smoothness, safety and response speed of the collaborative operation of multiple RGVs in industrial logistics scenarios, while reducing the risk of equipment idleness or collision due to communication failures, providing a feasible optical communication IoT solution for high-density, high-dynamic automated warehousing and production lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the overall process of the present invention; Figure 2 This is a control logic timing diagram of the present invention. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0018] See also Figure 1 and Figure 2 The present invention provides a technical solution: a method for connecting multiple RGVs based on optical communication, comprising the following steps: S1: The multi-directional tunable optical transceiver module on each RGV collects real-time location, motion status, and environmental obstacle information to generate dynamic optical relay priority weights; During implementation, the multi-directional tunable optical transceiver module installed on each RGV consists of at least three optical signal transceiver units evenly distributed in the horizontal plane. Each unit covers an angle of no less than 240 degrees. It uses laser ranging, inertial navigation, and environmental obstacle detection sensors to obtain its own position coordinates, movement speed, and direction data in real time, while also scanning the outline information of surrounding obstacles. The generation of dynamic optical relay priority weights is based on the following collaborative calculations: first, the real-time signal strength between adjacent RGVs is measured through the optical signal transceiver unit, and the number of alternative relay hops and link stability of the current transmission path are calculated in combination with the path redundancy model. Then, the remaining communication resources of each candidate relay node are evaluated based on its load balancing degree. Finally, the three parameters of signal strength, path redundancy, and load balancing are weighted and summed according to the preset weight coefficient to generate a dynamic priority ranking list, and the relay node with the highest comprehensive score is selected. When the movement of the RGV causes the main link signal attenuation to exceed the threshold, the system immediately calls the optimal node in the priority list, activates its optical transceiver unit to establish the relay link, and simultaneously updates the topological connection relationship in the path redundancy model to ensure that the relay switching process adapts to the dynamic environment. S2: Based on the dynamic optical relay priority weight, adjacent RGVs are selected as relay nodes to build a distributed dynamic optical relay network, achieving link coverage through optical signal relay transmission; During implementation, when the signal attenuation of the primary communication link exceeds a preset threshold due to RGV movement or environmental obstacles, the neighboring RGV nodes are screened according to the dynamic optical relay priority weight list; The process of screening neighboring RGV nodes based on the dynamic optical relay priority weight list is as follows: first, all candidate nodes within the current communication range are traversed, their priority scores, real-time locations, and signal strength data are extracted, and the path redundancy between each node and the target receiver is calculated using a topology-aware algorithm; the node with the highest score is selected as the relay node, and the optical signal transceiver unit in the corresponding direction of its multi-directional tunable optical transceiver module is activated, the transmission wavelength is adjusted to match the basic channel of the receiver, and a relay link is established; after the relay node receives the original optical signal, it dynamically allocates a new wavelength through the elastic channel wavelength pool and forwards the data to the next-hop node or target RGV, forming an optical signal relay transmission chain; at the same time, the system continuously monitors the load status and link stability of each relay node. When it detects that the node load exceeds the capacity threshold or the path redundancy decreases, it automatically triggers local topology reconstruction, reselects a backup node from the priority list, and switches the transmission path to ensure that there are no blind spots in the dynamic network coverage. In addition, after the relay link is established, the signal forwarding timing of the relay node is corrected in real time through the optical signal propagation delay and motion trajectory prediction model to avoid the cumulative delay caused by multi-hop transmission; S3: Divide the optical wavelength resources into basic channels and flexible channels, assign fixed wavelength transmission control instructions to the basic channels, and dynamically assign wavelength transmission data streams to the flexible channels based on mixed integer programming; During implementation, optical wavelength resources are divided into two parts: basic channels and flexible channels based on preset spectrum segmentation rules. The basic channels use a fixed wavelength allocation strategy, allocating a unique and non-overlapping fixed wavelength to each RGV for transmitting scheduling instructions and emergency shutdown signals. This low-latency, high-priority control information ensures the real-time and conflict-free transmission of critical instructions. The flexible channels retain a dynamic wavelength pool. Wavelength resources are not pre-bound to devices, but are dynamically allocated based on the real-time network topology and task requirements through a mixed integer programming model. The dynamic allocation process using a mixed integer programming model includes: first, establishing a mixed integer programming model with maximizing channel utilization as the objective function and subject to wavelength uniqueness constraints, load balancing constraints, and task urgency weights. After inputting data on the relative positions of RGVs, relay link load rates, and task types, a branch-and-bound algorithm is used to solve the optimal wavelength allocation scheme. Among them, the wavelength uniqueness constraint requires that wavelengths in the same area at the same time are not repeated; the load balancing constraint requires that the concurrent data flows of a single node do not exceed the capacity threshold; and the task urgency weight requires that material handling tasks take priority over status reports.

[0019] During the dynamic allocation process, the system periodically updates the wavelength pool status every 100 milliseconds. When a change in the RGV position or a new data transmission request is detected, the model is recalculated and the current optimal wavelength is allocated to the data stream. At the same time, the elastic channel wavelength and the basic channel wavelength are separated by a protection band in the spectrum to avoid signal crosstalk. In addition, to cope with sudden channel competition, the system monitors the usage status of each wavelength in real time and predicts conflicts for RGV data flows that are about to enter the same communication area. If the risk of wavelength overlap is detected, the elastic channel reallocation process is forcibly triggered to ensure the continuity of data stream transmission. S4: Identify potential channel contention through conflict prediction mechanisms and dynamically adjust wavelengths or switch transmission paths; During implementation, the conflict prediction mechanism is implemented through the following steps: First, based on the optical signal propagation delay calculation model and combined with the real-time position, speed and movement direction data of each RGV, the overlapping range of the communication areas of each RGV in the future time period is predicted; the RGV movement trajectory is modeled using the Kalman filter algorithm to estimate its expected position within the next 50 millisecond time window, and combined with the elastic channel wavelength allocation table, it is determined whether there is a risk of multiple RGVs in the same area using the same wavelength to transmit data; if a potential conflict is detected, a preemption release instruction is generated based on the task urgency and the current channel load status; The process of generating preemption release instructions based on task urgency and current channel load status includes: for low-priority tasks, forcibly releasing their current wavelengths and reallocating idle wavelengths; for high-priority tasks, forwarding their data streams to idle wavelength channels of adjacent relay nodes through a path switching mechanism; at the same time, the system verifies the spectral isolation between the new wavelength and the wavelengths in the adjacent area through simulated transmission before wavelength switching to avoid signal crosstalk; when path switching involves multi-hop relays, the link weights in the topology awareness algorithm are synchronously updated, and the signal forwarding timing of the relay nodes is adjusted to compensate for the transmission delay caused by path switching; in addition, after completing the wavelength or path adjustment, the system continuously monitors the conflict resolution status. If channel contention still exists within the preset time window, it triggers a local topology reconstruction or data packet retransmission mechanism until the conflict is completely eliminated; S5: Dual optical paths are used for asynchronous redundant transmission of critical data. The receiving end verifies and merges data packets based on timestamps and topology fingerprints. During the implementation process, key data packets are replicated into two copies and transmitted through two independent relay paths. Specifically, the transmitter selects a non-overlapping sequence of relay nodes for each path based on a dynamic optical relay priority weight list to ensure that the two paths do not share common nodes or parallel optical paths in the physical topology. Each data packet is attached with a timestamp and path topology fingerprint information accurate to the microsecond level. Specifically, the timestamp is generated synchronously by the transmitter's atomic clock, and the topology fingerprint is encrypted and encoded using a hash algorithm on the ID sequence of the relay nodes passed through. After asynchronously receiving data packets from the two paths within a preset time window, the receiver first extracts the timestamp and calculates the transmission delay difference. If the delay difference exceeds the fault tolerance threshold of 10 milliseconds, it is determined that the path is abnormal and an alarm is triggered. The topology fingerprint is then decrypted and verified to match the preset relay path node sequence. If there is any missing node ID or sequence tampering, it is marked as an abnormal data packet. Data packets that pass the verification enter the fusion stage, using a timestamp priority-based strategy. The strategy based on timestamp priority includes the following: if data on both paths is valid, the data packet with the latest timestamp is selected; if only one path is valid, that data is directly used; if there are content differences between the two paths, weighted fusion is performed in combination with the topology fingerprint weights; among the topology fingerprint weights, the primary path weight is set to 0.6 and the backup path weight is set to 0.4; after the fusion is completed, an integrity check code is generated. If the check fails, the data retransmission mechanism is triggered, and the dynamic optical relay priority weight is updated to reduce the probability of selecting abnormal paths; in addition, the system periodically collects statistics on the packet loss rate and delay fluctuation data of each path, dynamically adjusts the dual path selection strategy, and gives priority to relay node combinations with low load and high redundancy. S6: Real-time monitoring of the optical signal intensity fluctuation trend, determination of the obstruction type and triggering path switching or topology reconstruction; during implementation, the system collects optical signal intensity data in real time at a sampling frequency of thousands of times per second through a multi-directional tunable optical transceiver module, and calculates the short-term fluctuation trend and long-term attenuation slope of the signal intensity based on a sliding window algorithm; during this period, a 50-millisecond window can be selected for the short-term, and a 500-millisecond window can be selected for the long-term; when a sharp fluctuation in signal intensity is detected within the short-term window, for example, the intensity drop rate exceeds 50dB / ms, but the mean value within the long-term window is stable, it is determined to be an instantaneous obstruction and the path switching mechanism is immediately triggered, namely: the suboptimal node is selected as the backup path from the dynamic optical relay priority weight list, the transmission direction and wavelength of the multi-directional tunable optical transceiver module are adjusted, and the data stream is seamlessly switched to the new path, while the original path monitoring is retained until the signal is restored; if the signal strength is continuously lower than the preset threshold of 70dBm within the long-term window and the attenuation slope is If the rate approaches zero, it is determined to be permanent obstruction, and the local topology reconstruction process is initiated. First, the current relay link is frozen, and the path redundancy model is recalculated based on the real-time updated RGV position information. A new list of relay node candidates is generated, and elastic channel wavelengths are allocated to the affected data streams based on a mixed integer programming model. During the reconstruction process, a progressive switching strategy is adopted. First, a new path is established and its stability is verified. That is, after three consecutive signal strength samplings meet the standard, the old path resources are released to avoid communication interruption during the switching. In addition, the system uses a library of light intensity fluctuation patterns to match historical obstruction scenarios, such as forklifts passing or shelf displacement. If a known transient obstruction pattern is matched, the backup path parameters are preloaded in advance, shortening the switching response time to less than 5 milliseconds. After the switching or reconstruction is completed, the system continuously monitors the signal strength stability of the new path. If the same type of obstruction alarm is triggered again within 10 seconds, the area is marked as a high-risk communication blind spot, and global topology optimization is triggered to avoid it.

[0020] Generating dynamic optical relay priority weights includes: determining the priority ranking of relay nodes through weighted calculation according to the real-time position, signal strength, path redundancy and load balancing of RGV, and screening the optimal relay path; in the implementation process, the generation of dynamic optical relay priority weights is achieved through the following steps: each RGV obtains the optical signal strength data of the neighboring nodes in real time through the multi-directional tunable optical transceiver module, and combines its own coordinates and motion vector output by the laser ranging and inertial navigation module to build a local topology map centered on the current RGV; the calculation of path redundancy is based on the relay hop count and path redundancy model of the candidate nodes in the topology map, where the hop count is the number of relay hops and the path redundancy model of the candidate nodes in the topology map. The number is obtained by solving the shortest path using the Dijkstra algorithm, and link stability is dynamically adjusted based on historical communication packet loss rate and delay fluctuation data. Load balancing is evaluated by monitoring the number of concurrent data streams, remaining bandwidth resources, and processing delay of each candidate node, quantifying its available communication capacity and converting it into a 0.1 standardized score. In the weighted calculation stage, signal strength, path redundancy, and load balancing are linearly superimposed according to preset coefficients to generate a priority score list, which is then sorted from high to low based on the score. Among them, the preset coefficients include setting the signal strength weight to 0.4, the path redundancy weight to 0.3, and the load balancing weight to 0.3.

[0021] When selecting the optimal relay path, the system prioritizes the node with the highest score, while verifying whether its current load is below the capacity threshold and whether the number of path hops is less than the maximum allowed value. If the primary node does not meet the constraints, the system automatically selects the suboptimal node in descending order; the maximum allowed value is 3 hops.

[0022] In addition, when a network topology change is detected, the weight recalculation process is immediately triggered, the priority list is updated and synchronized to all associated nodes to ensure that relay decisions adapt to the dynamic environment in real time; among them, network topology changes include the addition of new RGVs or node offline.

[0023] Dynamic allocation of elastic channels involves constructing a mixed integer programming model to calculate wavelength allocation plans in real time based on the relative positions of RGVs, task urgency, and relay link load. The elastic channel wavelength pool is dynamically adjusted as network topology changes to maximize channel utilization. During implementation, dynamic allocation of elastic channels is achieved through the following steps: The system constructs the objective function and constraints of the mixed integer programming model based on real-time data collected on the relative positions of RGVs, task type identification, and relay link load rates. Task type identification includes setting material handling priority to 1 and status reporting priority to 2.

[0024] The process of constructing the objective function and constraints of the mixed integer programming model includes: the objective function is set to maximize channel utilization, and the constraints include wavelength uniqueness constraint within the same communication area, maximum number of concurrent data streams per node, and task urgency weight factor; after preprocessing the model input data, the branch and bound algorithm is used to solve the problem and output the optimal wavelength allocation scheme, including the elastic channel wavelength value and effective duration corresponding to each data stream; among which, the maximum number of concurrent data streams per node is limited to no more than 5, and the task urgency weight factor includes: priority 1 task weight is 0.7, and priority 2 task weight is 0.3.

[0025] During the allocation process, the system reserves a wavelength switching buffer for high-priority tasks, ensuring that urgent tasks can quickly switch wavelengths without interrupting transmission when the topology changes. During the dynamic adjustment phase, the network topology is scanned every 100 milliseconds. When RGV movement causes the overlap of communication areas to change by more than 15% or a new data transmission request is detected, the model is recalculated and the available resource list in the wavelength pool is updated based on the latest topology. At the same time, a 10nm spectrum guard interval is set between the elastic channel wavelength and the basic channel wavelength, and a bandpass filter is used to eliminate adjacent channel interference. For sudden wavelength conflicts, the system monitors the occupancy status of each wavelength in real time and predicts conflicts for RGV data flows entering the same area. If two data flows are detected to enter the overlapping area using the same wavelength within the next 30 milliseconds, the lower-priority data flow is forced to perform wavelength switching and rerouted to an idle wavelength of a neighboring relay node. Before the switch, the spectral isolation between the new wavelength and the basic channel of the target receiver is verified. After the allocation is completed, the system records the usage history of each wavelength to optimize the load balancing weight coefficient in subsequent models, forming a closed-loop optimization mechanism.

[0026] The conflict prediction mechanism includes: predicting potential channel contention based on the optical signal propagation delay and RGV movement trajectory; when a conflict risk is detected, triggering a preemption release instruction, switching wavelengths or releasing channel resources; during implementation, the realization of the conflict prediction mechanism includes the following steps: the system obtains the signal transmission rate corresponding to the current channel wavelength through the optical signal propagation delay calculation module, and combines the real-time coordinates, velocity vector and movement direction of each RGV to use the extended Kalman filter algorithm to predict the communication area boundaries of each RGV within the next 50 milliseconds; based on the elastic channel wavelength allocation table and predicted position data, a conflict probability matrix is ​​constructed to identify communication areas and corresponding time windows where wavelength overlap may occur, that is: if two or more RGV data streams will use the same wavelength in the same area and at the same time, it will be marked as a high-risk conflict event; for detected conflict events, the system generates differentiated processing strategies based on the task urgency label; among them, the task urgency label includes that the handling task priority is higher than the status report.

[0027] The system generates differentiated processing strategies based on task urgency labels, including the following steps: high-priority tasks retain the original wavelength and activate the path switching mechanism, forwarding their data streams through the idle wavelengths of adjacent relay nodes, while adjusting the receiving filter parameters of the target node to match the new wavelength; low-priority tasks trigger a preemptive release instruction, forcibly releasing the current wavelength and reallocating it to an available wavelength outside the spectrum protection interval. Before allocation, the simulation module is used to verify the spectrum isolation between the new wavelength and the wavelengths in the adjacent area.

[0028] During the path switching process, if a multi-hop relay link is involved, the system will synchronously correct the signal forwarding timing of the relay node. The process includes: calculating the transmission delay compensation value based on the newly added number of hops and injecting it into the timestamp field in the packet header to ensure the timing synchronization of the receiving end; after completing the wavelength or path adjustment, the system tracks the conflict resolution status through the real-time monitoring module. If wavelength overlap is still detected within 20 milliseconds after the adjustment, the local topology reconstruction process is initiated, the elastic channel allocation plan for the affected area is recalculated, and the packet retransmission mechanism is triggered to repair any lost information; in addition, the system records the time, location and processing results of historical conflict events to optimize the motion trajectory prediction accuracy of the Kalman filter algorithm, forming a conflict prediction self-learning closed loop.

[0029] Dual-optical path asynchronous redundant transmission involves sending the same data packet to the receiver via two independent relay paths. The receiver then removes abnormal data packets based on timestamp differences and topology fingerprint information, and completes data fusion. The implementation process for dual-optical path asynchronous redundant transmission is as follows: the sender replicates the critical data packet into two independent copies and assigns a different relay path to each copy.

[0030] The process of assigning a different relay path to each copy includes: the first path selects the main relay node sequence based on the dynamic optical relay priority weight list, and the second path selects a topologically non-intersecting backup relay node combination based on the path redundancy model to ensure that the two paths do not have a common optical transceiver unit or parallel optical path on the physical transmission link; each data packet header is embedded with a microsecond timestamp generated by a high-precision atomic clock, and the ID sequence of the relay node passed through is encrypted by the SHA256 hash algorithm to generate a unique topological fingerprint code; the receiving end listens to the data packets of the two paths within the preset asynchronous receiving window, first extracts the timestamp and calculates the transmission delay difference of the two paths. If the delay difference exceeds the preset threshold of 15 milliseconds, it is determined that the path is abnormal and the alarm log is triggered; then the topological fingerprint is decrypted and verified, and the relay node ID sequence is restored by reverse hash calculation and compared with the preset path node list. If it is detected that the node ID is missing, the order is disordered, or an illegal node is inserted, it will be The data packet is marked as abnormal and stored in the isolation buffer area; valid data packets that pass the verification enter the fusion stage and adopt a dynamic weight fusion strategy: if the data on both paths are valid and the content is consistent, the data packet with the latest timestamp is preferentially selected; if there is a difference in content, the data is weighted and fused according to the path reliability weight, that is, the main path weight is 0.7 and the backup path weight is 0.3, and the integrity of the fused data is verified by the cyclic redundancy check code; if the verification fails, a retransmission request is initiated to the sender, and the dynamic optical relay priority weight of the abnormal path is reduced; at the same time, the system periodically counts the transmission indicators of each path. When a path is abnormal for three consecutive times, it is automatically removed from the relay candidate list, and the path weight recalculation process is triggered to dynamically optimize the dual-path selection strategy; in addition, the receiving end maintains a topology fingerprint whitelist database and performs real-time blacklist filtering on unverified relay node IDs to prevent malicious node injection attacks; among them, the transmission indicators include packet loss rate, average delay, and signal fluctuation variance.

[0031] Judging the type of obstruction and triggering path switching or topology reconstruction includes: if the optical signal intensity fluctuation is a momentary drop, it is determined to be a momentary obstacle and the transmission path is dynamically switched; if the optical signal intensity is continuously lower than the threshold, it is determined to be a permanent obstruction and triggers local topology reconstruction and updates the relay node; during the implementation process, the system uses a multi-directional tunable optical transceiver module to monitor the optical signal intensity data in real time at a sampling frequency of 2000 times per second, and calculates the short-term and long-term intensity fluctuation characteristics based on the sliding window algorithm. When the signal intensity in the short-term window drops sharply and exceeds the preset threshold value of 60dB attenuation rate, but the long-term window mean fluctuates within the threshold range, it is determined to be a momentary obstacle obstruction, such as a mobile forklift passing through briefly, and the path switching is immediately triggered; among them, the short time corresponds to a 20 millisecond window and the long time corresponds to a 500 millisecond window.

[0032] The process of immediately triggering path switching includes: extracting the current suboptimal node information from the dynamic relay priority list, adjusting the transmission direction of the multi-directional tunable optical transceiver module to the corresponding angle of the backup path, synchronously switching to the pre-assigned protection wavelength in the elastic channel, completing the data stream forwarding switch within 5 milliseconds, and retaining the original path monitoring thread to continuously collect signal recovery data; if the signal strength in the long-term window is lower than 75dBm for 10 consecutive sampling cycles and the attenuation slope approaches zero, it is determined to be permanent obstruction, such as: adding a fixed shelf obstruction; initiating the local topology reconstruction process includes: freezing the affected relay link communication, calling the mixed integer programming model to recalculate the optimal relay node combination under the current network topology, giving priority to candidate nodes with path redundancy higher than 0.8 and load balancing scores in the top 20%, and generating a new relay path sequence; a two-stage verification mechanism is used in the reconstruction process.

[0033] The two-stage verification mechanism used in the reconstruction process includes: first simulating the signal strength stability of the new path through the simulation module, then sending test data packets in the actual link to verify the bit error rate =, and then gradually migrating the data flow to the new path after verification. After the migration is complete, the wavelength resources occupied by the old path are released. Among them, during the period when the simulation module simulates the signal strength stability of the new path, the sampling strength must be greater than 70dBm for five consecutive times; while during the period when sending test data packets in the actual link to verify the bit error rate, it must be less than 1E6.

[0034] For known obstruction patterns, such as periodic equipment movement, the system matches characteristic curves against a database of historical light intensity fluctuations. If the matching similarity exceeds 90%, the system preloads backup path parameters and pre-assigns protection wavelengths, compressing the switching response time to less than 3 milliseconds. After the switch or reconstruction is completed, the system starts a continuous monitoring thread. If the new path triggers the same type of alarm again within 30 seconds, it is determined that there is a structural communication defect in the area, triggering a global topology optimization algorithm to replan the RGV route and mark it as a high-risk avoidance area on the electronic map.

[0035] A multi-RGV IoT system based on optical communication, including: Relay management unit, used to generate and update the relay node network in real time, including topology perception submodule and path correction submodule; Wavelength assignment controller, which performs mixed integer programming and conflict prediction for elastic channels; Redundant check engine, used to achieve asynchronous check and fusion of dual optical path data; The multiple RGV IoT system based on optical communication is linked with each unit through a multi-directional tunable optical transceiver module to execute the steps of the multiple RGV IoT method based on optical communication; During the implementation process, multiple RGV IoT systems based on optical communication realize their functions through the coordinated operation of relay management unit, wavelength allocation controller and redundancy check engine: the relay management unit has a built-in topology perception submodule, which receives the RGV coordinates, signal strength and obstacle data uploaded by the multi-directional tunable optical transceiver module in real time, updates the global topology map every 50 milliseconds using the Dijkstra algorithm, and generates a list of relay node candidates based on the path redundancy model; the path correction submodule continuously analyzes the light intensity fluctuation data, and when it detects that the signal attenuation slope exceeds the threshold, it calls the protection wavelength parameters in the elastic channel wavelength pool, generates a path switching instruction and sends it to the multi-directional tunable optical transceiver module of the corresponding RGV. Tuning optical transceiver module; the wavelength allocation controller runs a mixed integer programming model, combines the topology map with the task queue data, calculates the elastic channel wavelength allocation plan every 100 milliseconds, and monitors the usage status of each wavelength through the conflict prediction module. If it is detected that the RGV movement trajectory causes the risk of wavelength overlap within the next 20 milliseconds, it triggers a preemptive resource release instruction, forcing low-priority tasks to switch to idle wavelengths; the redundancy check engine deploys a data packet timestamp comparison unit at the receiving end, calibrates the timing differences of the dual-path data packets through a high-precision clock synchronization protocol, and uses the topology fingerprint verification unit to perform hash inversion on the decrypted relay node ID sequence to match the preset path whitelist.

[0036] The system uses a bus architecture to achieve data interaction between various units. Specifically, the relay management unit pushes updated topology information to the wavelength allocation controller in real time. The wavelength allocation table generated by the latter is synchronized to the redundancy check engine for data fusion weight calculation. When the multi-directional tunable optical transceiver module dynamically adjusts the transmission direction and wavelength parameters according to instructions, it is necessary to verify the signal isolation after wavelength switching through closed-loop feedback and complete link reconstruction within 5 milliseconds. When local topology reconstruction is triggered, the system initiates a multi-module linkage protocol: the relay management unit freezes communication in the affected area, the wavelength allocation controller recalculates the wavelength mapping relationship, the redundancy check engine suspends data fusion and enables the cache mechanism. After the new path passes the two-stage verification, data transmission is gradually resumed and the global status log is updated. The two-stage verification includes simulation strength testing and actual bit error rate detection.

[0037] The multi-directional tunable optical transceiver module includes: At least three optical signal transceiver units evenly distributed in the horizontal plane, with a coverage angle of not less than 240 degrees; The transceiver unit supports dynamic wavelength tuning to adapt to the wavelength allocation requirements of basic channels and flexible channels; During the implementation process, the hardware realization of the multi-directional tunable optical transceiver module includes: three independent optical signal transceiver units are evenly distributed in the horizontal plane on the top of the RGV body, each unit covers an azimuth angle of 80 degrees, and the total coverage range reaches 240 degrees. Each unit realizes adaptive alignment of the pitch angle of ±15 degrees through a universal adjustment bracket; each transceiver unit integrates a tunable laser and a photodetector, and the laser wavelength tuning range covers the C band, which is 1530nm-1565nm, and supports fast switching between the fixed wavelength of the basic channel and the dynamic wavelength of the elastic channel, that is: the switching time is less than 1 millisecond; the dynamic wavelength tuning is achieved through the digital micromirror array DMD, which dynamically adjusts the incident angle of the diffraction grating to select the target wavelength according to the instruction parameters issued by the wavelength allocation controller, and at the same time through the closed loop The feedback circuit monitors the output wavelength deviation in real time and compensates for temperature drift; the optical path design of the transceiver unit adopts a dual-ring structure, namely: the inner ring is used for fixed-wavelength communication of the basic channel, and the outer ring carries the dynamic wavelength data stream of the elastic channel. The two rings are isolated by a wavelength division multiplexer to ensure that the spectrum interval is greater than 10nm; when the transmission direction needs to be switched, the module drives the universal bracket of the specified transceiver unit to rotate to the target azimuth angle, and synchronously activates the laser wavelength tuning module in the corresponding direction, completing the optical path alignment and wavelength locking within 2 milliseconds; in addition, the module has a built-in self-test function module, which scans the transmission power, receiving sensitivity and wavelength offset of each transceiver unit every 5 seconds. Abnormal data is reported to the relay management unit in real time, triggering the module hot backup switching or wavelength recalibration process to ensure the physical layer reliability of the communication link.

[0038] The topology perception submodule includes: an algorithm module for real-time calculation of the relative position between RGVs and the light path obstruction status; the path correction submodule includes an instruction generation module for dynamically adjusting the transmission path according to the light intensity fluctuation trend; during the implementation process, the topology perception submodule collects the optical signal intensity, azimuth and distance data of the adjacent RGV in real time through the multi-directional tunable optical transceiver module, and combines its own three-dimensional coordinates and motion vector output by the inertial navigation module, the motion vector includes speed and acceleration, to construct a dynamic topology map centered on the current RGV; the improved Dijkstra algorithm is run every 50 milliseconds to calculate the shortest path to the target node and its redundancy index, where redundancy is defined as the product of the number of alternative relay nodes in the path and the link stability, and the result is stored in the topology database; the determination of the light path obstruction status is achieved by fusing the optical signal intensity attenuation model with the environmental obstacle contour data: when the signal intensity drop rate is detected to exceed 40dB / ms and overlaps with the obstacle position scanned by the lidar, the direction is marked as an obstruction area and an optical path interruption is generated. Risk level (low / medium / high). The path correction submodule makes decisions based on topology perception and real-time light intensity fluctuation data. If the standard deviation of light intensity fluctuations within a 100-millisecond window exceeds a preset threshold of 10 dB, it calls upon the pre-stored protection wavelength parameters in the elastic channel wavelength pool to generate a path switching instruction. This instruction includes the target relay node ID, the transmission azimuth adjustment value, and the new wavelength allocation scheme. The switching instruction is sent via a low-latency bus to the multi-directional tunable optical transceiver module of the corresponding RGV. The module drives the gimbal to rotate to the target angle and synchronously tunes the laser wavelength. A closed-loop calibration process is also activated. A photodetector monitors the signal strength of the new path in real time. If the intensity does not reach 65 dBm for three consecutive samplings, it rolls back to the original path and triggers the topology perception submodule to recalculate candidate nodes. In the event of permanent obstruction, the path correction submodule uses a mixed integer programming model to complete elastic channel reallocation and relay path reconstruction in the affected area within 200 milliseconds. The coordinates of high-risk areas are marked on the electronic map for use by the global path planning algorithm.

[0039] The redundancy check engine includes: A packet timestamp comparison unit, used to identify differences in packet timing between two independent paths; A topology fingerprint verification unit, used to match the relay path topology information of the data packet source; During the implementation process, the redundancy check engine works together with the topology fingerprint verification unit through the data packet timestamp comparison unit: the timestamp comparison unit calibrates the receiving timestamps of the dual-path data packets based on the high-precision clock synchronization protocol, and calculates the transmission delay difference of the two paths. If the delay difference exceeds the 15 millisecond threshold, it triggers the abnormal path alarm and starts the data packet buffer isolation mechanism; the topology fingerprint verification unit performs a SHA256 hash inversion operation on the header encrypted field of the received data packet, decrypts the relay node ID sequence, and matches it with the preset legal path node whitelist database, that is: if the node ID is missing, the order is disordered, or there is an unauthorized node, it is determined to be a path tampering attack, the data packet is discarded, and a security log is generated; for the data packets that pass the verification, the engine executes a dynamic fusion strategy: when the dual-path data content is consistent, the data packet with the latest timestamp is selected; if there is a difference in the content, it is weighted and fused according to the path reliability history score, and the path reliability history score is used as the weighted fusion. The main path score weight is 0.7, the backup path is 0.3, and the integrity of the fused data is verified by the CRC32 check code. If the check fails, a retransmission request is initiated to the sender, and the dynamic optical relay priority weight of the abnormal path is reduced by 40%. At the same time, the relay node of the path is marked as temporarily disabled. The engine analyzes historical verification data every 5 seconds, counts the packet loss rate, number of tampering events and delay distribution of each path, and dynamically updates the path reliability score library. That is, when a path fails to be verified for three consecutive times, it is automatically removed from the relay candidate list, and the relay management unit is triggered to regenerate the dynamic optical relay priority weight. In addition, the topology fingerprint verification unit maintains a dynamic blacklist mechanism, implements real-time interception of paths with abnormal node IDs after decryption, and sends an alarm signal to the wavelength allocation controller through the bus, forcibly switching the wavelength and transmission path of the affected data stream, forming a closed-loop security protection. Among them, the abnormal node ID includes unregistered devices or offline nodes.

[0040] Through the deep integration of dynamic optical relay networks and adaptive wavelength allocation technology, the communication stability and collaborative efficiency of multiple RGVs in complex industrial scenarios have been significantly improved; the node selection mechanism based on dynamic optical relay priority weights, combined with the mixed integer programming model of elastic channels, has achieved uninterrupted transmission and efficient channel multiplexing of optical communication links in mobile obstruction scenarios; the coordinated application of conflict prediction mechanism and dual optical path redundancy verification technology has effectively avoided the risk of channel competition and ensured the integrity and real-time performance of key data, solving the problems of communication interruption, delay and data loss caused by signal obstruction or topology rigidity in traditional fixed node networking.

[0041] Through the closed-loop optimization mechanism and multi-module linkage design, the robustness and adaptability of the system in dynamic environments are greatly enhanced; the two-stage verification mechanism of path switching and topology reconstruction reduces the risk of network reconstruction, and the self-test function of the optical transceiver module and the dynamic weight adjustment strategy of the redundant check engine further improve the physical layer reliability and anti-interference capability; its technical effect is directly reflected in the improvement of the smoothness, safety and response speed of multi-RGV collaborative operations in industrial logistics scenarios, while reducing the risk of equipment idleness or collision due to communication failures, providing a feasible optical communication Internet of Things solution for high-density, high-dynamic automated warehousing and production lines.

[0042] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0043] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for connecting multiple RGVs based on optical communication, characterized in that: The steps include: S1: The multi-directional tunable optical transceiver module on each RGV collects real-time location, motion status, and environmental obstacle information to generate dynamic optical relay priority weights; S2: Based on the dynamic optical relay priority weight, adjacent RGVs are selected as relay nodes to build a distributed dynamic optical relay network, and link coverage is achieved through optical signal relay transmission; S3: Divide the optical wavelength resources into basic channels and flexible channels, assign fixed wavelength transmission control instructions to the basic channels, and dynamically assign wavelength transmission data streams to the flexible channels based on mixed integer programming; S4: Identify potential channel contention through conflict prediction mechanisms and dynamically adjust wavelengths or switch transmission paths; S5: Dual optical paths are used for asynchronous redundant transmission of critical data. The receiving end verifies and merges data packets based on timestamps and topology fingerprints. S6: Monitor the optical signal intensity fluctuation trend in real time, determine the type of obstruction, and trigger path switching or topology reconstruction.

2. The method for connecting multiple RGVs to the Internet of Things based on optical communication according to claim 1, characterized in that: Generating the dynamic optical relay priority weight includes: According to the real-time location, signal strength, path redundancy and load balance of RGV, the priority ranking of relay nodes is determined through weighted calculation, and the optimal relay path is selected.

3. The method for connecting multiple RGVs based on optical communication according to claim 1, characterized in that: The dynamic allocation of the elastic channel includes: Based on the relative positions between RGVs, task urgency, and relay link load status, a mixed integer programming model is constructed to calculate the wavelength allocation plan in real time. The elastic channel wavelength pool is dynamically adjusted as the network topology changes to maximize channel utilization.

4. The method for connecting multiple RGVs to the Internet of Things based on optical communication according to claim 1, characterized in that: The conflict prediction mechanism includes: Predict potential channel contention based on optical signal propagation delay and RGV motion trajectory; When a conflict risk is detected, a preemption release instruction is triggered to switch wavelengths or release channel resources.

5. The method for connecting multiple RGVs to the Internet of Things based on optical communication according to claim 1, characterized in that: Dual optical path asynchronous redundant transmission includes: Send the same data packet to the receiving end through two independent relay paths; The receiving end removes abnormal data packets based on timestamp differences and topology fingerprint information and completes data fusion.

6. The method for connecting multiple RGVs based on optical communication according to claim 1, characterized in that: Determining the occlusion type and triggering path switching or topology reconstruction includes: If the optical signal intensity fluctuation is a transient drop, it is determined to be a transient obstacle and the transmission path is dynamically switched; If the optical signal strength is continuously lower than the threshold, it is determined to be permanently blocked and triggers local topology reconstruction and updates the relay nodes.

7. A multi-RGV IoT system based on optical communication, characterized by: include: Relay management unit, used to generate and update the relay node network in real time, including topology perception submodule and path correction submodule; Wavelength assignment controller, which performs mixed integer programming and conflict prediction for elastic channels; Redundant check engine, used to achieve asynchronous check and fusion of dual optical path data; The multi-RGV IoT system based on optical communication is linked with each unit through a multi-directional tunable optical transceiver module to execute any method described in claims 1-6.

8. The optical communication-based multi-RGV IoT system according to claim 7, characterized in that: The multi-directional tunable optical transceiver module includes: At least three optical signal transceiver units evenly distributed in the horizontal plane, with a coverage angle of not less than 240 degrees; The transceiver unit supports dynamic wavelength tuning to adapt to the wavelength allocation requirements of the basic channel and the flexible channel.

9. The optical communication-based multi-RGV IoT system according to claim 7, characterized in that: The topology perception submodule includes: An algorithm module that calculates the relative position between RGVs and the light path occlusion status in real time; The path deviation correction submodule includes an instruction generation module for dynamically adjusting the transmission path according to the light intensity fluctuation trend.

10. The optical communication-based multi-RGV IoT system according to claim 7, characterized in that: The redundancy check engine includes: A packet timestamp comparison unit, used to identify differences in packet timing between two independent paths; The topology fingerprint verification unit is used to match the relay path topology information of the data packet source.

Images