Freight vehicle visual management method and storage medium

Abstract

The invention discloses a freight vehicle visual management method and a storage medium, and relates to the technical field of freight management, and the method comprises the steps: building a digital twin model of vehicle loading, and carrying out the real-time mapping of a carriage structure, a cargo position and the load condition of each region; a weight sensor, bar code scanning and image recognition data are fused, the comprehensive weight of each cargo is calculated by adopting weighted average, and the cargo placement position is dynamically optimized by utilizing a virtual stowage algorithm on the premise of integrating vehicle structure limitation and non-overlapping constraint, so that the overall center of gravity is close to a predetermined target as much as possible; the method comprises the following steps: constructing a vehicle internal space topology model by using a graph neural network, taking each cargo as a node, extracting spatial features through nonlinear mapping, inputting an aggregated feature sequence into a long-short-term memory network, performing time sequence analysis on historical loading data, predicting dynamic change of the gravity center in a few minutes in the future, and determining the dynamic change of the gravity center in the long-short-term memory network. And the conversion from static monitoring to dynamic early warning is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of freight management, and in particular to a visualization management method for freight vehicles and a storage medium. Background Art

[0002] Traditional visualization management systems for freight vehicles mainly rely on sensors and monitoring devices to reflect the vehicle's running status, loading conditions, and environmental information through real-time data collection and intuitive display.

[0003] However, in actual applications, it is difficult to capture the change in the vehicle's center of gravity during the loading and unloading process. For example, during the loading process of the vehicle, due to differences in the size, weight, and placement of the goods, the local load distribution may not be balanced, and this slight imbalance will be masked by the overall data or not monitored at all.

[0004] Although some traditional solutions use manual inspections or set up alarm mechanisms to address the problem of vehicle center-of-gravity offset, during the actual loading process, potential safety risks still cannot be warned in a timely manner, and there is a lag. If not discovered in time, it is likely to cause the vehicle to have a center-of-gravity offset during operation, leading to risks of rollover or vibration during the freight process of the vehicle, affecting the overall stability and safety. Therefore, there is an urgent need for a visualization management solution for freight vehicles to solve such problems. Summary of the Invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] The present invention provides a visualization management method for freight vehicles and a storage medium to solve the problem that when a freight vehicle is loaded, due to the diverse sizes, weights, and placement methods of the goods, the center of gravity often shifts, which may then lead to risks of rollover or vibration during operation. However, traditional systems only display real-time data and lack prediction and guidance for loading balance.

[0007] To solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a visualization management method for freight vehicles, which includes:

[0009] Step S1, establishing a digital twin model of vehicle loading, digitally mapping the carriage structure, cargo position, and load conditions in each area, and synchronously displaying the internal loading status of the vehicle in real time;

[0010] Step S2, using the vehicle loading status data output by S1, combining on-vehicle weight sensors, barcode scanning, and image recognition technologies to collect cargo data in real time, and calculating the distribution of each cargo and the overall center-of-gravity position through a virtual stowage algorithm;

[0011] Step S3: Based on the loading detail data determined in S2, use the graph neural network (GNN) to simulate the internal space relationship of the vehicle, and analyze the historical data in combination with the long short-term memory network (LSTM) to predict the dynamic change trend of the center of gravity in the next few minutes.

[0012] Step S4: According to the prediction result of S3, display the center of gravity distribution and safety level on the visualization platform, and generate warning information through the warning mechanism to notify the staff.

[0013] As a preferred solution of the freight vehicle visualization management method described in the present invention, wherein: the output of step S1 is digital vehicle loading status data.

[0014] As a preferred solution of the freight vehicle visualization management method described in the present invention, wherein: the loading detail data determined in step S2 includes the cargo distribution and the center of gravity position, which are used as the data input for S3.

[0015] As a preferred solution of the freight vehicle visualization management method described in the present invention, wherein: in step S2, the steps of calculating the distribution of each cargo and the overall center of gravity position by the virtual stowage algorithm include:

[0016] Fuse the cargo data from the weight sensor, barcode scanning, and image recognition to calculate the actual weight of each cargo.

[0017] According to the digital twin data of the vehicle loading, use the virtual stowage algorithm to optimize the placement of each cargo and determine the best position of each cargo in the carriage.

[0018] As a preferred solution of the freight vehicle visualization management method described in the present invention, wherein: in step S2, the method of calculating the distribution of each cargo and the overall center of gravity position is:

[0019] Perform cargo data fusion to obtain the comprehensive weight w of the i-th cargo i , and the fusion formula is:

[0020] w i =αa i +βb i +γc i ,

[0021] where α represents the weight sensor data weight, β represents the barcode scanning data weight, γ represents the image recognition data weight, a i represents the sensor reading of the i-th cargo, b i represents the barcode scanning result of the i-th cargo, c i represents the image recognition measurement data of the i-th cargo, and α + β + γ = 1.

[0022] Calculate the optimal position inside the cargo compartment using the formula:

[0023]

[0024] where x i and y i represent the horizontal and vertical positions of the i-th cargo respectively, n is the total number of cargos, and c0 is the predetermined target center-of-gravity position.

[0025] represents the set of cargo compartment positions obtained through optimization under the vehicle structure limit and non-overlapping constraints of the cargo box, minimizing the objective function where the objective function f is used to measure the deviation between the overall center of gravity and the predetermined target center of gravity c0.

[0026] Introduce the constraint conditions:

[0027]

[0028] where represents the two-dimensional position vector of the i-th cargo box in the vehicle coordinate system, represents the two-dimensional position vector of the j-th cargo box in the vehicle coordinate system,

[0029] x min 、x max 、y min 、y max represent the allowable cargo position ranges inside the vehicle respectively, and δ ij represents the minimum safety distance between the i-th and j-th cargos.

[0030] As a preferred solution of the visual management method for a freight vehicle described in the present invention, in step S3, the steps of using the graph neural network GNN to simulate the internal space relationship of the vehicle and combining the long short-term memory network LSTM to analyze historical data to predict the dynamic change trend of the center of gravity in the next few minutes include:

[0031] Based on the loading detail data determined in step S2, use the graph neural network GNN to construct a spatial topology model of the vehicle interior, taking each cargo as a node and the adjacent relationship as an edge to represent the spatial connection between cargos inside the vehicle;

[0032] Perform temporal analysis on the node feature sequence aggregated by GNN through the long short-term memory network LSTM to predict the dynamic change trend of the future overall center of gravity, and trigger the warning mechanism using the prediction result.

[0033] As a preferred solution of the visualization management method for a freight vehicle according to the present invention, wherein: in step S3, the arrangement of goods inside the vehicle is represented by nodes of a graph neural network GNN, and the initial feature representation is:

[0034] f i ′=σ(∑ j∈N(i) W f j +b),

[0035] wherein, f i represents the initial feature of node i, f j represents the initial feature of node j, N(i) is the neighbor set of node i, W represents the weight matrix, b is the bias vector, and σ is the non-linear activation function;

[0036] The aggregated feature sequence is fed into the LSTM model to predict the loading state, and the prediction formula includes:

[0037] h t =LSTM(f t ′,h t-1 ),

[0038] wherein, h t is the hidden state at time t, f t ′ represents the feature aggregated by GNN at time t, and h t-1 is the hidden state of the previous time,

[0039]

[0040] wherein, is the predicted centroid position at the future time t+T, Agg{f i ′} is the aggregation operation on the features of each node, and T is the prediction duration;

[0041] The loss function L is introduced and expressed as:

[0042]

[0043] wherein, L is the prediction loss, and c true represents the actually measured centroid position.

[0044] As a preferred solution of the visualization management method for a freight vehicle according to the present invention, wherein: in step S4, the step of displaying the centroid distribution and safety level on the visualization platform is

[0045] Calculate the deviation between the predicted centroid and the predetermined target centroid c according to the predicted centroid position output in step S3 target , and the calculation formula is:

[0046]

[0047] Where D represents the actual center of gravity deviation,

[0048] Construct the safety index S, expressed as:

[0049] S = exp(-λD),

[0050] Among them, S is the safety index, λ is the adjustment parameter, which reflects the sensitivity of the center of gravity deviation to safety.

[0051] Determine the safety level G:

[0052] If S ≥ θ2, then G = high security,

[0053] If θ1≤S<θ2, then G = medium safety,

[0054] If S<θ1, then G = low security,

[0055] Among them, θ1 and θ2 are low safety and high safety thresholds respectively,

[0056] Generate early warning indicator A, expressed as:

[0057] If S<θ1, then A=1,

[0058] If S ≥ θ1, then A = 0,

[0059] Among them, A is the warning indicator, and 1 means that a warning needs to be triggered.

[0060] In a second aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for visual management of freight vehicles as described in the first aspect of the present invention is implemented.

[0061] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for visual management of freight vehicles as described in the first aspect of the present invention is implemented.

[0062] The beneficial effects of the present invention are as follows: the present invention establishes a digital twin model of vehicle loading, and maps the car body structure, cargo position and load conditions of each area in real time; by fusing weight sensors, barcode scanning and image recognition data, a weighted average is used to calculate the comprehensive weight of each cargo, and a virtual loading algorithm is used to dynamically optimize the cargo placement position under the premise of comprehensive vehicle structure restrictions and non-overlapping constraints, so that the overall center of gravity is as close to the predetermined target as possible, thereby significantly reducing the risk of overloading caused by uneven distribution.

[0063] In the present invention, a topological model of the vehicle interior space is constructed using a graph neural network. Each cargo is regarded as a node, and spatial features are extracted through a non-linear mapping. Then, the aggregated feature sequence is input into a long short-term memory network for time-series analysis of historical loading data to predict the dynamic change of the center of gravity in the next few minutes, realizing the transformation from static monitoring to dynamic warning.

[0064] In the present invention, the center-of-gravity deviation is calculated based on the prediction result of step S3, and a safety index is constructed. The center-of-gravity distribution and safety level are displayed on a visualization platform through an early warning mechanism, enabling timely notification to the staff for adjustment.

[0065] In summary, the present invention significantly improves the real-time monitoring and prediction accuracy of the loading state, effectively reduces the accident risk, and significantly enhances the safety and stability of the operation of freight vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0067] Figure 1 It is a schematic flow chart of the visualization management method for freight vehicles of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0068] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention will be made with reference to the accompanying drawings of the specification.

[0069] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0070] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments.

[0071] Embodiment 1, referring to Figure 1 , this embodiment provides a visualization management method for freight vehicles, including the following steps:

[0072] Step S1, establish a digital twin model of vehicle loading, digitally map the carriage structure, cargo position, and load conditions in each area, and synchronously display the internal loading status of the vehicle in real time;

[0073] Step S1 outputs digital vehicle loading status data;

[0074] Step S2, utilize the vehicle loading status data output by S1, combine on-vehicle weight sensors, barcode scanning, and image recognition technologies to collect cargo data in real time, and calculate the distribution of each cargo and the overall center of gravity position through a virtual stowage algorithm;

[0075] The loading detail data determined in Step S2, including cargo distribution and center of gravity position, serves as the data input for S3;

[0076] In Step S2, the steps of calculating the distribution of each cargo and the overall center of gravity position through the virtual stowage algorithm include:

[0077] Fuse the cargo data from weight sensors, barcode scanning, and image recognition, and calculate the actual weight of each cargo,

[0078] According to the digital twin data of vehicle loading, optimize the placement of each cargo using the virtual stowage algorithm to determine the best position of each cargo in the carriage;

[0079] In Step S2, the method of calculating the distribution of each cargo and the overall center of gravity position is:

[0080] Perform cargo data fusion to obtain the comprehensive weight w of the i-th cargo i , and the fusion formula is:

[0081] w i =αa i +βb i +γc i ,

[0082] where α represents the weight sensor data weight, β represents the barcode scanning data weight, γ represents the image recognition data weight, a i represents the sensor reading of the i-th cargo, b i represents the barcode scanning result of the i-th cargo, c i represents the image recognition measurement data of the i-th cargo, and α + β + γ = 1,

[0083] Calculate the best position of the cargo in the carriage, and the formula is:

[0084]

[0085] where x i and y irepresent the horizontal and vertical positions of the i-th cargo respectively, n is the total number of cargos, and c0 is the predetermined target center of gravity position.

[0086] represents the set of cargo box positions obtained through optimal solution under the conditions of meeting the vehicle structure limitations and the non-overlap constraints of the cargo box, minimizing the objective function where the objective function f is used to measure the deviation between the overall center of gravity and the predetermined target center of gravity c0.

[0087] Introduce the constraint conditions:

[0088]

[0089] where represents the two-dimensional position vector of the i-th cargo box in the vehicle coordinate system, represents the two-dimensional position vector of the j-th cargo box in the vehicle coordinate system,

[0090] x min 、x max 、y min 、y max represent the allowable cargo position ranges in the vehicle respectively, and δ ij represents the minimum safety distance between the i-th and j-th cargos.

[0091] Specifically, effective fusion of different data sources is carried out here. The sensor, barcode, and image recognition data are integrated into the accurate weight information of each cargo through weighted average, and then the virtual stowage algorithm is used to optimize the allocation of the cargo positions in the vehicle. The objective function adopted aims to make the overall center of gravity calculated after all cargos are arranged as close as possible to the predetermined target position, thereby reducing the risk of partial load caused by uneven distribution; constraint conditions are introduced to ensure that the cargo positions in the carriage are legal and non-overlapping.

[0092] Step S3, based on the loading detail data determined in S2, use the graph neural network GNN to simulate the internal space relationship of the vehicle, and combine the long short-term memory network LSTM to analyze the historical data to predict the dynamic change trend of the center of gravity in the next few minutes.

[0093] In step S3, the steps of using the graph neural network GNN to simulate the internal space relationship of the vehicle and combining the long short-term memory network LSTM to analyze the historical data to predict the dynamic change trend of the center of gravity in the next few minutes include:

[0094] Based on the loading detail data determined in step S2, use the graph neural network GNN to construct the spatial topology model of the vehicle interior, taking each cargo as a node and the adjacent relationship as an edge to express the spatial connection between the cargos in the vehicle interior.

[0095] Perform temporal analysis on the node feature sequence obtained by GNN aggregation through the long short-term memory network LSTM, predict the dynamic change trend of the overall center of gravity in the future, and use the prediction result to trigger the early warning mechanism;

[0096] In step S3, the cargo layout inside the vehicle is represented by the nodes of the graph neural network GNN, and the initial feature representation is:

[0097] f i ′ = σ(∑ j∈N(i) W f j + b),

[0098] where f i represents the initial feature of node i, f j represents the initial feature of node j, N(i) is the neighbor set of node i, W represents the weight matrix, b is the bias vector, and σ is the non-linear activation function;

[0099] Send the aggregated feature sequence into the LSTM model to predict the loading state. The prediction formula includes:

[0100] h t = LSTM(f t ′, h t-1 ),

[0101] where h t is the hidden state at time t, f t ′ represents the feature aggregated by GNN at time t, and h t-1 is the hidden state of the previous time;

[0102]

[0103] where, is the predicted center of gravity position at the future time t + T, Agg{f i ′} is the aggregation operation on the features of each node, and T is the prediction duration;

[0104] Introduce the loss function L, which is expressed as:

[0105]

[0106] where L is the prediction loss, and c true represents the actually measured center of gravity position;

[0107] Specifically, in step S3, a graph neural network is introduced to abstract the cargo layout inside the vehicle into a graph structure, fully expressing the spatial connection relationship between each node. After using GNN to perform non-linear mapping on the node features, the aggregated feature sequence is sent into the LSTM model to realize the prediction of the loading state changing with time;

[0108] During the whole process, the spatial features of the goods are first extracted, and then the dynamic patterns in the historical data are captured through the time series network to predict the future center of gravity position. The fusion strategy adopts the method of GNN first and then LSTM, enabling the model to reflect both the static spatial structure and capture the dynamic change trend. The loss function is used to measure the deviation between the prediction result and the actual observation value, guiding the continuous optimization of the model parameters;

[0109] Step S4: According to the prediction result of S3, display the center of gravity distribution and safety level on the visualization platform, and generate warning information through the warning mechanism to notify the staff;

[0110] In step S4, the steps of displaying the center of gravity distribution and safety level on the visualization platform are as follows:

[0111] According to the predicted center of gravity position output by step S3 Calculate the deviation between the predicted center of gravity and the predetermined target center of gravity c target The deviation formula is:

[0112]

[0113] where D represents the actual center of gravity deviation,

[0114] Construct a safety index S, expressed as:

[0115] S = exp(-λD),

[0116] where S is the safety index and λ is the adjustment parameter, reflecting the sensitivity of the center of gravity deviation to safety,

[0117] Determine the safety level G:

[0118] If S ≥ θ2, then G = high safety,

[0119] If θ1 ≤ S < θ2, then G = medium safety,

[0120] If S < θ1, then G = low safety,

[0121] where θ1 and θ2 are the low safety and high safety thresholds respectively,

[0122] Generate a warning index A, expressed as:

[0123] If S < θ1, then A = 1,

[0124] If S ≥ θ1, then A = 0,

[0125] where A is the warning index, and 1 indicates that a warning needs to be triggered;

[0126] Specifically, the deviation between the predicted center of gravity and the predetermined target is calculated here to quantify the unsafe factors of the center of gravity change during the loading process. An exponential function is used to construct a safety index, which can quickly reflect the impact of the center of gravity deviation on the overall safety. The greater the deviation, the faster the safety index drops. In addition, according to the safety index, the safety level is divided through a preset threshold, clearly divided into high, medium, and low safety states, and a binary warning index is used to judge whether the condition for triggering a warning is met, effectively improving the stability and safety of the overall operation of the freight vehicle.

[0127] This embodiment also provides a computer device applicable to the visualization management method of freight vehicles, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement a visualization management method and storage medium of freight vehicles as proposed in the above embodiment.

[0128] This computer device can be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the outer shell of the computer device, or an external keyboard, a touchpad, or a mouse, etc.

[0129] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the freight vehicle visualization management method proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (abbreviated as SRAM), electrically erasable programmable read-only memory (abbreviated as EEPROM), erasable programmable read-only memory (abbreviated as EPROM), programmable read-only memory (abbreviated as PROM), read-only memory (abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0130] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A visualization management method for freight vehicles, characterized in that: including Step S1: Establish a digital twin model of the vehicle load, digitally map the carriage structure, cargo position, and load conditions in each area, and synchronously display the internal load status of the vehicle in real time; Step S2: Utilize the vehicle load status data output by S1, combine on-vehicle weight sensors, barcode scanning, and image recognition technologies to collect cargo data in real time, and calculate the distribution of each cargo and the overall center-of-gravity position through a virtual stowage algorithm; Step S3: Based on the loading detail data determined in S2, use a graph neural network GNN to simulate the internal space relationship of the vehicle, and combine a long short-term memory network LSTM to analyze historical data to predict the dynamic change trend of the center of gravity in the next few minutes; Step S4: According to the prediction result of S3, display the center-of-gravity distribution and safety level on a visualization platform, and generate a warning message through a warning mechanism to notify the staff.

2. The visualization management method for a freight vehicle according to claim 1, wherein: Step S1 outputs digital vehicle load status data.

3. The visualization management method for a freight vehicle according to claim 2, wherein: The loading detail data determined in Step S2 includes cargo distribution and center-of-gravity position, which are used as the data input for S3.

4. The visualization management method for a freight vehicle according to claim 3, characterized in that: In Step S2, the steps of calculating the distribution of each cargo and the overall center-of-gravity position through a virtual stowage algorithm include: Fuse the cargo data from weight sensors, barcode scanning, and image recognition, and calculate the actual weight of each cargo; According to the digital twin data of the vehicle load, use a virtual stowage algorithm to optimize the placement of each cargo and determine the best position of each cargo in the carriage.

5. The visualization management method for a freight vehicle according to claim 4, characterized in that: In Step S2, the method of calculating the distribution of each cargo and the overall center-of-gravity position is: Perform cargo data fusion to obtain the comprehensive weight w of the i-th cargo i , and the fusion formula is: w i = αa i + βb i + γc i , Among them, α represents the weight sensor data weight, β represents the barcode scanning data weight, γ represents the image recognition data weight, a i represents the sensor reading of the i-th cargo, b i represents the barcode scanning result of the i-th cargo, c i represents the image recognition measurement data of the i-th cargo, and α + β + γ = 1, Calculate the best position of the cargo in the carriage, and the formula is: where x i and y i represent the horizontal and vertical positions of the i-th cargo respectively, n is the total number of cargos, and c0 is the predetermined target center of gravity position. It represents the set of cargo box positions obtained through optimization and solution under the conditions of meeting the vehicle structure limitations and the non-overlap constraints of the cargo box, which minimizes the objective function where the objective function f is used to measure the deviation between the overall center of gravity and the predetermined target center of gravity c0; Introduce constraint conditions: Among them, represents the two-dimensional position vector of the i-th cargo box in the vehicle coordinate system, represents the two-dimensional position vector of the j-th cargo box in the vehicle coordinate system, x min 、x max 、y min 、y max respectively represent the allowable range of cargo positions inside the vehicle, and δ ij represents the minimum safety distance between the i-th and j-th cargos.

6. The visualization management method for a freight vehicle according to claim 5, wherein: In Step S3, the steps of using a graph neural network GNN to simulate the internal space relationship of the vehicle and combining a long short-term memory network LSTM to analyze historical data to predict the dynamic change trend of the center of gravity in the next few minutes include: Based on the loading detail data determined in Step S2, use a graph neural network GNN to construct a spatial topology model inside the vehicle, with each cargo as a node and the adjacent relationship as an edge to express the spatial connection between the cargos inside the vehicle; Conduct a time-series analysis on the node feature sequence aggregated by GNN through a long short-term memory network LSTM to predict the dynamic change trend of the future overall center of gravity, and use the prediction result to trigger a warning mechanism.

7. The method for visual management of a freight vehicle according to claim 6, characterized in that: In Step S3, the cargo layout inside the vehicle is represented by the nodes of a graph neural network GNN, and the initial feature representation is: f i ′ = σ(∑ j∈N(i) Wf j + b), Among them, f i represents the initial feature of node i, and f j represents the initial feature of node j. N(i) is the neighbor set of node i, W represents the weight matrix, b is the bias vector, and σ is the non-linear activation function; Send the aggregated feature sequence into the LSTM model to predict the loading status, and the prediction formula includes: h t = LSTM(f t ′ , h t-1 ), where h t is the hidden state at time t, and f t ′ represents the feature after aggregation by GNN at time t, and h t-1 is the hidden state at the previous time Among them, is the center-of-gravity position at the predicted future moment t+T, Agg{f i ′} is the aggregation operation on the features of each node, and T is the prediction duration; Introduce a loss function L, which is expressed as: Among them, L is the prediction loss, and c true represents the actually measured center-of-gravity position.

8. The visualization management method for a freight vehicle according to claim 7, wherein: In Step S4, the step of displaying the center-of-gravity distribution and safety level on a visualization platform is: The predicted center-of-gravity position output according to step S3 Calculate the deviation between the predicted center of gravity and the predetermined target center of gravity c target The deviation calculation formula is as follows: where D represents the actual center-of-gravity deviation; Construct a safety index S, which is expressed as: S = exp(-λD), where S is the safety index and λ is an adjustment parameter, reflecting the sensitivity of the center-of-gravity deviation to safety; Determine the safety level G: If S ≥ θ2, then G = high safety; If θ1 ≤ S < θ2, then G = medium safety; If S < θ1, then G = low safety; where θ1 and θ2 are the low-safety and high-safety thresholds respectively; Generate a warning index A, which is expressed as: If S < θ1, then A = 1, If S ≥ θ1, then A = 0, where A is an early warning indicator, and 1 indicates that an early warning needs to be triggered.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the freight vehicle visualization management 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 the processor, it implements the steps of the freight vehicle visualization management method according to any one of claims 1 to 8.

Images