Data processing method and device for VR interactive training of cross-country container forklift

Abstract

The invention discloses a data processing method and device for VR interactive training of a cross-country container forklift. The method comprises the steps that VR interactive training data information of a user is acquired; the user VR interactive training data information comprises an action data set and an expression data set; processing the action data set to obtain a user action evaluation result; processing the expression data set to obtain a user emotion evaluation result; and according to the user action evaluation result and the user emotion evaluation result, performing personalized training suggestion and training content dynamic adjustment on the VR interactive training system. According to the invention, through efficient acquisition, deep processing and intelligent analysis of multi-dimensional data generated in the training process, the forklift driving operation level and safety awareness of trainees are accurately evaluated, the training result is fed back in real time, and the training strategy is dynamically adjusted, so that the quality and efficiency of the forklift driving VR interactive training are improved, and the training experience is improved. And the actual training cost and the safety risk are reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of virtual reality, in particular to a data processing method and device for off-road container forklift VR interactive training. BACKGROUND

[0002] In modern logistics and industrial production, forklifts are important handling equipment, and their operation standardization and safety are crucial. Traditional forklift driving training has problems such as high training cost, high safety risk, and low training efficiency, while VR interactive training, with its immersive experience and simulation of real working environment, has gradually become a new direction for forklift driving training. During the VR interactive training of forklift driving, a large amount of complex data is generated, such as the steering angle, fork lifting height, and driving speed of the forklift when the trainee operates it, the operation process data such as starting, braking, and gear shifting, and the reaction data and emotional state data of the trainee when facing emergency situations and during the training process. However, the existing data processing method has obvious shortcomings. On the one hand, multi-source data lacks effective integration and analysis, and cannot accurately evaluate the trainee's performance in forklift operation standardization, safety awareness, and emergency handling ability; on the other hand, it cannot provide dynamic feedback and training content adjustment based on the trainee's real-time operation, making it difficult to meet individual training needs, resulting in poor training effectiveness and failing to fully realize the advantages of VR interactive training in forklift driving. SUMMARY

[0003] The technical problem to be solved by the present application is to provide a data processing method and device for off-road container forklift VR interactive training, which efficiently collects, deeply processes, and intelligently analyzes the multi-dimensional data generated during the training process, accurately evaluates the trainee's forklift driving operation level and safety awareness, provides real-time feedback on training results and dynamically adjusts training strategies, thereby improving the quality and efficiency of forklift driving VR interactive training and reducing actual training costs and safety risks.

[0004] To solve the above technical problems, the first aspect of the present application discloses a data processing method for off-road container forklift VR interactive training, comprising: S1, obtaining user VR interactive training data information; the user VR interactive training data information includes a set of action data and a set of expression data; S2, processing the set of action data to obtain a user action evaluation result; S3, processing the set of expression data to obtain a user emotion evaluation result; S4, making individualized training recommendations and dynamically adjusting the training content of the VR interactive training system based on the user action evaluation result and the user emotion evaluation result.

[0005] As an optional implementation, in the first aspect of the present invention, processing the action data set to obtain user action evaluation results includes: S21, Perform feature extraction on the action data set to obtain a set of action feature parameters; S22, Process the action feature set to obtain an action semantic network graph set; S23, process the action semantic network graph set to obtain user action evaluation results.

[0006] As an optional implementation, in the first aspect of the present invention, the step of extracting features from the action data set to obtain an action feature parameter set includes: S211, for any action data in the action data set Divide into, and obtain Data segments , The number of action data points contained in the action data set, where the action data set is... , ; S212, for any data segment Processing is performed to obtain motion feature parameters. ; S213, regarding the action feature parameters The parameters are integrated to obtain a set of motion feature parameters.

[0007] As an optional implementation, in the first aspect of the present invention, the step of processing any data segment... Processing is performed to obtain motion feature parameters. ,include: S2121, for any data segment Processing is performed to obtain the first parameter information. ; The first parameter information The expression is: for Length, , for Time index, for Time index, for The k-th component; S2122, regarding the first parameter information Processing is performed to obtain the second parameter information. ; The second parameter information The expression is: for The k-th component; S2123, regarding the second parameter information Processing is performed to obtain the third parameter information. ; The third parameter information The expression is: for The k-th component; S2124, integrate the first parameter information, the second parameter information, and the third parameter information to obtain action feature parameters. , med indicates that the intermediate value is to be calculated.

[0008] As an optional implementation, in the first aspect of the present invention, processing the action feature set to obtain an action semantic network graph set includes: S221, Extract the encoded vector from the action feature set to obtain the action space vector set; S222, Process the set of action space vectors to obtain a set of action label information; S223, Process the set of action space vectors and the set of action label information to obtain a set of action semantic network graphs.

[0009] As an optional implementation, in the first aspect of the present invention, processing the expression data set to obtain a user emotion evaluation result includes: S31, perform data augmentation on the expression data set to obtain an enhanced expression data set; S32, normalize the enhanced facial expression data set to obtain a normalized facial expression data set; S33, process the normalized expression data set to obtain the user emotion evaluation result.

[0010] As an optional implementation, in the first aspect of the present invention, the step of data augmentation of the expression data set to obtain an augmented expression data set includes: S311, Process the expression data set to obtain an expression tag set; S312, using a preset generator network, the set of facial expression tags and random noise are processed to obtain a generated facial expression image; S313, Process the generated facial expression image and the set of facial expression tags to obtain associated parameter information; S314, Process the associated parameter information to obtain the maximum value of the associated parameter information; S315, The maximum value of the associated parameter information is processed to obtain the first optimized loss function; S316, The generator is trained using the first optimization loss function to obtain an optimized generator; S317, using a preset discriminator, the generated facial expression image, the set of facial expression tags, and the set of facial expression data are processed to obtain a second optimized loss function; S318, The discriminator is trained using the second optimized loss function to obtain an optimized discriminator; S319, using the optimization generator and the optimization discriminator, the expression data set and the expression tag set are processed to obtain an enhanced expression data set.

[0011] A second aspect of this invention discloses a data processing device for VR interactive training of off-road container forklifts, the device comprising: The data acquisition module is used to acquire user VR interaction training data information; the user VR interaction training data information includes a set of action data and a set of facial expression data. The action evaluation module is used to process the action data set to obtain user action evaluation results; The emotion evaluation module is used to process the expression data set to obtain the user's emotion evaluation result; The system optimization and adjustment module is used to provide personalized training suggestions and dynamically adjust the training content of the VR interactive training system based on the user action evaluation results and the user emotion evaluation results.

[0012] As an optional implementation, in the second aspect of the present invention, processing the action data set to obtain user action evaluation results includes: S21, Perform feature extraction on the action data set to obtain a set of action feature parameters; S22, Process the action feature set to obtain an action semantic network graph set; S23, process the action semantic network graph set to obtain user action evaluation results.

[0013] As an optional implementation, in a second aspect of the present invention, the step of extracting features from the action data set to obtain an action feature parameter set includes: S211, for any action data in the action data set Divide into, and obtain m Data segments , N The number of action data points contained in the action data set, where the action data set is... , ; S212, for any data segment Processing is performed to obtain motion feature parameters. ; S213, regarding the action feature parameters The parameters are integrated to obtain a set of motion feature parameters.

[0014] As an optional implementation, in the second aspect of the present invention, the step of processing any data segment... Processing is performed to obtain motion feature parameters. ,include: S2121, for any data segment Processing is performed to obtain the first parameter information. ; The first parameter information The expression is: for Length, , for Time index, for Time index, for The k-th component; S2122, regarding the first parameter information Processing is performed to obtain the second parameter information. ; The second parameter information The expression is: for The k-th component; S2123, regarding the second parameter information Processing is performed to obtain the third parameter information. ; The third parameter information The expression is: for The k-th component; S2124, integrate the first parameter information, the second parameter information, and the third parameter information to obtain action feature parameters. , med indicates that the intermediate value is to be calculated.

[0015] As an optional implementation, in the second aspect of the present invention, the processing of the action feature set to obtain an action semantic network graph set includes: S221, Extract the encoded vector from the action feature set to obtain the action space vector set; S222, Process the set of action space vectors to obtain a set of action label information; S223, Process the set of action space vectors and the set of action label information to obtain a set of action semantic network graphs.

[0016] As an optional implementation, in a second aspect of the present invention, processing the expression data set to obtain a user emotion evaluation result includes: S31, perform data augmentation on the expression data set to obtain an enhanced expression data set; S32, normalize the enhanced facial expression data set to obtain a normalized facial expression data set; S33, process the normalized expression data set to obtain the user emotion evaluation result.

[0017] As an optional implementation, in a second aspect of the present invention, the step of data augmentation of the expression data set to obtain an augmented expression data set includes: S311, Process the expression data set to obtain an expression tag set; S312, using a preset generator network, the set of facial expression tags and random noise are processed to obtain a generated facial expression image; S313, Process the generated facial expression image and the set of facial expression tags to obtain associated parameter information; S314, Process the associated parameter information to obtain the maximum value of the associated parameter information; S315, The maximum value of the associated parameter information is processed to obtain the first optimized loss function; S316, The generator is trained using the first optimization loss function to obtain an optimized generator; S317, using a preset discriminator, the generated facial expression image, the set of facial expression tags, and the set of facial expression data are processed to obtain a second optimized loss function; S318, The discriminator is trained using the second optimized loss function to obtain an optimized discriminator; S319, using the optimization generator and the optimization discriminator, the expression data set and the expression tag set are processed to obtain an enhanced expression data set.

[0018] A third aspect of the present invention discloses another data processing apparatus for VR interactive training, the apparatus comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the data processing method for VR interactive training of off-road container forklifts disclosed in the first aspect of the present invention.

[0019] The fourth aspect of the present invention discloses a computer-storable medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the data processing method for VR interactive training of off-road container forklifts disclosed in the first aspect of the present invention.

[0020] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention improves the quality and efficiency of VR interactive training for forklift driving by efficiently collecting, deeply processing, and intelligently analyzing multi-dimensional data generated during training. It accurately assesses trainees' forklift driving skills and safety awareness, provides real-time feedback on training results, and dynamically adjusts training strategies. This reduces actual training costs and safety risks. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating a data processing method for VR interactive training of off-road container forklifts disclosed in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a data processing device for VR interactive training of off-road container forklifts disclosed in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of another data processing device for VR interactive training of off-road container forklifts disclosed in an embodiment of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0025] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0026] This invention discloses a data processing method and apparatus for VR interactive training of off-road container forklifts. The method includes: acquiring user VR interactive training data information; the user VR interactive training data information includes a set of action data and a set of facial expression data; processing the set of action data to obtain user action evaluation results; processing the set of facial expression data to obtain user emotion evaluation results; and providing personalized training suggestions and dynamically adjusting training content for the VR interactive training system based on the user action evaluation results and the user emotion evaluation results. This invention, through efficient collection, in-depth processing, and intelligent analysis of multi-dimensional data generated during training, accurately assesses trainees' forklift driving operation skills and safety awareness, provides real-time feedback on training results, and dynamically adjusts training strategies, thereby improving the quality and efficiency of VR interactive training for forklift driving and reducing actual training costs and safety risks. Detailed explanations follow.

[0027] Example 1 Please see Figure 1 , Figure 1 This is a flowchart illustrating a data processing method for VR interactive training of off-road container forklifts disclosed in an embodiment of the present invention. Figure 1 The data processing method for VR interactive training of off-road container forklifts described herein is applied in the field of virtual reality technology, and this invention is not limited thereto. For example... Figure 1 As shown, the data processing method for VR interactive training of off-road container forklifts may include the following operations: S1, acquire user VR interaction training data information; the user VR interaction training data information includes a set of motion data and a set of facial expression data. S2, process the action data set to obtain user action evaluation results; S3, process the expression data set to obtain the user's emotion evaluation result; S4. Based on the user action evaluation results and the user emotion evaluation results, provide personalized training suggestions and dynamically adjust the training content for the VR interactive training system.

[0028] Optionally, processing the action data set to obtain user action evaluation results includes: S21, Perform feature extraction on the action data set to obtain a set of action feature parameters; S22, Process the action feature set to obtain an action semantic network graph set; S23, process the action semantic network graph set to obtain user action evaluation results, including: The action semantic network graphs in the action semantic network graph set are compared with the standard action graph, and the graph node matching degree and edge weight difference between the two are calculated. The graph node matching degree is calculated by counting the number of key action nodes covered in the action semantic network graph to obtain the completion degree value; Completion score = (Number of matched key action nodes / Total number of key action nodes) × 100% Edge weight difference calculation: The difference in edge weights between the action semantic network graph and the standard action graph; Edge weight difference value = ( ) in, The number of edges, Let be the weight of the i-th edge in the action semantic network graph. Let be the weight of the i-th edge in the standard motion graph; The completion score and edge weight difference are processed to output the user action evaluation result as a quantitative score, such as an action standardization score of 80 points.

[0029] Optionally, the step of extracting features from the action data set to obtain an action feature parameter set includes: S211, for any action data in the action data set Divide into, and obtain m Data segments , N The number of action data points contained in the action data set, where the action data set is... , ; S212, for any data segment Processing is performed to obtain motion feature parameters. ; S213, regarding the action feature parameters The parameters are integrated to obtain a set of motion feature parameters.

[0030] Optionally, the above applies to any data segment Processing is performed to obtain motion feature parameters. ,include: S2121, for any data segment Processing is performed to obtain the first parameter information. ; The first parameter information The expression is: for Length, , for Time index, for Time index, for The k-th component; S2122, regarding the first parameter information Processing is performed to obtain the second parameter information. ; The second parameter information The expression is: for The k-th component; S2123, regarding the second parameter information Processing is performed to obtain the third parameter information. ; The third parameter information The expression is: for The k-th component; S2124, integrate the first parameter information, the second parameter information, and the third parameter information to obtain action feature parameters. , med indicates that the intermediate value is to be calculated.

[0031] Optionally, the processing of the action feature set to obtain an action semantic network graph set includes: S221, Extract the encoded vector from the action feature set to obtain the action space vector set; Using a pre-trained feature extraction model (Transformer encoder), the action features from the action feature set are input into the model, and the feature vectors of the output layer are taken as action space vectors to form a set of action space vectors.

[0032] S222, Process the set of action space vectors to obtain a set of action label information; A fully connected neural network classification model is trained using a pre-defined labeled action dataset (action features + corresponding labels); the action space vector set is input into the trained fully connected neural network classification model, and the corresponding action label information, such as category and attribute, is output. After integration, an action label information set is obtained.

[0033] S223, Process the action space vector set and the action label information set to obtain an action semantic network graph set, including: Using action tags as nodes and semantic relationships as edges; The weights of the edges are the cosine similarity and temporal co-occurrence frequency (representing the order of actions) of the action space vectors. Using the NetworkX tool, action nodes and relation edges (with weights) are combined into a graph structure to obtain a set of action semantic network graphs.

[0034] Optionally, processing the facial expression data set to obtain the user's emotion evaluation result includes: S31, perform data augmentation on the expression data set to obtain an enhanced expression data set; S32, normalize the enhanced facial expression data set to obtain a normalized facial expression data set; S33, process the normalized expression data set to obtain the user emotion evaluation result.

[0035] User emotion ratings include seven categories: happy, sad, surprised, disgusted, fearful, painful, and neutral. In the normalized facial expression dataset, each facial expression data is divided into N frames. N CNNs are used to extract features from each facial expression data. Each CNN layer is connected to a single LSTM layer, and the N LSTM layers above and below are interconnected.

[0036] Each facial expression data point is processed by a CNN network to obtain a feature vector. , take N feature vectors As input, it enters the sequence learning stage and learns to recognize the user's emotion evaluation results with N two-layer LSTM networks.

[0037] Optionally, the step of performing data augmentation on the expression data set to obtain an augmented expression data set includes: S311, Process the expression data set to obtain an expression tag set; S312, using a preset generator network, the set of facial expression tags and random noise are processed to obtain a generated facial expression image; The generator G is fed with random noise, a set of facial expression labels, and the feature templates corresponding to the labels to generate facial expression images; the feature templates corresponding to the labels are semantic features trained in advance using real label images. S313, Process the generated facial expression image and the set of facial expression tags to obtain associated parameter information; The generated facial expression images were processed using CNN to obtain shallow, medium, and deep features. The cyclic cross-correlation coefficients between the shallow, medium, and deep features and the seven labeled images were calculated to obtain three groups of seven coefficients each. The maximum value of each group of coefficients was taken and fused according to the following values: 0.1 for shallow features, 0.3 for medium features, and 0.6 for deep features to obtain the association parameter information. S314, Process the associated parameter information to obtain the maximum value of the associated parameter information; Extract the maximum value of the associated parameter information ; S315, The maximum value of the associated parameter information is processed to obtain the first optimized loss function; The first optimization loss function is: n Indicates the number of images generated; The input generator represents the first... i Zhang noisy image, Indicates that the label is y The i Zhang generates an image. Indicates the discriminator pair The processing results For multi-scale fusion of cross-correlation coefficients, This represents the second-highest value of the multi-scale fusion cross-correlation coefficient. , A value between 0 and 1; S316, The generator is trained using the first optimization loss function to obtain an optimized generator; S317, using a preset discriminator, the generated facial expression image, the set of facial expression tags, and the set of facial expression data are processed to obtain a second optimized loss function; Image pyramids are used to decompose real images, generated expression images, and corresponding labels into feature representations at multiple scales, resulting in large-scale (global), medium-scale (local), and small-scale (detail) features. Image features and label features at the same scale are processed to obtain their cross-correlation coefficients; the cross-correlation coefficients at all scales are then weighted and averaged to obtain the maximum value. The second maximum value is The second optimization loss function is: m This represents the number of real images (images in the facial expression dataset). For the tag y The j A real image, To use a discriminator to The processing and the resulting outcome; S318, The discriminator is trained using the second optimized loss function to obtain an optimized discriminator; S319, using the optimization generator and the optimization discriminator, the expression data set and the expression tag set are processed to obtain an enhanced expression data set.

[0038] As can be seen, this invention improves the quality and efficiency of VR interactive training for forklift driving by efficiently collecting, deeply processing, and intelligently analyzing multi-dimensional data generated during the training process, accurately assessing trainees' forklift driving skills and safety awareness, providing real-time feedback on training results, and dynamically adjusting training strategies, thereby reducing actual training costs and safety risks.

[0039] Example 2 Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of a data processing device for VR interactive training of off-road container forklifts disclosed in an embodiment of the present invention. Figure 2 The data processing device for VR interactive training of off-road container forklifts described herein is applied in the field of virtual reality technology, and the embodiments of this invention are not limited thereto. Figure 2As shown, the data processing device for VR interactive training of off-road container forklifts can include the following operations: S301, Data acquisition module, used to acquire user VR interaction training data information; the user VR interaction training data information includes a set of motion data and a set of facial expression data. S302, Action evaluation module, used to process the action data set to obtain user action evaluation results; S303, Emotion Evaluation Module, used to process the expression data set to obtain user emotion evaluation results; S304, System Optimization and Adjustment Module, used to provide personalized training suggestions and dynamically adjust the training content of the VR interactive training system based on the user action evaluation results and the user emotion evaluation results.

[0040] Example 3 Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of another data processing device for VR interactive training of off-road container forklifts disclosed in an embodiment of the present invention. Figure 3 The data processing device for VR interactive training of off-road container forklifts described herein is applied in the field of virtual reality technology, and the embodiments of this invention are not limited thereto. Figure 3 As shown, the data processing device for VR interactive training of off-road container forklifts can include the following operations: Memory 401 storing executable program code; Processor 402 coupled to memory 401; The processor 402 calls the executable program code stored in the memory 401 to execute the steps in the data processing method for VR interactive training of off-road container forklifts described in Embodiment 1.

[0041] Example 4 This invention discloses a computer-readable storage medium storing a computer program for electronic data interchange, wherein the computer program enables a computer to perform the steps in the data processing method for VR interactive training of off-road container forklifts described in Embodiment 1.

[0042] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0043] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0044] Finally, it should be noted that the data processing method and apparatus for VR interactive training of off-road container forklifts disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data processing method for off-road container forklift VR interaction training, characterized in that, The method comprises: S1, obtaining user VR interaction training data information; the user VR interaction training data information comprises a motion data set and an expression data set; S2, processing the motion data set to obtain a user motion evaluation result; S3, processing the expression data set to obtain a user emotion evaluation result; S4, making personalized training suggestions and dynamically adjusting training content for a VR interaction training system according to the user motion evaluation result and the user emotion evaluation result.

2. The data processing method for the VR interaction training of the off-road container forklift according to claim 1, characterized in that, The processing of the motion data set to obtain a user motion evaluation result comprises: S21, extracting features from the motion data set to obtain a motion feature parameter set; S22, processing the motion feature set to obtain a motion semantic network graph set; S23, processing the motion semantic network graph set to obtain a user motion evaluation result.

3. The data processing method for VR interaction training of off-road container forklifts according to claim 2, characterized in that, The feature extraction from the motion data set to obtain a motion feature parameter set comprises: S211, any action data in the action data set is divided to obtain m data segments , N is the number of action data contained in the action data set, and the action data set is , ; S212, processing any data segment to obtain action feature parameters ; S213, integrating the action feature parameters to obtain an action feature parameter set.

4. The data processing method for VR interaction training of off-road container forklifts according to claim 3, characterized in that, The pair of any data segments Processing to obtain action characteristic parameters , comprising: S2121, processing any data segment to obtain first parameter information ; The first parameter information The expression is: is of length , , is of time index , is of time index , is the kth component in ; S2122, processing the first parameter information to obtain second parameter information ; The second parameter information The expression is: For the kth component in S2123, processing the second parameter information to obtain third parameter information ; The third parameter information The expression is: For the kth component in S2124, integrating the first parameter information, the second parameter information and the third parameter information to obtain an action feature parameter , , med represents finding a median value.

5. The data processing method for VR interaction training of off-road container forklifts according to claim 3, characterized in that, The processing of the motion feature set to obtain a motion semantic network graph set comprises: S221, extracting an encoding vector from the motion feature set to obtain a motion space vector set; S222, processing the motion space vector set to obtain an action label information set; S223, processing the motion space vector set and the action label information set to obtain a motion semantic network graph set.

6. The data processing method for off-road container forklift VR interaction training according to claim 1, characterized in that, The processing of the expression data set to obtain a user emotion evaluation result comprises: S31, performing data enhancement on the expression data set to obtain an enhanced expression data set; S32, normalizing the enhanced expression data set to obtain a normalized expression data set; S33, processing the normalized expression data set to obtain a user emotion evaluation result.

7. The data processing method for off-road container forklift VR interaction training according to claim 6, characterized in that, The data enhancement on the expression data set to obtain an enhanced expression data set comprises: S311, processing the expression data set to obtain an expression label set; S312, using a preset generator network to process the expression label set and random noise to obtain a generated expression image; S313, processing the generated expression image and the expression label set to obtain association parameter information; S314, processing the association parameter information to obtain a maximum value of the association parameter information; S315, processing the maximum value of the association parameter information to obtain a first optimization loss function; S316, using the first optimization loss function to train a generator to obtain an optimized generator; S317, using a preset discriminator to process the generated expression image, the expression label set, and the expression data set to obtain a second optimization loss function; S318, using the second optimization loss function to train the discriminator to obtain an optimized discriminator; S319, using the optimized generator and the optimized discriminator to process the expression data set and the expression label set to obtain an enhanced expression data set.

8. A data processing device for VR interaction training of off-road container forklifts, characterized by The device comprises: The data acquisition module is configured to acquire user VR interaction training data information; the user VR interaction training data information comprises a motion data set and an expression data set; The motion evaluation module is configured to process the motion data set to obtain a user motion evaluation result. The emotion evaluation module is configured to process the expression data set to obtain a user emotion evaluation result. The system optimization adjustment module is configured to make individualized training suggestions and dynamic adjustments of training content for the VR interaction training system according to the user motion evaluation result and the user emotion evaluation result.

9. A data processing device for VR interaction training of off-road container forklifts, characterized by The device comprises: a memory storing executable program codes; a processor coupled with the memory; The processor invokes the executable program codes stored in the memory to execute the data processing method for the off-road container forklift VR interaction training according to any one of claims 1-7.

10. A computer storable medium, characterized by The computer storage medium stores computer instructions, which are invoked to execute the data processing method for the off-road container forklift VR interaction training according to any one of claims 1-7.

Images