A trajectory prediction-based seamless switching system and method for visible light communication link of factory automated guided vehicle

Abstract

The application discloses a kind of factory automatic guided vehicle visible light communication link seamless switching system and method based on trajectory prediction, comprising: according to the current motion state of automatic guided vehicle, the motion state model of automatic guided vehicle is established, the motion trajectory of automatic guided vehicle is predicted using automatic guided vehicle motion state model, and the predicted position of automatic guided vehicle future time is obtained;According to the three-dimensional space data of factory workshop, a visible light communication channel model is established;According to the predicted position and the channel model, when the automatic guided vehicle triggers the communication link switching instruction, the communication quality of the candidate communication link between each candidate visible light communication light source and the automatic guided vehicle is evaluated by a predetermined evaluation index, and the optimal target communication link is determined;Through the switching instruction pre-generated and cached, switching is realized between the current communication link and the optimal target communication link, to ensure that the automatic guided vehicle and the automatic guided vehicle dispatch center always maintain communication.

Description

Technical Field

[0001] This invention relates to the fields of visible light communication and industrial intelligent logistics communication technology, specifically to a system and method for seamless switching of visible light communication links for automated guided vehicles in factories based on trajectory prediction. Background Technology

[0002] With the rapid development of industrial automation and intelligent manufacturing technologies, AGVs have been widely used in factory material transportation, warehouse scheduling, and flexible production line logistics systems. During operation, AGV clusters need to continuously communicate with the factory scheduling system in real time to perform functions such as path planning, obstacle avoidance, task scheduling, and status feedback. Therefore, the stability and real-time performance of the communication system directly affect the operational efficiency and safety of the industrial logistics system.

[0003] Currently, factory AGV communication mainly relies on wireless communication methods such as WiFi and Bluetooth. However, the large number of metal equipment, motors, and complex mechanical structures in industrial settings generate severe electromagnetic interference, leading to problems such as unstable communication links, disordered switching timing, and severe communication fluctuations. Furthermore, network congestion can easily occur when multiple AGVs communicate concurrently, affecting the real-time performance of the scheduling system.

[0004] Visible light communication utilizes LED lighting equipment for data transmission, offering advantages such as no electromagnetic interference, abundant spectrum resources, high security, and integrated lighting and communication, gradually becoming an important research direction for industrial AGV communication. However, most existing visible light communication systems employ fixed coverage and passive switching, meaning the system only initiates link switching when the quality of the current communication link to the AGV deteriorates to a set threshold. Since the AGV is in continuous motion, when switching is triggered, the current link is often already in a weak coverage area.

[0005] Furthermore, industrial workshop environments contain numerous shelves, equipment, and obstructions, resulting in complex AGV operating trajectories. Traditional switching methods based on fixed signal thresholds are ill-suited to the continuous communication requirements of dynamic industrial scenarios. Especially in areas where AGVs are operating at high speeds or where multiple vehicles converge, frequent link switching can easily lead to communication instability, affecting the reliable operation of industrial scheduling systems.

[0006] Therefore, how to combine the AGV's running trajectory and motion status to predict the AGV's future position in advance and complete the target link switch before the link quality deteriorates, so as to achieve continuous and stable transmission of visible light communication for AGVs in industrial scenarios, has become an important problem that needs to be solved. Summary of the Invention

[0007] The purpose of this invention is to provide, on the one hand, a seamless switching system for visible light communication links of automated guided vehicles (AGVs) in factories based on trajectory prediction, and on the other hand, a seamless switching method for visible light communication links of AGVs in factories based on trajectory prediction. This system and method can achieve seamless link switching and communication continuity assurance for AGVs in complex industrial workshop environments by establishing a dynamic motion model of AGVs, a visible light communication coverage model, and a trajectory prediction model. It can be applied to scenarios such as intelligent manufacturing, flexible warehousing, industrial logistics handling, and unmanned transportation systems.

[0008] To achieve this objective, the present invention provides a seamless switching system for visible light communication links of automated guided vehicles (AGVs) in factories based on trajectory prediction, characterized in that it comprises: The motion prediction module is used to establish a motion state model of the automated guided vehicle (AGV) based on its current motion state, and to predict the trajectory of the AGV using the motion state model to obtain the predicted position of the AGV at future moments. The visible light communication modeling module is used to build a visible light communication channel model based on the three-dimensional spatial data of the factory workshop; The communication link switching module is used to determine whether the automated guided vehicle (AGV) has triggered a communication link switching command based on the predicted future position of the AGV and the channel model of the visible light communication source. When the AGV triggers the communication link switching command, all available visible light communication sources within the factory area are selected as candidate visible light communication sources. The communication quality of the candidate communication links between each candidate visible light communication source and the AGV is evaluated using preset evaluation indicators to determine the optimal target communication link. The switching between the current communication link and the optimal target communication link is achieved through pre-generated and cached switching commands, ensuring that the AGV and the AGV dispatch center maintain communication at all times.

[0009] Furthermore, based on the current motion state of the automated guided vehicle (AGV), a motion state model of the AGV is established to predict its trajectory and obtain its predicted future position. This method includes: obtaining the AGV's state vector based on its current position, speed, and direction of motion. ; in, Let k be the state vector of the automated guided vehicle at time k. Let k be the x-coordinate of the current position of the automated guided vehicle. Let k be the ordinate of the current position of the automated guided vehicle. Let k be the velocity component of the automated guided vehicle in the x-axis direction at time k. Let k be the velocity component of the automated guided vehicle (AGV) along the y-axis. Based on the AGV's state vector, a Kalman filter algorithm is used to calculate the AGV's motion state in real time, and the future position state of the AGV is predicted based on the state transition model. in, Let A be the state vector of the automated guided vehicle at time k+1, and let A be the state transition matrix. This is system noise; The prediction time window Δt between time k and time k+1 is: in, For communication transmission delay; This refers to the link switching execution time; Get the future location of the automated guided vehicle for: .

[0010] Furthermore, the method for establishing a visible light communication channel model based on the three-dimensional spatial data of the factory workshop includes: constructing a factory spatial geometric model based on the three-dimensional spatial data of the factory workshop; and, based on the factory spatial geometric model, calculating the optical power received by the automated guided vehicle from the visible light communication light source using the Lambertian light source model to obtain the visible light communication channel model. ; in, For the automated guided vehicle to receive the optical power of the i-th visible light communication source, Let be the Lambert order. The receiving area of ​​the photodetector. The propagation distance of light from a visible light communication source. The radiation angle of the visible light communication light source. The incident angle of light from a visible light communication light source. This refers to the emission power of a visible light communication light source. For optical filter gain, For the concentrator gain, the light propagation distance of the visible light communication light source. The radiation angle of visible light communication light sources The angle of incidence of light from visible light communication light sources It was calculated using a factory spatial geometric model.

[0011] Furthermore, methods for determining whether an automated guided vehicle (AGV) has triggered a communication link switching command include: The automated guided vehicle (AGV) is deemed to need to perform a communication link switch when any of the following conditions are met: or ; in, The signal-to-interference-plus-noise ratio (SIR) of the visible light communication source with ID s at time k+Δt. The power was calculated based on the optical power received by the automated guided vehicle from the visible light communication source. To maintain a minimum signal-to-interference-plus-noise ratio (SIR) threshold for stable communication transmission, Let be the horizontal distance between the automated guided vehicle (AGV) and the currently serving visible light communication source at time k+Δt. It is calculated from the predicted position of the automated guided vehicle at a future time. The maximum effective horizontal connection distance between the automated guided vehicle and the visible light communication source to maintain stable communication transmission.

[0012] Furthermore, when the visible light communication light source numbered s is used as the service light source, the signal-to-interference-plus-noise ratio of the corresponding communication link is... The calculation formula is as follows: ; in, The optical power received by the automated guided vehicle from the serving visible light communication source numbered s is the visible light communication source that contributes the most to the received power. This refers to the noise power of the automated guided vehicle receiver. This refers to the interference power received by the automated guided vehicle (AGV), which is the total optical power generated by non-serving visible light communication sources on the AGV. It is calculated based on the optical power received by the automated guided vehicle from the visible light communication light source; Then the signal-to-interference-plus-noise ratio (SIR) of the visible light communication source with service number s at time k+Δt is: ; in, The optical power received by the automated guided vehicle from the visible light communication source numbered s at time k+Δt; Let k be the noise power of the automated guided vehicle receiver at time k+Δt; Let be the interference power received by the automated guided vehicle at time k+Δt, which is the total optical power generated by the non-serving visible light communication light source on the automated guided vehicle.

[0013] Furthermore, interference power It can be calculated as: ; Correspondingly, Where N is the number of visible light communication light sources, Let be the optical power generated by the i-th visible light communication light source for the automated guided vehicle. Let be the optical power generated by the i-th visible light communication light source on the automated guided vehicle at time k+Δt.

[0014] Furthermore, the method for evaluating the communication quality of candidate communication links between each candidate visible light communication source and the automated guided vehicle by using preset evaluation indicators, and determining the optimal target communication link, includes: Construct a comprehensive decision function for link switching based on the communication signal-to-interference-plus-noise ratio and the cost of communication link switching: ; in, This is the link handover evaluation value. This is the predicted signal-to-interference-plus-noise ratio (SIR / NNR) evaluation term for candidate communication links. For the cost of link switching, These are the weighting coefficients for the predicted signal-to-interference-plus-noise ratio (SIR) evaluation term of candidate communication links. The weighting coefficient for link switching cost is used to select the target visible light communication source with the highest link switching evaluation value to establish a communication link with the automated guided vehicle.

[0015] Furthermore, the method for switching between the current communication link and the optimal target communication link by pre-generating and caching switching instructions includes: the communication controller pre-generates and caches communication link switching instructions based on the predicted target visible light communication light source; when the horizontal connection distance between the automated guided vehicle and the currently serving visible light communication light source is greater than or equal to the maximum effective horizontal connection distance that can maintain stable communication transmission between the automated guided vehicle and the visible light communication light source at a certain future time, or when it is detected that the signal-to-interference-plus-noise ratio (SIR) of the communication link is lower than the minimum SIR threshold for maintaining stable communication transmission at a certain future time, the cached communication link switching instructions are invoked to perform the switching operation from the current communication link to the target communication link.

[0016] Furthermore, a method for seamless switching of visible light communication links for automated guided vehicles (AGVs) in factories based on trajectory prediction according to the system includes: Based on the current motion state of the automated guided vehicle (AGV), an AGV motion state model is established. The AGV motion state model is then used to predict the AGV's trajectory and obtain the predicted position of the AGV at future moments. A visible light communication channel model was established based on the three-dimensional spatial data of the factory workshop; Based on the predicted future position of the automated guided vehicle (AGV) and the channel model of the visible light communication source, it is determined whether the AGV will trigger a communication link switching command. When the AGV triggers the communication link switching command, all available visible light communication sources within the factory area are selected as candidate visible light communication sources. The communication quality of the candidate communication links between each candidate visible light communication source and the AGV is evaluated using preset evaluation indicators to determine the optimal target communication link. Through pre-generated and cached switching commands, the switching between the current communication link and the optimal target communication link is realized, ensuring that the AGV and the AGV dispatch center maintain communication at all times.

[0017] The beneficial effects of this invention are as follows: Addressing the problems in existing visible light communication systems for automated guided vehicles (AGVs) such as delayed link switching, low signal tracking accuracy, significant signal jitter in dynamic motion scenarios, difficulty in adapting traditional fixed-threshold switching methods to dynamic industrial scenarios, and poor link stability due to frequent ping-pong switching, this invention establishes an AAV motion state model and a visible light communication coverage model. It utilizes Kalman filtering to predict the future trajectory of the AAV in real time and pre-calculates the signal-to-interference-plus-noise ratio (SIR) of the current communication link based on the predicted position. When the predicted SIR is lower than a preset threshold or... When the measured distance exceeds the effective coverage radius, the target link pre-selection process is triggered in advance. Based on this, the system traverses all available visible light communication sources within the plant area, comprehensively predicts signal-to-interference-plus-noise ratio (SIR) and link switching costs, constructs a comprehensive decision function, selects the optimal target communication link, and pre-generates switching instructions cached in the controller. When the automated guided vehicle (AGV) reaches the preset switching node or detects that the current link's SIR is below a preset threshold, the controller immediately invokes the cached instructions to complete the communication link switch, thereby effectively shortening the switching control delay and avoiding switching lag and communication interruption caused by the AGV's high-speed movement. Simultaneously, this invention dynamically adjusts the prediction time window based on the AGV's operating speed, expanding the prediction range at high speeds to increase switching lead and reducing the prediction window at low speeds to reduce computational overhead. By organically combining the above-mentioned trajectory prediction and pre-buffering switching mechanism, this invention can significantly improve signal tracking accuracy, suppress signal jitter and ping-pong switching phenomena, reduce signal jump amplitude during switching, reduce the number of invalid reciprocating switches, and greatly reduce switching timing deviation. Thus, it can realize predictive seamless switching and continuous stable communication of the visible light communication link of automated guided vehicles in complex industrial scenarios, effectively ensuring the reliable operation of industrial scheduling systems. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the indoor VLC system of the present invention. Figure 2 This is a projection diagram of the AGV trajectory and LED layout of the present invention; Figure 3This is a comparison curve of SINR under the traditional handover and predictive seamless handover schemes of the present invention; Figure 4 This is a comparison chart of the LED switching smoothness of the present invention; Figure 5 This is a comparison diagram of the LED switching timing of the present invention; Figure 6 This is a schematic diagram of the structure of the present invention; Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to represent selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0020] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: Example 1 like Figure 6 As shown, a seamless switching system for visible light communication links of automated guided vehicles (AGVs) in factories based on trajectory prediction is characterized by comprising: The motion prediction module is used to establish a motion state model of the automated guided vehicle (AGV) based on its current motion state, and to predict the trajectory of the AGV using the motion state model to obtain the predicted position of the AGV at future moments. The visible light communication modeling module is used to build a visible light communication channel model based on the three-dimensional spatial data of the factory workshop; The communication link switching module is used to determine whether the automated guided vehicle (AGV) has triggered a communication link switching command based on the predicted future position of the AGV and the channel model of the visible light communication source. When the AGV triggers the communication link switching command, all available visible light communication sources within the factory area are selected as candidate visible light communication sources. The communication quality of the candidate communication links between each candidate visible light communication source and the AGV is evaluated using preset evaluation indicators to determine the optimal target communication link. The switching between the current communication link and the optimal target communication link is achieved through pre-generated and cached switching commands, ensuring that the AGV and the AGV dispatch center maintain communication at all times.

[0021] In some embodiments, the visible light communication light source includes an LED light source. An Automated Guided Vehicle (AGV) is a transport vehicle equipped with electromagnetic or optical automatic guidance devices, capable of traveling along a predetermined guidance path, and possessing safety protection and various transfer functions.

[0022] A schematic diagram of an indoor VLC (Visible Light Communication) system is shown below. Figure 1 As shown, the system constructs a three-dimensional Cartesian coordinate system based on the three-dimensional spatial data of the factory workshop, with the X, Y, and Z axes as the coordinates. Multiple visible light communication light sources are deployed in the indoor environment of the factory workshop. The coordinates of the first visible light communication light source are marked as follows: The coordinates of the i-th visible light communication light source Multiple visible light communication light sources constitute the indoor lighting and communication network, all installed at a height of H (e.g., 6 meters). The width and length of the factory workshop are defined as W and L, respectively. Automated Guided Vehicles (AGVs) operating on the workshop floor serve as mobile terminals, and their three-dimensional spatial coordinates at time t are defined as... The AGV (Automated Guided Vehicle) is equipped with a PD (photodetector) on its top to receive signals from a visible light communication light source above. The diagram uses a yellow cone-shaped beam to illustrate the downlink transmission link of the visible light signal, visually demonstrating the line-of-sight connection between the light source and the AGV receiver. Furthermore, the diagram also indicates the AGV's motion state, including the velocity component v along the positive X-axis. and velocity components along the positive Y-axis It is used to characterize the motion characteristics of AGV in a two-dimensional plane.

[0023] In some technical solutions, methods for establishing an automated guided vehicle (AGV) motion state model based on its current motion state, predicting its trajectory, and obtaining its predicted future position include: obtaining the AGV's state vector based on its current position, speed, and direction of motion. ; in, Let k be the state vector of the automated guided vehicle at time k. Let k be the x-coordinate of the current position of the automated guided vehicle. Let k be the ordinate of the current position of the automated guided vehicle. Let k be the velocity component of the automated guided vehicle in the x-axis direction at time k. Let k be the velocity component of the automated guided vehicle (AGV) along the y-axis. Based on the AGV's state vector, a Kalman filter algorithm is used to calculate the AGV's motion state in real time, and the future position state of the AGV is predicted based on the state transition model. in, Let A be the state vector of the automated guided vehicle at time k+1, and let A be the state transition matrix. This is system noise; The prediction time window Δt between time k and time k+1 is: in, For communication transmission delay; This refers to the link switching execution time; Get the future location of the automated guided vehicle for: .

[0024] By representing the motion state of the automated guided vehicle (AGV) as a four-dimensional state vector containing position and velocity components, and introducing a Kalman filter algorithm for state estimation and trajectory prediction, the interference of sensor noise and motion disturbances on position perception can be effectively suppressed, improving the accuracy and robustness of future position prediction. Furthermore, by setting the prediction time window to the sum of communication transmission delay and link handover execution time, the predicted position is aligned temporally with the actual AGV position at the moment of handover, avoiding handover timing mismatch caused by predictions being too early or too late. Thus, the system can obtain an accurate estimate of the AGV's future position before it enters a weak coverage area, providing a reliable spatiotemporal basis for subsequent link handover decisions.

[0025] AGV trajectory and LED layout projection, as shown Figure 2The graph illustrates the movement trajectory and link coverage effect of an automated guided vehicle (AGV) in a visible light communication environment. The horizontal axis represents the X-axis coordinate (unit: m), and the vertical axis represents the Y-axis coordinate (unit: m), both ranging from 0 to 20. In the graph, the black solid line with a circle represents the ideal trajectory (the AGV's ideal operating trajectory without any interference), the green solid line with a square represents the predicted trajectory (the predicted operating trajectory of the AGV), and the blue solid line with a triangle represents the observed trajectory (the observed actual operating trajectory of the AGV). Yellow dots represent the positions of LED light sources, and cyan dashed circles represent the coverage area of ​​the LED light sources. The graph contains nine LED light sources, indicated by yellow dots. Each light source is surrounded by a circular coverage area drawn with a cyan dashed line, and the coverage areas of adjacent light sources overlap to form a continuous communication area. The ideal trajectory (black solid line), the predicted trajectory (green solid line), and the observed trajectory (blue solid line) all exhibit an approximately straight line shape extending from the lower left to the upper right. To facilitate intuitive differentiation of the three curves and clearly demonstrate the details of trajectory deviation, the trajectory offset was artificially magnified for visualization during plotting: the green solid line (predicted trajectory) deviates upward by +0.5m from the ideal trajectory at approximately 7m in the X direction (upward offset in the Y direction); the blue solid line (observed trajectory) deviates downward by -0.5m from the ideal trajectory at approximately 14m in the X direction (downward offset in the Y direction). Under actual operating conditions, both the observed and predicted trajectories closely follow the ideal trajectory, and the actual motion error is stably controlled within ±0.5m, verifying the effectiveness of the motion prediction module.

[0026] In some technical solutions, the method for establishing a visible light communication channel model based on the three-dimensional spatial data of the factory workshop includes: constructing a factory spatial geometric model based on the three-dimensional spatial data of the factory workshop; and, based on the factory spatial geometric model, using a Lambertian light source model to calculate the optical power received by the automated guided vehicle from the visible light communication light source, thereby obtaining the visible light communication channel model. ; in, For the automated guided vehicle to receive the optical power of the i-th visible light communication source, Let be the Lambert order. The receiving area of ​​the photodetector. The propagation distance of light from a visible light communication source. The radiation angle of the visible light communication light source. The incident angle of light from a visible light communication light source. This refers to the emission power of a visible light communication light source. For optical filter gain, For the concentrator gain, the light propagation distance of the visible light communication light source. The radiation angle of visible light communication light sources The angle of incidence of light from visible light communication light sources It was calculated using a factory spatial geometric model.

[0027] By constructing a spatial geometric model using three-dimensional spatial data of a factory workshop and calculating the optical power of each visible light communication source received by the automated guided vehicle (AGV) based on a Lambertian light source model, physical factors such as light source layout, obstacle obstruction, and illumination distribution in the actual industrial environment can be incorporated into the channel modeling process. In this model, the light propagation distance, radiation angle, and incident angle are all directly solved from spatial geometric relationships, avoiding model bias caused by fitting empirical parameters, making the calculated received optical power at different locations more consistent with the real scene. The visible light communication channel model established in this way can not only support the link quality assessment at the current moment, but also, combined with the predicted location, enable the advance prediction of the channel state at future moments, providing a fundamental guarantee for the accuracy of communication link switching triggering.

[0028] In some embodiments, within the factory spatial geometry model, the calculation of the light propagation distance, radiation angle, and incident angle of the visible light communication light source is based on the three-dimensional spatial positional relationship between the light source and the receiver. First, based on the three-dimensional spatial data of the factory workshop (including the workshop's length, width, and height, coordinates and dimensions of obstructions such as shelves and equipment, the AGV's preset travel path coordinate sequence, and the height of the AGV's onboard receiver), the installation position coordinates of each LED light source and the receiver position coordinates of each sampling point on the AGV path are determined. The light propagation distance is the straight-line Euclidean distance between the center point of the LED light source and the center point of the AGV receiver, which can be obtained by taking the square root of the sum of the squares of the differences in the three-dimensional coordinates between the two points. The light radiation angle is defined as the angle between the normal direction of the LED light source's emitting surface and the line connecting the light source to the receiver. This angle depends on the installation orientation of the light source (usually vertically downward) and the direction vector of the receiver relative to the light source. The light incident angle is the angle between the normal direction of the receiver's receiving surface and the line connecting the receiver to the light source. This angle depends on the receiver's orientation (usually horizontally upward) and the direction vector of the light source relative to the receiver. Both of the aforementioned angles can be obtained through vector operations on the difference in three-dimensional coordinates between the light source and the receiver, as well as their respective normal directions. By introducing a factory space geometric model, abstract communication channel parameters can be transformed into concrete, calculable geometric quantities, thereby supporting subsequent communication coverage analysis and layout optimization.

[0029] In some technical solutions, methods for determining whether an automated guided vehicle (AGV) has triggered a communication link switching command include: The automated guided vehicle (AGV) is deemed to need to perform a communication link switch when any of the following conditions are met; otherwise, the AGV does not need to perform a communication link switch: or ; in, The signal-to-interference-plus-noise ratio (SIR) of the visible light communication source with ID s at time k+Δt. The power was calculated based on the optical power received by the automated guided vehicle from the visible light communication source. To maintain a minimum signal-to-interference-plus-noise ratio (SIR) threshold for stable communication transmission, Let be the horizontal distance between the automated guided vehicle (AGV) and the currently serving visible light communication source at time k+Δt. It is calculated from the predicted position of the automated guided vehicle at a future time. The maximum effective horizontal connection distance between the automated guided vehicle and the visible light communication source to maintain stable communication transmission.

[0030] By setting a dual decision based on predicted signal-to-interference-plus-noise ratio (SNR) and predicted distance, the system can comprehensively assess link handover needs from both signal quality and spatial coverage dimensions, avoiding potential missed or false judgments that might occur with a single threshold decision in complex industrial environments. When the Automated Guided Vehicle (AUV) has not exceeded the coverage radius but the SNR deteriorates prematurely due to increased interference, the SNR condition can trigger handover first. Conversely, when the SNR still meets the requirements but the AUV is about to leave the effective coverage area, the predicted distance condition can intervene in a timely manner. This complementary design of the two decision conditions ensures that handover triggering is neither too early nor too late, reducing unnecessary handover attempts while maintaining communication continuity and improving the system's adaptability to different movement trajectories and environmental interference.

[0031] In some embodiments, the method for calculating the horizontal distance between the automated guided vehicle (AGV) at time k+Δt and the currently serving visible light communication light source based on the predicted future position of the AGV includes: First, obtain the fixed plane coordinates of the current visible light communication light source in the factory's three-dimensional spatial coordinate system; Secondly, based on the predicted position coordinates of the automated guided vehicle at time k+Δt (these coordinates are calculated in real time by the motion state prediction module based on the Kalman filter algorithm); Finally, the Euclidean distance between the two points is calculated using the method of calculating the difference between the predicted position coordinates of the automated guided vehicle (AGV) and the planar coordinates of the serving visible light communication source along the horizontal and vertical axes, summing the results, and then taking the square root. This horizontal distance is used to determine whether the AGV is about to leave the effective coverage radius of the current serving visible light communication source, thereby triggering a link switching decision.

[0032] In some technical solutions, when the visible light communication light source numbered 's' is used as the service light source, the signal-to-interference-plus-noise ratio (SIR) of the corresponding communication link is... The calculation formula is as follows: ; in, The optical power received by the automated guided vehicle from the serving visible light communication source numbered s is the visible light communication source that contributes the most to the received power. This refers to the noise power of the automated guided vehicle receiver. This refers to the interference power received by the automated guided vehicle (AGV), which is the total optical power generated by non-serving visible light communication sources on the AGV. It is calculated based on the optical power received by the automated guided vehicle from the visible light communication light source; Then the signal-to-interference-plus-noise ratio (SIR) of the visible light communication source with service number s at time k+Δt is: ; in, The optical power received by the automated guided vehicle from the visible light communication source numbered s at time k+Δt; Let k be the noise power of the automated guided vehicle receiver at time k+Δt; Let be the interference power received by the automated guided vehicle at time k+Δt, which is the total optical power generated by the non-serving visible light communication light source on the automated guided vehicle.

[0033] An AGV onboard receiver is a specialized optoelectronic device installed on an Automated Guided Vehicle (AGV) to receive visible light communication (VLC) signals. It converts the modulated light signal emitted by an LED light source into an electrical signal, and then demodulates it to recover AGV scheduling instructions, position data, and other information. In a visible light communication (VLC) system, a concentrator is an optical device installed at the receiving end (such as an AGV onboard receiver). Its core function is to focus the light and increase the intensity of the received signal.

[0034] By calculating the signal-to-interference-plus-noise ratio (SIR) at future moments and clearly distinguishing between the three components—the received power of the serving light source, the total interference power of non-serving light sources, and the receiver noise power of the automated guided vehicle (AGV)—the quantitative assessment of link quality gains clear physical meaning and calculability. Specifically, the serving light source is selected as the visible light communication source that contributes the most to the received power, ensuring that the current communication link always operates under the optimal signal source. The introduction of interference power reflects the actual signal competition situation when multiple light sources coexist in industrial scenarios. The predictive calculation of the SIR at future moments allows the system to anticipate the degradation trend of the link quality before it actually deteriorates, providing a forward-looking quantitative basis for switching decisions.

[0035] SINR comparison curves for traditional handover and predictive seamless handover schemes are as follows: Figure 3As shown in the figure, this graph illustrates the changes in signal-to-interference-plus-noise ratio (SINR) for three modes—ideal, conventional, and predictive—at different time steps. The vertical axis represents the SINR value (in dB), and the horizontal axis represents the evolution of time steps. In the figure, the black solid line with a diamond marker represents the SINR change curve for communication link switching under the ideal state; the red solid line with a square marker represents the SINR for communication link switching under the conventional method; and the blue solid line with a triangle marker represents the SINR for communication link switching under the predictive method. The graph shows that the SINR of the three modes fluctuates periodically over time, with significant troughs occurring between approximately 200 and 300 time steps and around 400 time steps. To more clearly illustrate the subtle differences, the figure also includes magnified local areas and offset indicators: At the low point near time step 400, the figure magnifies the SINR change in that area using a local box, displaying a corresponding magnified comparison within the box; the figure also includes captions "+0.5 (Y upward offset)" and "-0.5 (Y downward offset)," indicating the relative positions of the predicted and traditional curves in specific areas, visually demonstrating the subtle differences in SINR values ​​between the different methods. Overall, the figure clearly compares the communication quality performance of the three modes in dynamic environments. The predicted mode exhibits a relatively high SINR level in most time steps, while the traditional mode shows lower SINR values ​​in some low-point areas. This indicates that the prediction mechanism, by anticipating environmental changes and triggering handover, significantly outperforms the traditional passive handover mechanism in terms of communication link quality (SINR), closely approaching ideal communication performance.

[0036] LED switching smoothness, for example Figure 4 As shown in the figure, this diagram presents the performance comparison experimental results of two different schemes in a visible light communication system, aiming to quantitatively evaluate the impact of introducing a prediction mechanism on communication stability. The vertical axis in the figure is defined as the SINR jump (dB), used to measure the drastic degree of instantaneous signal changes in the communication link, while the horizontal axis compares the test data under the traditional scheme and the prediction scheme. Experimental data shows that in the traditional scheme without the prediction mechanism, the SINR jump value is relatively high, approximately 0.78dB, indicating that the communication link quality fluctuates drastically, easily leading to unstable data transmission. However, in the optimized scenario using the prediction scheme, the SINR jump value is significantly reduced to approximately 0.30dB. This data comparison proves that the motion prediction-based link switching mechanism can effectively suppress sudden changes in the signal-to-interference-plus-noise ratio (SINR) and significantly smooth out link fluctuations, thus verifying the substantial technical effect of this solution in maintaining the communication continuity of automated guided vehicles (AGVs).

[0037] In some technical solutions, interference power It can be calculated as: ; Correspondingly, Where N is the number of visible light communication light sources, Let be the optical power generated by the i-th visible light communication light source for the automated guided vehicle. Let be the optical power generated by the i-th visible light communication light source on the automated guided vehicle at time k+Δt.

[0038] By defining interference power as the total optical power generated by non-serving visible light communication sources on the automated guided vehicle (AGV), i.e., the sum of the received power of all visible light communication sources minus the received power of the serving source, the calculation process for interference power is simplified. An effective estimate of the global interference level can be obtained without modeling each interference source separately. Full utilization of the calculated received power data of each source avoids additional sensing or computational overhead, making it suitable for scenarios where rapid iterative calculations are required during AGV operation. The consistent structure of the calculation formulas for the current and future times facilitates the reuse of the same computational logic in the prediction phase, improving the system's real-time performance and engineering feasibility.

[0039] In some technical solutions, methods for evaluating the communication quality of candidate communication links between each candidate visible light communication source and the automated guided vehicle by setting preset evaluation indicators, and determining the optimal target communication link, include: Construct a comprehensive decision function for link switching based on the communication signal-to-interference-plus-noise ratio and the cost of communication link switching: ; in, This is the link handover evaluation value. This is the predicted signal-to-interference-plus-noise ratio (SIR / NNR) evaluation term for candidate communication links. For the cost of link switching, These are the weighting coefficients for the predicted signal-to-interference-plus-noise ratio (SIR) evaluation term of candidate communication links. The weighting coefficient for link switching cost is used to select the target visible light communication source with the highest link switching evaluation value to establish a communication link with the automated guided vehicle.

[0040] By constructing a weighted comprehensive decision function that includes a predicted signal-to-interference-plus-noise ratio (SIR) evaluation term and link handover cost, communication quality and handover cost are incorporated into a unified optimization objective, avoiding the pursuit of the highest SIR while ignoring the system overhead caused by frequent handovers. The SIR evaluation term reflects the signal quality of candidate links at future moments, while the handover cost encompasses the control signaling overhead required to re-establish the link, the risk of connection interruption, and potential data retransmission costs. The introduction of weighting coefficients allows for flexible adjustment of the priorities of both based on different application scenarios or operational phases. For example, the handover cost weight can be appropriately increased in high-speed areas to reduce handover frequency, or the SIR weight can be emphasized in areas with severe signal interference to ensure communication reliability. Finally, the candidate link with the highest evaluation value is selected as the target link, achieving globally optimal decision-making under multiple candidate conditions.

[0041] In some technical solutions, the method of switching between the current communication link and the optimal target communication link by pre-generating and caching switching instructions includes: the communication controller pre-generates and caches communication link switching instructions based on the predicted target visible light communication light source; when the horizontal connection distance between the automated guided vehicle and the currently serving visible light communication light source is greater than or equal to the maximum effective horizontal connection distance that can maintain stable communication transmission between the automated guided vehicle and the visible light communication light source at a certain future time, or when it is detected that the signal-to-interference-plus-noise ratio (SIR) of the communication link is lower than the minimum SIR threshold for maintaining stable communication transmission at a certain future time, the cached communication link switching instructions are invoked to perform the switching operation from the current communication link to the target communication link.

[0042] By pre-generating and caching handover commands for the target communication link, the handover decision and handover execution are decoupled in time, allowing the generation of handover commands to be completed before the link quality deteriorates. When the automated guided vehicle (AGV) reaches the preset handover node or detects that the current link's signal-to-interference-plus-noise ratio (SNR) is below the threshold, the controller does not need to re-execute the link selection and command generation process; instead, it directly calls the cached handover command to complete the link handover. This effectively reduces the control latency between the fulfillment of the trigger condition and the execution of the handover action, avoiding additional waiting time introduced by real-time calculation and signaling interaction. This is particularly suitable for the stringent requirements for handover response speed in high-speed AGV scenarios. Simultaneously, the pre-generated command caching strategy also reduces the instantaneous computational load on the controller in high-frequency handover scenarios, improving the overall operational stability and scalability of the system.

[0043] After the link switch is completed, the system connects the new LED communication link to the factory scheduling network, enabling data interaction between the AGV and the scheduling center. The scheduling center sends instructions such as path planning, obstacle avoidance scheduling, speed control, and task allocation to the AGV via the visible light communication network. At the same time, the AGV uploads its location, operating status, power information, and fault information in real time.

[0044] Furthermore, this invention dynamically adjusts the link prediction time window based on the AGV's operating speed. When the AGV is running at high speed, the system expands the prediction range in advance to increase the lead time for link switching; when the AGV is running at low speed or in a waiting state, the prediction window is reduced, thereby reducing the system's computational overhead.

[0045] Through the above methods, predictive seamless switching and continuous stable communication of the visible light communication link of AGVs in complex industrial scenarios can be achieved.

[0046] LED switching timing example Figure 5 As shown in the figure, this graph compares the trends of LED numbers connected to the Automated Guided Vehicle (AGV) over time during time steps 200 to 600. The vertical axis represents the LED number, and the horizontal axis represents the time step. The ideal state of the LED switching sequence is represented by a solid black line with a filled dot, serving as the theoretically optimal switching benchmark. The traditional strategy for LED switching sequence is represented by a red dashed line with a diamond mark. The predictive strategy for LED switching sequence is represented by a blue dashed line with a triangle mark. Observing the curve trend, four LED switching events occurred at approximately time steps 250, 380, 420, and 630. During the transition phase of each switching, the predictive strategy (blue dashed line) performed the smoothest, locking in and stably connecting to the next target LED number in advance, essentially maintaining synchronization with the ideal state (solid black line), effectively avoiding drastic fluctuations in the numbering. In contrast, the traditional strategy (red dashed line) exhibited significant drastic fluctuations and instability at the moment of switching, with its LED number jumping sharply near the target value before finally stabilizing. This comparison intuitively demonstrates that adopting a predictive mechanism can effectively improve the smoothness and foresight of AGV communication link switching.

[0047] In some embodiments, the present invention achieves continuous and stable communication during AGV movement through trajectory prediction and seamless switching mechanisms, which can effectively correct signal tracking deviations, improve switching lag, suppress jitter and ping-pong switching, and improve the stability of visible light communication and the operating efficiency of AGV scheduling systems in complex industrial environments.

[0048] Based on the proposed method, a 20m×20m×6m factory scene was constructed in this simulation. Nine LEDs were evenly arranged in a 3×3 configuration, with a single LED having an effective coverage radius of 4.4m, basically achieving overlapping coverage without blind spots. The AGV moved diagonally at a constant speed. The observation noise and 0.5s system transmission delay were superimposed to compare the performance of the traditional light source switching strategy and the predictive seamless switching strategy proposed in this invention.

[0049] (1) The overall communication baseline quality performance is excellent. The average SINR of the ideal, traditional handover and predictive seamless handover strategies are 2.67dB, 2.65dB and 2.66dB respectively, and the average values ​​of the three are basically the same. The lowest SINR of the traditional handover and predictive seamless handover schemes are -4.01dB and -3.65dB respectively. The SINR is improved under the predictive seamless handover scheme, but the physical space loss determines the lower limit of the signal, and the handover logic cannot break through this inherent limit.

[0050] (2) In terms of signal tracking accuracy, due to time delay and observation noise interference, the mean square error of traditional switching SINR is 0.123dB, while that of predictive seamless switching is only 0.034dB. The prediction strategy relies on position prediction, which reduces the signal deviation caused by noise and time delay, and the waveform fits the ideal state to a higher degree.

[0051] (3) In terms of switching stability, the traditional strategy has an average signal jump of 0.78dB per switch, and the signal fluctuation is relatively large; the prediction scheme has a jump of only 0.30dB, and the link signal transition is smoother during dynamic movement, with stronger anti-disturbance capability.

[0052] (4) The handover timing and decision stability differ significantly. The time delay causes the traditional handover timing to be seriously delayed, with a time deviation of 75.1 steps. Frequent decision swings lead to 8.0 ping-pong handovers, and the total number of handovers is as high as 17.4. Predictive seamless handover anticipates future positions in advance, compresses the timing deviation to 39.1 steps, reduces the number of ping-pong handovers to 1.2, and optimizes the total number of handovers to 6.9, greatly reducing ineffective back-and-forth handovers.

[0053] In summary, under the conditions of AGV motion trajectory, predetermined noise level and 0.5s delay, the predictive seamless handover strategy can maintain excellent communication SINR, effectively suppress signal jitter, reduce handover lag, and avoid frequent ping-pong handover. Its overall handover stability and practicality are significantly better than traditional delayed handover schemes.

[0054] Example 2 A method for seamless switching of visible light communication links for automated guided vehicles (AGVs) in factories based on trajectory prediction, according to the system, includes: Based on the current motion state of the automated guided vehicle (AGV), an AGV motion state model is established. The AGV motion state model is then used to predict the AGV's trajectory and obtain the predicted position of the AGV at future moments. A visible light communication channel model was established based on the three-dimensional spatial data of the factory workshop; Based on the predicted future position of the automated guided vehicle (AGV) and the channel model of the visible light communication source, it is determined whether the AGV will trigger a communication link switching command. When the AGV triggers the communication link switching command, all available visible light communication sources within the factory area are selected as candidate visible light communication sources. The communication quality of the candidate communication links between each candidate visible light communication source and the AGV is evaluated using preset evaluation indicators to determine the optimal target communication link. Through pre-generated and cached switching commands, the switching between the current communication link and the optimal target communication link is realized, ensuring that the AGV and the AGV dispatch center maintain communication at all times.

[0055] Example 3 The present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the method described in Embodiment 2.

[0056] This invention can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented in whole or in part as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0057] Those skilled in the art will readily understand that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, combinations, substitutions, improvements, etc., made under the spirit and principles of the present invention are included within the protection scope of the present invention.

[0058] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. A seamless switching system for visible light communication links of automated guided vehicles (AGVs) in factories based on trajectory prediction, characterized in that it include: The motion prediction module is used to establish a motion state model of the automated guided vehicle (AGV) based on its current motion state, and to predict the trajectory of the AGV using the motion state model to obtain the predicted position of the AGV at future moments. The visible light communication modeling module is used to build a visible light communication channel model based on the three-dimensional spatial data of the factory workshop; The communication link switching module is used to determine whether the automated guided vehicle (AGV) has triggered a communication link switching command based on the predicted future position of the AGV and the channel model of the visible light communication source. When the AGV triggers the communication link switching command, all available visible light communication sources within the factory area are selected as candidate visible light communication sources. The communication quality of the candidate communication links between each candidate visible light communication source and the AGV is evaluated using preset evaluation indicators to determine the optimal target communication link. The switching between the current communication link and the optimal target communication link is achieved through pre-generated and cached switching commands, ensuring that the AGV and the AGV dispatch center maintain communication at all times.

2. The visible light communication link seamless switching system for automated guided vehicles (AGVs) in factories based on trajectory prediction according to claim 1, characterized in that, The method for establishing an automated guided vehicle (AGV) motion state model based on its current motion state and predicting its trajectory to obtain its predicted future position includes: obtaining the AGV's state vector based on its current position, speed, and direction of motion. ; in, Let k be the state vector of the automated guided vehicle at time k. Let k be the x-coordinate of the current position of the automated guided vehicle. Let k be the ordinate of the current position of the automated guided vehicle. Let k be the velocity component of the automated guided vehicle in the x-axis direction at time k. Let k be the velocity component of the automated guided vehicle (AGV) along the y-axis. Based on the AGV's state vector, a Kalman filter algorithm is used to calculate the AGV's motion state in real time, and the future position state of the AGV is predicted based on the state transition model. ； in, Let A be the state vector of the automated guided vehicle at time k+1, and let A be the state transition matrix. This is system noise; The prediction time window Δt between time k and time k+1 is: in, For communication transmission delay; This refers to the link switching execution time; Get the future location of the automated guided vehicle for: .

3. The visible light communication link seamless switching system for automated guided vehicles (AGVs) in factories based on trajectory prediction according to claim 1, characterized in that, The method for establishing a visible light communication channel model based on the three-dimensional spatial data of the factory workshop includes: constructing a factory spatial geometric model based on the three-dimensional spatial data of the factory workshop; and, based on the factory spatial geometric model, calculating the optical power received by the automated guided vehicle from the visible light communication light source using the Lambertian light source model to obtain the visible light communication channel model. ； in, For the automated guided vehicle to receive the optical power of the i-th visible light communication source, Let be the Lambert order. The receiving area of ​​the photodetector. The propagation distance of light from a visible light communication source. The radiation angle of the visible light communication light source. The incident angle of light from a visible light communication light source. This refers to the emission power of a visible light communication light source. For optical filter gain, For the concentrator gain, the light propagation distance of the visible light communication light source. The radiation angle of visible light communication light sources The angle of incidence of light from visible light communication light sources It was calculated using a factory spatial geometric model.

4. The visible light communication link seamless switching system for automated guided vehicles in factories based on trajectory prediction according to claim 1, characterized in that, Methods for determining whether an automated guided vehicle (AGV) has triggered a communication link switching command include: The automated guided vehicle (AGV) is deemed to need to perform a communication link switch when any of the following conditions are met: or ; in, The signal-to-interference-plus-noise ratio (SIR) of the visible light communication source with ID s at time k+Δt. The power was calculated based on the optical power received by the automated guided vehicle from the visible light communication source. To maintain a minimum signal-to-interference-plus-noise ratio (SIR) threshold for stable communication transmission, Let be the horizontal distance between the automated guided vehicle (AGV) and the currently serving visible light communication source at time k+Δt. It is calculated from the predicted position of the automated guided vehicle at a future time. The maximum effective horizontal connection distance between the automated guided vehicle and the visible light communication source to maintain stable communication transmission.

5. A seamless switching system for visible light communication links of automated guided vehicles (AGVs) based on trajectory prediction, as described in claim 4, is characterized in that... When the visible light communication light source numbered s is used as the service light source, the signal-to-interference-plus-noise ratio of the corresponding communication link. The calculation formula is as follows: ； in, The optical power received by the automated guided vehicle from the serving visible light communication source numbered s is the visible light communication source that contributes the most to the received power. This refers to the noise power of the automated guided vehicle receiver. This refers to the interference power received by the automated guided vehicle (AGV), which is the total optical power generated by non-serving visible light communication sources on the AGV. It is calculated based on the optical power received by the automated guided vehicle from the visible light communication light source; Then the signal-to-interference-plus-noise ratio (SIR) of the visible light communication source with service number s at time k+Δt is: ； in, The optical power received by the automated guided vehicle from the visible light communication source numbered s at time k+Δt; Let k be the noise power of the automated guided vehicle receiver at time k+Δt; Let be the interference power received by the automated guided vehicle at time k+Δt, which is the total optical power generated by the non-serving visible light communication light source on the automated guided vehicle.

6. A seamless switching system for visible light communication links of automated guided vehicles (AGVs) based on trajectory prediction, as described in claim 5, is characterized in that... Interference power It can be calculated as: ； Correspondingly, the interference power received by the automated guided vehicle at time k+Δt for: ； Where N is the number of visible light communication light sources, Let be the optical power generated by the i-th visible light communication light source for the automated guided vehicle. Let be the optical power generated by the i-th visible light communication light source on the automated guided vehicle at time k+Δt.

7. A seamless switching system for visible light communication links of automated guided vehicles (AGVs) based on trajectory prediction, as described in claim 5, is characterized in that... The method for evaluating the communication quality of candidate communication links between each candidate visible light communication source and the automated guided vehicle by using preset evaluation indicators, and determining the optimal target communication link, includes: Construct a comprehensive decision function for link switching based on the communication signal-to-interference-plus-noise ratio and the cost of communication link switching: ； in, This is the link handover evaluation value. This is the predicted signal-to-interference-plus-noise ratio (SIR / NNR) evaluation term for candidate communication links. For the cost of link switching, These are the weighting coefficients for the predicted signal-to-interference-plus-noise ratio (SIR) evaluation term of candidate communication links. The weighting coefficient for link switching cost is used to select the target visible light communication source with the highest link switching evaluation value to establish a communication link with the automated guided vehicle.

8. A seamless switching system for visible light communication links of automated guided vehicles (AGVs) based on trajectory prediction, as described in claim 7, is characterized in that... The method for switching between the current communication link and the optimal target communication link by pre-generating and caching switching instructions includes: the communication controller pre-generates and caches communication link switching instructions based on the predicted target visible light communication light source; when the horizontal connection distance between the automated guided vehicle and the currently serving visible light communication light source is greater than or equal to the maximum effective horizontal connection distance that can maintain stable communication transmission between the automated guided vehicle and the visible light communication light source at a certain future time, or when it is detected that the signal-to-interference-plus-noise ratio (SIR) of the communication link is lower than the minimum SIR threshold for maintaining stable communication transmission at a certain future time, the cached communication link switching instructions are invoked to perform the switching operation from the current communication link to the target communication link.

9. A method for seamless switching of visible light communication links for automated guided vehicles (AGVs) in factories based on trajectory prediction, as described in claim 1, characterized in that... include: Based on the current motion state of the automated guided vehicle (AGV), an AGV motion state model is established. The AGV motion state model is then used to predict the AGV's trajectory and obtain the predicted position of the AGV at future moments. A visible light communication channel model was established based on the three-dimensional spatial data of the factory workshop; Based on the predicted future position of the automated guided vehicle (AGV) and the channel model of the visible light communication source, it is determined whether the AGV will trigger a communication link switching command. When the AGV triggers the communication link switching command, all available visible light communication sources within the factory area are selected as candidate visible light communication sources. The communication quality of the candidate communication links between each candidate visible light communication source and the AGV is evaluated using preset evaluation indicators to determine the optimal target communication link. Through pre-generated and cached switching commands, the switching between the current communication link and the optimal target communication link is realized, ensuring that the AGV and the AGV dispatch center maintain communication at all times.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method of claim 9.

Images