Carrier transport control method, garment hanging system, and readable storage medium

Abstract

The application relates to a carrier transportation control method, a clothing hanging system and a readable storage medium. Target order information and a first carrier transportation path matched with the target order information are acquired. In the process of driving a target carrier associated with the target order information to transport along the first carrier transportation path, if a current path efficiency index of the target carrier is less than a same-order average efficiency threshold, path deduction is performed based on the target order information and a current state vector corresponding to each target carrier associated with the target order information, a second carrier transportation path is determined, and a target carrier transportation path is determined from the first carrier transportation path and the second carrier transportation path to drive the target carrier to transport along the target carrier transportation path. The application avoids the problem of waste of computing resources caused by triggering global recalculation of all carriers without distinction in the related art due to congestion, and effectively reduces the computing cost of carrier transportation control.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technology, and in particular to a vehicle transportation control method, a clothing hanging system, and a readable storage medium. Background Technology

[0002] In a smart garment manufacturing factory, the production logistics system can be divided into three core links: the pre-sewing stage (cluster of cutting and sewing workstations), the bridging transport stage (connection and convergence across production lines), and the post-production sorting stage (finished product classification, quality inspection, packaging, and warehousing). The collaborative efficiency of these three links directly determines the overall factory capacity and delivery cycle. Among them, the bridging transport stage, as the core hub connecting the pre-sewing stage and the post-production sorting stage, has a critical bottleneck in terms of the control precision and response speed of its transport path, which determines whether the entire chain can flow smoothly.

[0003] In relevant garment hanging systems, the path control of the bridging transport channel generally adopts a static preset control scheme, which pre-plans fixed transport routes and sorting point allocation schemes for the vehicles. Once local congestion occurs, the system triggers global replanning to recalculate the path for all vehicles, resulting in a significant increase in computational overhead.

[0004] There is currently no effective solution to the problem of high computational overhead in vehicle transportation control in related technologies. Summary of the Invention

[0005] Therefore, it is necessary to provide a vehicle transportation control method, a clothing hanging system, and a readable storage medium that can reduce computational overhead and avoid resource waste, in order to address the above-mentioned technical problems.

[0006] In a first aspect, this application provides a vehicle transportation control method applied to a garment hanging system, the method comprising:

[0007] Obtain target order information and a first vehicle transportation route that matches the target order information;

[0008] During the process of driving the target vehicle associated with the target order information to transport along the first vehicle transport path, if it is identified that the current path efficiency index of the target vehicle is less than the average efficiency threshold of the same order, then the second vehicle transport path is determined based on the target order information and the current state vector of each target vehicle associated with the target order information.

[0009] A target vehicle transport path is determined from the first vehicle transport path and the second vehicle transport path to drive the target vehicle along the target vehicle transport path.

[0010] In one embodiment, obtaining a first vehicle transport route matching the target order information includes:

[0011] Obtain the current state vector corresponding to each carrier in the garment hanging system;

[0012] Based on the target order information and the current state vector of each vehicle, path planning is performed to determine the first vehicle transportation path that matches the target order information.

[0013] In one embodiment, obtaining the current state vector corresponding to each carrier in the garment hanging system includes:

[0014] Acquire multi-source heterogeneous data corresponding to each of the aforementioned vehicles; the multi-source heterogeneous data is collected by multi-source heterogeneous sensors deployed in the garment hanging system.

[0015] The multi-source heterogeneous data is fused to obtain the current state vector corresponding to each of the vehicles.

[0016] In one embodiment, the step of performing path planning based on the target order information and the current state vector corresponding to each of the vehicles, and determining the first vehicle transportation path matching the target order information, includes:

[0017] Obtain the geometric topology model corresponding to the garment hanging system;

[0018] In a digital twin simulation environment that matches the garment hanging system, path planning is performed based on the geometric topology model and the current state vectors of each of the vehicles to determine the initial optimal bridging path between the grid line and the rack line in the garment hanging transportation system.

[0019] Based on the initial optimal bridging path and the target order information, a first vehicle transportation path matching the target order information is determined.

[0020] In one embodiment, determining a first vehicle transport route matching the target order information based on the initial optimal bridging path and the target order information includes:

[0021] Based on the target order information, a set of candidate transportation origin and destination stations corresponding to the target order information is determined; the set of candidate transportation origin and destination stations includes at least one candidate sewing pre-exit station and at least one candidate sorting station.

[0022] Based on the station status corresponding to each of the candidate sewing pre-line exits and each of the candidate sorting ports in the candidate transportation start and end station set, the target sewing pre-line exit and the target sorting port are determined.

[0023] Based on the initial optimal bridging path, the target sewing pre-exit station, and the target sorting station, a first vehicle transportation path matching the target order information is determined.

[0024] In one embodiment, the method further includes:

[0025] Determine the current path efficiency index for each of the target vehicles in the transportation process;

[0026] Based on the mean of each current path efficiency index and a preset deviation factor, the average efficiency threshold of the same production order is determined.

[0027] In one embodiment, the step of determining the second vehicle transportation path based on the target order information and the current state vector corresponding to each of the target vehicles associated with the target order information includes:

[0028] Obtain the current state vector corresponding to each of the target vehicles associated with the target order information, as well as the set of candidate transportation origin and destination stations corresponding to the target order information;

[0029] Based on the current state vector corresponding to each of the target vehicles, the station status of each candidate sorting port in the candidate transportation start and end station set is determined.

[0030] Based on the site status of each candidate sorting point, a set of candidate transportation routes matching the target order information is determined; the set of candidate transportation routes includes at least one candidate transportation route.

[0031] Based on the current state vector corresponding to the target vehicle and each of the candidate transportation paths, calculate the path cost corresponding to each of the candidate transportation paths.

[0032] The candidate transportation path with the lowest path cost is determined as the second vehicle transportation path.

[0033] In one embodiment, determining the target vehicle transport path from the first vehicle transport path and the second vehicle transport path includes:

[0034] If the first vehicle transport route coincides with the second vehicle transport route, then the second vehicle transport route is determined as the target vehicle transport route;

[0035] If the first vehicle transport route does not overlap with the second vehicle transport route, then the first vehicle transport route is determined as the target vehicle transport route.

[0036] Secondly, this application provides a garment hanging system for performing the vehicle transport control method described above.

[0037] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0038] The aforementioned vehicle transportation control method, garment hanging system, and readable storage medium acquire target order information and a first vehicle transportation path matching the target order information. During the process of driving the target vehicle associated with the target order information along the first vehicle transportation path, by identifying whether the current path efficiency index of the target vehicle is less than the average efficiency threshold for the same order, the inefficient transport target vehicle can be accurately located. Furthermore, if the current path efficiency index of the target vehicle is identified as being less than the average efficiency threshold for the same order, a path deduction is performed based on the target order information and the current state vectors corresponding to each target vehicle associated with the target order information to determine a second vehicle transportation path. Thus, only the current... The vehicle-triggered replanning is significantly less efficient than the average of the same order group, thus avoiding the waste of computing resources caused by indiscriminate global recalculation of all vehicles due to congestion in related technologies, and effectively reducing the computational overhead of vehicle transportation control. At the same time, path deduction is performed based on the current state vector of each target vehicle, realizing online dynamic correction during transportation, improving the real-time performance of path control and the dynamic adaptability of the production line. Furthermore, the target vehicle transportation path is determined from the first vehicle transportation path and the second vehicle transportation path to drive the target vehicle along the target vehicle transportation path, which can ensure that the target vehicle always travels along the optimal path and ensure the stability of vehicle transportation. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating a vehicle transportation control method in one embodiment;

[0041] Figure 2 This is a flowchart illustrating the first vehicle transportation path acquisition step in one embodiment;

[0042] Figure 3 This is a flowchart illustrating the steps for determining the transportation path of the first vehicle in one embodiment.

[0043] Figure 4 This is a flowchart illustrating the steps for determining the transportation path of the second vehicle in one embodiment. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0045] In relevant garment hanging systems, path control of bridging transport channels generally adopts a static preset control scheme, pre-planning fixed transport routes and sorting point allocation schemes for the vehicles. Once local congestion occurs, the system triggers global replanning to recalculate the paths for all vehicles, resulting in a significant increase in computational overhead. To address this, this application provides a vehicle transport control method aimed at reducing the computational overhead of vehicle transport control and avoiding resource waste.

[0046] The vehicle transportation control method provided in this application will be described in detail below.

[0047] Before formally introducing the vehicle transportation control method of this application, the application scenario of the vehicle transportation control method of this application will be explained first. The vehicle transportation control method of this application is applied to a clothing hanging system; the clothing hanging system adopts a three-layer collaborative architecture of "end-edge-cloud", and its specific composition is as follows:

[0048] The edge-side (production line execution layer) includes intelligent bridging controllers, intelligent sorting controllers, servo drive modules, multi-source heterogeneous sensors, human-machine interface terminals (such as HMI touch screens), and industrial 5G communication gateways. The servo drive modules include grid drives, bridging drives, rack drives, and sorting drives; the multi-source heterogeneous sensors include RFID readers, photoelectric sensor arrays, infrared temperature measurement modules, industrial vision units, and vibration sensors. The intelligent bridging controller and intelligent sorting controller receive scheduling commands from the edge-side and drive the servo drive modules to perform specific actions such as release, waiting, speed change, and sorting port allocation.

[0049] The edge-side (production line edge computing layer) includes edge computing nodes deployed on each production line. The cloud-side (factory-level management and control layer) includes a factory digital twin platform, a global scheduling optimization engine, a big data analytics platform, and a visualization monitoring center. The factory digital twin platform is used to build a high-fidelity simulation model of the entire factory, supporting physically-level rendering accuracy for 3D visualization of the production line. The global scheduling optimization engine, based on a multi-agent reinforcement learning framework, utilizes the high-fidelity simulation environment built on the factory digital twin platform to train bridging-sorting collaborative scheduling strategies offline. The trained strategy network is then lightweighted and compressed before being periodically distributed to the edge computing nodes of each production line for continuous strategy updates. The big data analytics platform performs in-depth analysis of data such as vehicle transportation efficiency and equipment health, providing data support for parameter tuning of path planning algorithms. The visualization monitoring center provides a panoramic view of the entire factory's real-time operational status.

[0050] For example, the edge side is responsible for real-time data acquisition and command execution, uploading multi-source heterogeneous sensors to the edge side with a delay of ≤10ms; the edge side undertakes real-time inference and local decision-making, runs the vehicle transportation control method of this application, and can be autonomous offline when the network is interrupted; the cloud side is responsible for offline training and continuous optimization of the global scheduling strategy.

[0051] In one example embodiment, the hardware topology of the garment hanging system bridging channel is as follows: sewing pre-channel exit → grid line → bridging section (including merging track) → rack line → sorting port array. Multi-source heterogeneous sensors are deployed at key nodes throughout the bridging section to form a seamless sensing network. Specifically, this includes: RFID readers at the grid line exit and rack line merging point to collect carrier identification information, hanger codes, style information, and order data; photoelectric sensor arrays at the bridging section to collect real-time carrier position and speed; industrial vision units at the sorting port to identify hanger posture, garment type, and color attributes; and vibration sensors and infrared temperature measurement modules at key drive nodes to collect motor operating status and temperature data. Data collected by each sensor is uploaded to the edge computing node via an industrial 5G communication gateway, enabling the edge computing node to perform fusion processing based on multi-source heterogeneous data and output a unified state vector. The state vector refers to a set of multi-dimensional parameters obtained by data fusion processing from multi-source heterogeneous sensors, which characterizes the real-time operating status and business attributes of the vehicle. The state vector includes vehicle position, vehicle speed, vehicle load weight, selectable set of sewing pre-exit points, selectable set of sorting points, order priority, and estimated arrival time, which are used for path planning and dynamic scheduling decisions.

[0052] The vehicle transportation control method provided in this application is described below.

[0053] In one embodiment, such as Figure 1 As shown, Figure 1This is a flowchart illustrating a vehicle transport control method in one embodiment; the vehicle transport control method in this embodiment is applied to a garment hanging system and includes the following steps:

[0054] Step S101: Obtain the target order information and the first vehicle transportation route that matches the target order information.

[0055] The target order information may include, but is not limited to, order number, style information, number of vehicles, set of optional exits before sewing, set of optional sorting ports, and order priority. The set of optional exits before sewing and the set of optional sorting ports are used to constrain the range of start and end stations selected by vehicles in route planning, and the order priority is used for arbitration of release when vehicles merge.

[0056] The first vehicle transport path includes an access path from the target sewing pre-track exit to the initial optimal bridging path entrance, the initial optimal bridging path, and an exit path from the initial optimal bridging path exit to the target sorting port. The target sewing pre-track exit refers to the vehicle's starting departure station, determined after screening based on the selectable set of sewing pre-track exits in the target order information and the station status of each sewing pre-track exit. The target sorting port refers to the vehicle's destination station, determined after screening based on the selectable set of sorting ports in the target order information and the station status of each sorting port. The station status may include queue status, processing time, estimated waiting time, and congestion level. The initial optimal bridging path refers to the bridging path with the shortest travel time between the grid line and the rack line in the garment hanging transport system.

[0057] In an exemplary embodiment, the method for obtaining the first vehicle transportation path matching the target order information may be as follows: Based on the garment hanging system, a digital twin simulation environment matching the garment hanging system is constructed. In the digital twin simulation environment, using the current state vector corresponding to each vehicle as the initial condition, the production line CAD layout as the geometric constraint, and the optional sets of sewing pre-line exits and sorting ports in the target order information as boundary conditions, discrete event simulation is run to determine the initial optimal bridging path between the grid line and the rack line in the garment hanging transportation system; then, based on the initial optimal bridging path and the target order information, the first vehicle transportation path matching the target order information is determined. Here, the current state vector corresponding to the vehicle refers to a multi-dimensional parameter set obtained by data fusion processing from multi-source heterogeneous sensors, used to characterize the real-time operating state and business attributes of the vehicle; the current state vector includes vehicle position, vehicle speed, vehicle load weight, optional sets of sewing pre-line exits and sorting ports, order priority, and estimated arrival time, to achieve real-time state reconstruction of the entire production line link.

[0058] In step S102, during the process of driving the target vehicle associated with the target order information to transport along the first vehicle transport path, if it is identified that the current path efficiency index of the target vehicle is less than the average efficiency threshold of the same order, then the path is deduced based on the target order information and the current state vector of each target vehicle associated with the target order information to determine the second vehicle transport path.

[0059] The number of target vehicles is related to the order requirements corresponding to the target order information, and is not specifically limited here. For each target vehicle, the intelligent bridging controller and the intelligent sorting controller work together to control the target vehicle to transport along the first vehicle transport path. During the transportation process, the current path efficiency index and the average efficiency threshold of the same order are calculated in real time, and the current path efficiency index is compared with the average efficiency threshold of the same order to determine whether the path deduction process needs to be triggered.

[0060] The current path efficiency index is used to characterize the ideal remaining passage efficiency of the target vehicle on the first vehicle transport path. The current path efficiency index is calculated as follows: the ideal remaining time required for the target vehicle to reach the destination from the current position along the first vehicle transport path is used as the numerator, and the actual estimated remaining time required for the target vehicle to reach the destination from the current position along the first vehicle transport path is used as the denominator. Among them, the ideal remaining time is the time required for the target vehicle to pass through the remaining section of the first vehicle transport path without obstruction at the maximum operating speed, and the actual estimated remaining time is the time required to pass through the remaining section of the first vehicle transport path based on the target vehicle's current position, current speed, and the status of the queue at the stations ahead. The value range of the current path efficiency index is (0,1]. The closer it is to 1, the smoother the remaining section of the path is and the higher the efficiency. The closer it is to 0, the more congested the remaining section of the path is and the lower the efficiency.

[0061] The average efficiency threshold for the same order is determined based on the average value of the current path efficiency index of the target vehicle in the process of transportation, and is used to reflect the benchmark level of the passage efficiency of the vehicle group within the same order.

[0062] Understandably, by dynamically determining the average efficiency threshold of the same order, the benchmark level of the passage efficiency of the vehicle group within the same order can be reflected in real time. Furthermore, only vehicles whose current path efficiency index is lower than the average efficiency threshold of the same order are identified as inefficient vehicles and the path deduction process is triggered. This can avoid the problem of misjudgment or omission caused by using a preset fixed threshold, and improve the accuracy of vehicle transportation path control.

[0063] It should be noted that the path deduction described in this application starts from the current position of the target vehicle, uses the selectable set of sorting ports in the target order information as the endpoint constraint, and uses the production line CAD layout as the geometric constraint. It progressively advances the forward prediction of each candidate transportation path with a fixed step size, calculates the prediction time corresponding to each candidate transportation path, and determines the candidate transportation path with the shortest prediction time as the second vehicle transportation path, thus achieving online dynamic correction of the vehicle transportation process. The candidate transportation path is determined based on the current position of the target vehicle, the selectable set of sorting ports in the target order information, and the queue status of each sorting port.

[0064] It is understandable that the second vehicle transportation path is the current optimal remaining transportation path obtained after path deduction, and the starting point of the second vehicle transportation path is the current position of the target vehicle.

[0065] In an exemplary embodiment, a target vehicle associated with the target order information is driven to transport along a first vehicle transport path. For each target vehicle, during the transportation process along the first vehicle transport path, it is determined whether the current path efficiency index corresponding to the target vehicle is less than the average efficiency threshold of the same order. If the current path efficiency index corresponding to the target vehicle is less than the average efficiency threshold of the same order, it indicates that the passage efficiency of the target vehicle is significantly lower than the average level of the same order group, and there is room for path optimization. At this time, the path deduction process is triggered, that is, based on the target order information and the current state vector of each target vehicle associated with the target order information, the path deduction is performed to determine the second vehicle transport path. If the current path efficiency index corresponding to the target vehicle is greater than or equal to the average efficiency threshold of the same order, it indicates that the passage efficiency of the target vehicle is within the reasonable range of the same order group, and the current path is still relatively good. At this time, the target vehicle can continue to transport along the first vehicle transport path without triggering path deduction and path switching, thus avoiding unnecessary computational overhead.

[0066] Understandably, under the same order, the path efficiency of each target vehicle should be maintained within a certain deviation range from the group average level. Those that deviate need to trigger path deduction and adjustment to bring the entire order group to a globally better level. Path deduction is only triggered for those lagging behind with efficiency lower than the group average, thereby avoiding unnecessary path switching and reducing system computing overhead and resource waste.

[0067] Step S103: Determine the target vehicle transportation path from the first vehicle transportation path and the second vehicle transportation path, so as to drive the target vehicle to transport along the target vehicle transportation path.

[0068] It should be noted that since the current state vectors corresponding to each target vehicle and the station states corresponding to each sorting point are dynamically changing, there are two scenarios for the first and second vehicle transport paths: First, the first and second vehicle transport paths overlap. In this case, it indicates that the optimal remaining path derived from the current state is consistent with the initial planned path, and the current path is still relatively good, requiring no adjustment. The first vehicle transport path is then determined as the target vehicle transport path, and the target vehicle is driven to continue transporting along the first vehicle transport path. Second, the first and second vehicle transport paths do not overlap. In this case, it indicates that the optimal remaining path derived from the current state is better than the initial planned path, and the current path has room for optimization. The second vehicle transport path is then determined as the target vehicle transport path, and the target vehicle is driven to transport along the second vehicle transport path starting from its current position, achieving dynamic path switching and online correction.

[0069] In the above embodiments, target order information and a first vehicle transportation path matching the target order information are obtained. During the process of driving the target vehicle associated with the target order information along the first vehicle transportation path, by identifying whether the current path efficiency index of the target vehicle is less than the average efficiency threshold for the same order, the inefficient transport target vehicle can be accurately located. Furthermore, if the current path efficiency index of the target vehicle is identified as being less than the average efficiency threshold for the same order, path deduction is performed based on the target order information and the current state vectors corresponding to each target vehicle associated with the target order information to determine the second vehicle transportation path. Thus, only vehicles with significantly lower current path efficiency than the average efficiency threshold for the same order are transported. The vehicle-triggered replanning based on the average value of the order group avoids the waste of computing resources caused by the indiscriminate global recalculation of all vehicles due to congestion in related technologies, effectively reducing the computational overhead of vehicle transportation control. At the same time, path deduction is performed based on the current state vector corresponding to each target vehicle, realizing online dynamic correction during transportation, improving the real-time performance of path control and the dynamic adaptability of the production line. Furthermore, the target vehicle transportation path is determined from the first vehicle transportation path and the second vehicle transportation path to drive the target vehicle to travel along the target vehicle transportation path, which can ensure that the target vehicle always travels along the optimal path and ensure the smoothness of vehicle transportation.

[0070] In one embodiment, such as Figure 2 As shown, Figure 2 This is a flowchart illustrating the first vehicle transportation route acquisition step in one embodiment; based on the above embodiment, acquiring the first vehicle transportation route matching the target order information includes the following steps:

[0071] Step S201: Obtain the current state vector corresponding to each vehicle in the garment hanging system.

[0072] The current state vector is used to characterize the real-time operating status and business attributes of the corresponding vehicle, including vehicle location, vehicle speed, vehicle load weight, selectable set of sewing pre-exit outlets, selectable set of sorting outlets, order priority, and estimated arrival time.

[0073] In an exemplary embodiment, step S201 obtains the current state vector corresponding to each carrier in the garment hanging system, including the following steps:

[0074] Step 1: Acquire multi-source heterogeneous data corresponding to each carrier. This multi-source heterogeneous data is collected by multi-source heterogeneous sensors deployed in the garment hanging system; the specific limitations of these sensors are described above and will not be repeated here. In some embodiments, the multi-source heterogeneous data may include, but is not limited to, carrier identification information, hanger codes, style information, and order data collected by RFID readers; carrier position and speed collected by photoelectric sensor arrays; hanger posture, garment type, and color attributes collected by industrial vision units; and motor operating status and temperature data collected by vibration sensors and infrared temperature measurement modules.

[0075] Step 2: Perform data fusion processing on the multi-source heterogeneous data to obtain the current state vector corresponding to each vehicle.

[0076] For example, multi-source heterogeneous data is collected based on multi-source heterogeneous sensors. After time-stamp alignment and dimension normalization, the multi-source heterogeneous data is input into an extended Kalman filter for multi-sensor data fusion to generate the current state vector corresponding to each vehicle. The current state vector format can be S=[x, v, m, Go, Gr, p, ETA], where x is the current position of the vehicle (mm), v is the current speed of the vehicle (mm / s), m is the load weight of the vehicle (kg), Go is the selectable set of exit points before sewing, Gr is the selectable set of sorting points, p is the order priority (1~5), and ETA is the estimated arrival time (s). It should be noted that the extended Kalman filter outputs the current state vector through iterative fusion of prediction steps and update steps (correction of observations from each sensor).

[0077] It is understood that this embodiment obtains the current state vector of each vehicle by fusing multi-source heterogeneous data through an extended Kalman filter. This ensures that the multi-source heterogeneous data is uniformly fused into a single vector, guaranteeing the consistency of input for subsequent path planning and dynamic scheduling decisions, and avoiding the data fragmentation and information loss caused by independent processing by each sensor.

[0078] Step S202: Based on the target order information and the current state vector of each vehicle, perform path planning to determine the first vehicle transportation path that matches the target order information.

[0079] In one exemplary embodiment, such as Figure 3 As shown, Figure 3 This is a flowchart illustrating the first vehicle transportation path determination step in one embodiment; step S202 performs path planning based on the target order information and the current state vector corresponding to each vehicle to determine the first vehicle transportation path matching the target order information, including the following steps:

[0080] Step S301: Obtain the geometric topology model corresponding to the garment hanging system.

[0081] The geometric topology model refers to a digital track network model constructed based on the CAD layout of the garment hanging system production line. It includes the track topology (coordinates and connections of straight segments, curved segments, confluence points, and branching points), the coordinates of each station location, track segment length, slope information, sorting port number, and physical dimensions. The model stores nodes (stations / convergence points / branching points) and edges (track segments) in a graph structure, with length and slope attributes attached to the edges.

[0082] Step S302: In the digital twin simulation environment that matches the garment hanging system, path planning is performed based on the geometric topology model and the current state vector of each vehicle to determine the initial optimal bridging path between the grid line and the rack line in the garment hanging transportation system.

[0083] The digital twin model is built on a discrete event simulation framework, using real-time fused state vectors as input to continuously simulate the future operating state of the production line at a 1:1 time scale. It should be noted that the digital twin model is continuously compared with the actual production line, and physical parameters such as friction coefficient, motor response delay, and vehicle inertia in the model are automatically calibrated through online parameter identification algorithms (such as recursive least squares method) to ensure simulation accuracy (position error ≤ 5cm).

[0084] For example, using a geometric topology model as the basis of the simulation space and the current state vector corresponding to each vehicle as the initial condition, discrete event simulation is performed in a digital twin simulation environment that matches the garment hanging system to determine the set of feasible bridging paths between the grid line and the rack line in the garment hanging transportation system. Then, based on the path cost corresponding to each feasible bridging path, the feasible bridging path with the minimum path cost is selected as the initial optimal bridging path.

[0085] Step S303: Based on the initial optimal bridging path and the target order information, determine the first vehicle transportation path that matches the target order information.

[0086] In an exemplary embodiment, step S303, based on the initial optimal bridging path and the target order information, determines a first vehicle transport path that matches the target order information, including the following steps:

[0087] Step 1: Based on the target order information, determine the set of candidate transportation origin and destination stations corresponding to the target order information.

[0088] The target order information may include, but is not limited to, order number, style information, number of carriers, a set of optional pre-sewing exits, a set of optional sorting exits, and order priority. The candidate transport origin and destination station set is determined based on the set of optional pre-sewing exits and the set of optional sorting exits. The candidate transport origin and destination station set includes at least one candidate pre-sewing exit and at least one candidate sorting exit. A candidate pre-sewing exit refers to a selectable exit point where the target carrier corresponding to the target order information can exit from the sewing area and enter the bridging transport channel after completing the pre-sewing process. A candidate sorting exit refers to a selectable sorting exit point where the target carrier corresponding to the target order information can enter the downstream sorting area after completing the transfer in the bridging transport channel. It should be noted that carriers under the same order are not fixedly assigned to a single exit or sorting exit; the optimal one needs to be dynamically selected based on the status of each station to achieve load balancing and efficiency coordination among carriers of the same order.

[0089] Step 2: Based on the station status of each candidate sewing pre-exit and each candidate sorting point in the candidate transportation origin and destination station set, determine the target sewing pre-exit and the target sorting point.

[0090] The station status can include queue status, processing time, estimated waiting time, and congestion level. The target sewing front exit is the optimal starting point for the target vehicle to enter the bridging transport channel at the current moment, determined based on the station status assessment. The target sorting point is the optimal destination for the target vehicle to complete the downstream sorting operation at the current moment, determined based on the station status assessment.

[0091] Step 3: Based on the initial optimal bridging path, the target sewing pre-exit station, and the target sorting station, determine the first vehicle transportation path that matches the target order information.

[0092] For example, starting from the exit of the target sewing pre-line station, the route is connected to the entrance of the initial optimal bridging path via the access path, traversing the bridging transport middle road along the initial optimal bridging path, and then connected to the target sorting port via the outgoing path, forming a complete point-to-point transport path from the starting point to the end point, which serves as the first vehicle transport path matching the target order information.

[0093] Understandably, the initial optimal bridging path is only the optimal path between the grid line and the rack line, excluding the access segment from the sewing pre-track exit to the grid line exit, and also excluding the exit segment from the rack line merging point to the target sorting point. In actual transportation, the complete transportation path of the vehicle is formed by sequentially splicing together the access segment, the bridging segment, and the exit segment. The access segment is the path from the target sewing pre-track exit to the entrance of the initial optimal bridging path; the bridging segment is the initial optimal bridging path itself; and the exit segment is the path from the exit of the initial optimal bridging path to the target sorting point. Accordingly, after determining the initial optimal bridging path, it is necessary to further determine the target sewing pre-track exit and the target sorting point, and then splice the three paths together to obtain the complete first vehicle transportation path that matches the target order information.

[0094] In the above embodiments, by constructing a digital twin simulation environment that matches the garment hanging system, using a geometric topology model as the basis of the simulation space and the current state vector of each vehicle as the initial condition for discrete event simulation and deduction, the initial optimal bridging path between the grid line and the rack line can be accurately determined. Then, based on the optional set constraints in the target order information, the complete point-to-point first vehicle transportation path can be determined. This avoids the path rigidity and uneven load caused by using fixed mapping rules, ensures the matching of the path planning results with the order business constraints, and improves the response speed and execution feasibility of path planning.

[0095] In one embodiment, the vehicle transportation control method further includes the following steps:

[0096] Step 1: Determine the current path efficiency index for each target vehicle in the transportation process.

[0097] The current path efficiency index for each target vehicle in the transportation process is calculated as follows: the ideal remaining time required for the target vehicle to reach the destination from its current position along the first vehicle transportation path is used as the numerator, and the actual estimated remaining time required for the target vehicle to reach the destination from its current position along the first vehicle transportation path is used as the denominator. The ideal remaining time is the time required for the target vehicle to pass through the remaining section of the first vehicle transportation path without obstruction at its maximum operating speed, and the actual estimated remaining time is the time required to pass through the remaining section of the first vehicle transportation path based on the target vehicle's current position, current speed, and the status of the queue at the stations ahead. The value range of the current path efficiency index is (0,1]. The closer it is to 1, the smoother and more efficient the remaining section of the path is; the closer it is to 0, the more congested and less efficient the remaining section of the path is.

[0098] Step 2: Determine the average efficiency threshold for the same production order based on the mean of each current path efficiency index and the preset deviation factor.

[0099] The preset deviation factor is used to control the leniency of the average efficiency threshold of the same order, that is, the maximum deviation range of the current path efficiency of the vehicle relative to the average level of the same order. The larger the preset deviation factor, the lower the average efficiency threshold of the same order, and the less likely the vehicle is to trigger path simulation. The smaller the preset deviation factor, the higher the average efficiency threshold of the same order, and the more likely the vehicle is to trigger path simulation.

[0100] For example, assuming the mean of each current path efficiency index is η_avg, and η_avg is compared with a preset deviation factor δ, the average efficiency threshold for the same order is obtained as η_avg×(1-δ), where 0<δ<1, for example, δ can be taken as 15%. When the current path efficiency index η_i of a target vehicle is less than η_avg×(1-δ), it is determined that the efficiency of the vehicle is lagging behind the average level of the same order, and path deduction is triggered; otherwise, it continues to travel along the current path.

[0101] In the above embodiments, the average efficiency threshold of the same production order is determined by the average current path efficiency index of all target vehicles in transit of the same production order and the preset deviation factor. This allows the triggering condition to adaptively fluctuate with the overall operating status of the production line, avoiding the misjudgment problem caused by using a single fixed absolute threshold to trigger path deduction. By using the average efficiency threshold of the same production order as a dynamic reference benchmark for path deduction and judgment, the accurate identification and directional correction of truly inefficient vehicles are achieved, improving the accuracy of vehicle transportation path control and the rationality of system resource scheduling.

[0102] In one embodiment, such as Figure 4 As shown, Figure 4 This is a flowchart illustrating the steps for determining the transportation path of the second vehicle in one embodiment. Based on the above embodiment, the transportation path of the second vehicle is determined by path deduction based on the target order information and the current state vectors of each target vehicle associated with the target order information, including the following steps:

[0103] Step S401: Obtain the current state vector of each target vehicle associated with the target order information and the set of candidate transportation origin and destination stations corresponding to the target order information.

[0104] The candidate transport origin and destination station set includes at least one candidate pre-sewing exit and at least one candidate sorting exit. A candidate pre-sewing exit refers to a selectable exit point where the target vehicle corresponding to the target order information can exit from the sewing area and enter the bridging transport middle channel after completing the pre-sewing process; a candidate sorting exit refers to a selectable sorting exit where the target vehicle corresponding to the target order information can enter the downstream sorting area after completing the transfer in the bridging transport middle channel.

[0105] It should be noted that the method for determining the current state vector of each target vehicle can be found in the limitations mentioned above, and will not be repeated here.

[0106] Step S402: Based on the current state vector corresponding to each target vehicle, determine the station status of each candidate sorting port in the candidate transportation start and end station set.

[0107] For example, based on the current state vector corresponding to each target vehicle, the real-time queue length of each candidate sorting port is calculated, and combined with the average processing time of each sorting port, a site status including queue length, expected waiting time and congestion level is generated.

[0108] Step S403: Based on the site status of each candidate sorting point, determine the set of candidate transportation routes that match the target order information.

[0109] The candidate transportation route set includes at least one candidate transportation route.

[0110] It should be noted that each candidate sorting point is a set of selectable endpoints determined based on the target order information. However, not all sorting points in this set are included in the construction of candidate transportation routes. For example, if the site status of a candidate sorting point indicates that it is currently in a congested state (such as queue length exceeding a preset threshold, expected waiting time being too long, or equipment failure), then the candidate sorting point will be removed from the candidate transportation route set and will no longer participate in the route enumeration as a candidate endpoint. Only when the site status of a candidate sorting point indicates that it is available (i.e., not congested, not overflowing, and not malfunctioning) will a candidate transportation route be constructed based on that candidate sorting point. This ensures the feasibility and timeliness of the second vehicle transportation route and avoids invalid inferences and erroneous corrections caused by selecting unavailable endpoints.

[0111] Step S404: Based on the current state vector of the target vehicle and each candidate transportation path, calculate the path cost corresponding to each candidate transportation path.

[0112] The path cost characterizes the overall time consumption of a candidate transportation path. A lower path cost indicates a shorter overall time consumption and higher efficiency for the candidate transportation path. In an exemplary embodiment, the path cost is calculated as: C(r) = Σ(L / v) + Σ(Q×t + τ); where L is the track length of the candidate transportation path, v is the current speed of the target vehicle, Q is the current queue length of the candidate sorting station corresponding to the candidate transportation path, t is the average processing time per item at the candidate sorting station corresponding to the candidate transportation path, and τ is the processing time of the candidate sorting station for the target vehicle. The current speed of the target vehicle is extracted from the current state vector corresponding to the target vehicle.

[0113] For example, for each candidate transport path, the current speed of the target vehicle is determined based on the current state vector corresponding to the target vehicle; based on the candidate transport path, the track length of the candidate transport path, the current queue length of the candidate sorting port corresponding to the candidate transport path, and the average processing time per item of the candidate sorting port corresponding to the candidate transport path are determined. Based on the above parameters, the path cost corresponding to the candidate transport path is calculated according to the formula C(r)=Σ(L / v)+Σ(Q×t+τ).

[0114] Step S405: The candidate transportation route with the lowest path cost is determined as the second vehicle transportation route.

[0115] It is understandable that the smaller the path cost, the shorter the overall time required for the target vehicle to reach the destination from the current position along the candidate transportation path, and the higher the traffic efficiency. Determining the candidate transportation path with the minimum path cost as the second vehicle transportation path can ensure that the target vehicle continues transportation from the current position as the starting point and along the optimal remaining path, thereby realizing online dynamic correction and efficiency optimization of the transportation process.

[0116] In the above embodiments, when the current path efficiency index of the target vehicle is found to be less than the average efficiency threshold of the same order, the current state vector of each target vehicle associated with the target order information and the set of candidate transportation start and end stations corresponding to the target order information are obtained. Then, based on the current state vector of each target vehicle, the station status of each candidate sorting port in the set of candidate transportation start and end stations is determined. This allows for real-time perception of the queue length, expected waiting time, and congestion level of each candidate sorting port, providing a dynamic data basis for path selection and avoiding blind decision-making due to information lag. Based on the station status of each candidate sorting port, a set of candidate transportation paths matching the target order information is determined, which can filter out congested or unavailable candidate sorting ports and ensure the feasibility of candidate transportation paths. Based on the current state vector of the target vehicle and each candidate transportation path, the path cost of each candidate transportation path is calculated. The candidate transportation path with the minimum path cost is determined as the second vehicle transportation path, which ensures that the target vehicle continues transportation along the optimal remaining path, improving transportation efficiency and the dynamic adaptability of the production line.

[0117] In one embodiment, determining the target vehicle transportation path from the first vehicle transportation path and the second vehicle transportation path includes the following steps:

[0118] Step 1: If the first vehicle transportation route coincides with the second vehicle transportation route, then the second vehicle transportation route is determined as the target vehicle transportation route.

[0119] Step 2: If the first vehicle transportation route does not overlap with the second vehicle transportation route, then the first vehicle transportation route is determined as the target vehicle transportation route.

[0120] It should be noted that if the first and second vehicle transport paths overlap, it means that the optimal remaining path derived from the current state is consistent with the initial planned path, and the current path is still relatively good, requiring no adjustment. The first vehicle transport path is then designated as the target vehicle transport path, and the target vehicle is driven to continue transporting along the first vehicle transport path. If the first and second vehicle transport paths do not overlap, it means that the optimal remaining path derived from the current state is better than the initial planned path, and the current path has room for optimization. In this case, the second vehicle transport path is designated as the target vehicle transport path, and the target vehicle is driven to transport along the second vehicle transport path starting from its current position, achieving dynamic path switching and online correction.

[0121] In the above embodiments, by determining whether the first vehicle transportation path and the second vehicle transportation path overlap, the switching is only performed when the path changes, avoiding path oscillations caused by meaningless switching and ensuring the stability of vehicle transportation. When the first vehicle transportation path and the second vehicle transportation path overlap, the first vehicle transportation path is determined as the target vehicle transportation path, which can reduce unnecessary computational overhead and command issuance. When the first vehicle transportation path and the second vehicle transportation path do not overlap, the second vehicle transportation path is determined as the target vehicle transportation path, and the target vehicle is driven to transport along the second vehicle transportation path starting from its current position, realizing online dynamic correction and efficiency optimization of the transportation process.

[0122] In some embodiments, a convergence conflict avoidance method is provided: obtaining the state vectors of each vehicle about to arrive at the same convergence point and the position coordinates of the convergence point; predicting the arrival time of each vehicle at the convergence point according to the formula t=(x_merge-x) / v, where x_merge is the position coordinate of the convergence point, x is the current position of the vehicle, v is the current speed of the vehicle, and t is the estimated arrival time of the vehicle at the convergence point; and generating an arrival sequence by arranging the arrival times in ascending order; if the arrival time difference between adjacent vehicles is less than the safe time window t_safe=d_safe / v_ If a vehicle's speed is determined by a predefined safety distance threshold (d_safe) and the average speed of adjacent vehicles (v_avg), a confluence conflict risk is identified. Upon detection of a confluence conflict risk, a release order is determined based on one or more combinations of vehicle priority, estimated arrival time, and first-in-first-out (FIFO) sequence. Vehicles are then controlled to pass through the confluence point according to the release order. If, during travel, the distance to the vehicle ahead is detected to be less than the predefined safety distance threshold, the vehicle is slowed down. If the downstream queue length is detected to exceed the predefined threshold, reverse pressure deceleration is implemented to control vehicle deceleration or stop release. It is understood that confluence points include, but are not limited to, sorting inlet entrances, intersections of bridging sections and rack lines, and merging points where multiple grid lines converge at the bridging section.

[0123] In some embodiments, based on the above embodiments, the bridging-sorting collaborative scheduling of the garment hanging system is implemented using a multi-agent reinforcement learning (MARL) framework. Each intelligent bridging controller and each intelligent sorting controller are modeled as an independent decision-making agent, collectively forming a multi-agent system. Each agent's state space includes the distribution of vehicles within its jurisdiction, a summary of the load states of adjacent controllers, and the order structure within the current time window; its action space includes the bridging agent's release / waiting / speed-shifting cruise (speed gears 1-5) and the sorting agent's allocation to sorting ports 1-K or temporary buffers. The reward function is designed as a weighted combination of multiple objectives:

[0124] ;

[0125] Wherein, w1+w2+w3+w4+w5=1, and the values ​​of w1~w5 can be adjusted and determined according to production line requirements; T is throughput, W is average waiting time, V is sorting port load variance, E is energy consumption, and O is the number of timed-out orders; throughput is the number of carts that complete sorting per unit time; average waiting time is the average time taken for a cart to complete sorting from entering the line; sorting port load variance is the variance of the queue length of each port (the smaller the variance, the more balanced the performance); energy consumption is the integral of the power of each motor over time; and the number of timed-out orders is the number of orders that exceed the delivery date commitment. All indicators are obtained from digital twin simulation statistics. The reward function's role is to guide the agent to learn the optimal balance strategy among the five objectives of throughput, waiting time, load balancing, energy consumption, and on-time delivery, avoiding the deterioration of other indicators due to optimization of a single objective.

[0126] Training employs a centralized training-distributed execution (CTDE) paradigm: offline pre-training is performed in a cloud-based global scheduling optimization engine using historical data and a digital twin simulation environment to generate an initial policy network. During deployment, the policy network is distributed to each edge computing node after knowledge distillation, INT8 quantization, and pruning for lightweighting. The compressed model is less than 500KB, and the single inference latency does not exceed 2ms. Each agent makes decisions based solely on local observations, achieving near-global optimum without requiring global real-time communication. During online operation, edge nodes continuously collect experience data, periodically upload it to the cloud for policy fine-tuning, and then distribute updates, forming a continuous learning loop. Simulation results demonstrate that compared to traditional fixed-rule scheduling, the average waiting time is reduced by 45%, and throughput is increased by 35%. The experience data is represented by transition tuples (s, a, r, s'), where s is the agent's local observation at the decision moment (including vehicle distribution, queue length, and adjacent loads), a is the actual scheduling action executed (release / wait / speed change / sorting point allocation), r is the immediate reward calculated by the reward function, and s' is the new local observation after the action is executed. Local observation refers to the fact that each Agent can only perceive the status information within its own jurisdiction, including the vehicle position and speed distribution within the controller's jurisdiction segment, the entry / exit queue length of the segment, and the load status summary of adjacent controllers (periodically broadcast through the 5G gateway, not real-time full data).

[0127] In one specific embodiment, the vehicle transport control method is applied to a garment hanging system and includes the following steps:

[0128] Step 1: Obtain the target order information and the multi-source heterogeneous data corresponding to each carrier in the garment hanging system. The multi-source heterogeneous data is collected by multi-source heterogeneous sensors deployed in the garment hanging system.

[0129] Step 2: Perform data fusion processing on the multi-source heterogeneous data to obtain the current state vector corresponding to each vehicle.

[0130] Step 3: Obtain the geometric topology model corresponding to the garment hanging system.

[0131] Step 4: In the digital twin simulation environment that matches the garment hanging system, path planning is performed based on the geometric topology model and the current state vector of each vehicle to determine the initial optimal bridging path between the grid line and the rack line in the garment hanging transportation system.

[0132] Step 5: Based on the initial optimal bridging path and the target order information, determine the first vehicle transportation path that matches the target order information.

[0133] Step 6: Drive the target vehicle associated with the target order information to transport along the first vehicle transport path.

[0134] Step 7: During the transportation of the target vehicle along the first vehicle transportation path, if the current path efficiency index of the target vehicle is found to be less than the average efficiency threshold of the same order, then the path is deduced based on the target order information and the current state vector of each target vehicle associated with the target order information to determine the second vehicle transportation path.

[0135] Step 8: If the first vehicle transport path coincides with the second vehicle transport path, then the second vehicle transport path is determined as the target vehicle transport path, so as to drive the target vehicle to transport along the target vehicle transport path.

[0136] Step 9: If the first vehicle transport path does not overlap with the second vehicle transport path, then the first vehicle transport path is determined as the target vehicle transport path, so as to drive the target vehicle to transport along the target vehicle transport path.

[0137] In the above embodiments, the current state vector corresponding to each vehicle is obtained by fusing multi-source heterogeneous data. This ensures that the multi-source heterogeneous data is uniformly fused into a single vector, guaranteeing the input consistency for subsequent path planning and dynamic scheduling decisions, and avoiding data fragmentation and information loss caused by independent processing by each sensor. During the process of driving the target vehicle associated with the target order information along the first vehicle transportation path, by identifying whether the current path efficiency index of the target vehicle is less than the average efficiency threshold of the same order, the inefficient transport target vehicle can be accurately located. Then, if the current path efficiency index of the target vehicle is identified as being less than the average efficiency threshold of the same order, path deduction is performed based on the target order information and the current state vectors corresponding to each target vehicle associated with the target order information to determine the second vehicle transportation path. Thus, replanning is only triggered for vehicles whose current path efficiency is significantly lower than the average of the same order group, thereby avoiding... This approach avoids the waste of computational resources caused by indiscriminate global recalculation of all vehicles due to congestion in related technologies, effectively reducing the computational overhead of vehicle transportation control. Simultaneously, by performing path deduction based on the current state vector of each target vehicle, online dynamic correction during transportation is achieved, improving the real-time performance of path control and the dynamic adaptability of the production line. Furthermore, determining the target vehicle transportation path from the first and second vehicle transportation paths to drive the target vehicle along that path ensures that the target vehicle always travels along the optimal path, guaranteeing the stability of vehicle transportation.

[0138] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0139] In one embodiment, a garment hanging system is also provided for performing the vehicle transport control method described in the above embodiments.

[0140] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0141] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0142] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0143] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0144] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A vehicle transportation control method, characterized in that, Applied to garment hanging systems, the method includes: Obtain target order information and a first vehicle transportation route that matches the target order information; During the process of driving the target vehicle associated with the target order information to transport along the first vehicle transport path, if it is identified that the current path efficiency index of the target vehicle is less than the average efficiency threshold of the same order, then the second vehicle transport path is determined based on the target order information and the current state vector of each target vehicle associated with the target order information. A target vehicle transport path is determined from the first vehicle transport path and the second vehicle transport path to drive the target vehicle along the target vehicle transport path.

2. The method according to claim 1, characterized in that, Obtaining the first vehicle transport route matching the target order information includes: Obtain the current state vector corresponding to each carrier in the garment hanging system; Based on the target order information and the current state vector of each vehicle, path planning is performed to determine the first vehicle transportation path that matches the target order information.

3. The method according to claim 2, characterized in that, The step of obtaining the current state vector corresponding to each carrier in the garment hanging system includes: Acquire multi-source heterogeneous data corresponding to each of the aforementioned vehicles; the multi-source heterogeneous data is collected by multi-source heterogeneous sensors deployed in the garment hanging system. The multi-source heterogeneous data is fused to obtain the current state vector corresponding to each of the vehicles.

4. The method according to claim 2, characterized in that, The step of performing path planning based on the target order information and the current state vector corresponding to each of the vehicles, and determining the first vehicle transportation path that matches the target order information, includes: Obtain the geometric topology model corresponding to the garment hanging system; In a digital twin simulation environment that matches the garment hanging system, path planning is performed based on the geometric topology model and the current state vectors of each of the vehicles to determine the initial optimal bridging path between the grid line and the rack line in the garment hanging transportation system. Based on the initial optimal bridging path and the target order information, a first vehicle transportation path matching the target order information is determined.

5. The method according to claim 4, characterized in that, The step of determining a first vehicle transport route matching the target order information based on the initial optimal bridging path and the target order information includes: Based on the target order information, a set of candidate transportation origin and destination stations corresponding to the target order information is determined; the set of candidate transportation origin and destination stations includes at least one candidate sewing pre-exit station and at least one candidate sorting station. Based on the station status corresponding to each of the candidate sewing pre-line exits and each of the candidate sorting ports in the candidate transportation start and end station set, the target sewing pre-line exit and the target sorting port are determined. Based on the initial optimal bridging path, the target sewing pre-exit station, and the target sorting station, a first vehicle transportation path matching the target order information is determined.

6. The method according to claim 1, characterized in that, The method further includes: Determine the current path efficiency index for each of the target vehicles in the transportation process; Based on the mean of each current path efficiency index and a preset deviation factor, the average efficiency threshold of the same production order is determined.

7. The method according to claim 1, characterized in that, The step of determining the second vehicle transportation path based on the target order information and the current state vector of each target vehicle associated with the target order information includes: Obtain the current state vector corresponding to each of the target vehicles associated with the target order information, as well as the set of candidate transportation origin and destination stations corresponding to the target order information; Based on the current state vector corresponding to each of the target vehicles, the station status of each candidate sorting port in the candidate transportation start and end station set is determined. Based on the site status of each candidate sorting point, a set of candidate transportation routes matching the target order information is determined; the set of candidate transportation routes includes at least one candidate transportation route. Based on the current state vector corresponding to the target vehicle and each of the candidate transportation paths, calculate the path cost corresponding to each of the candidate transportation paths. The candidate transportation path with the lowest path cost is determined as the second vehicle transportation path.

8. The method according to claim 1, characterized in that, Determining the target vehicle transportation route from the first vehicle transportation route and the second vehicle transportation route includes: If the first vehicle transport route coincides with the second vehicle transport route, then the second vehicle transport route is determined as the target vehicle transport route; If the first vehicle transport route does not overlap with the second vehicle transport route, then the first vehicle transport route is determined as the target vehicle transport route.

9. A garment hanging system, characterized in that, The garment hanging system is used to perform the vehicle transportation control method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images