A vehicle maintenance simulation model demonstration method and system based on dynamic scheduling

Abstract

This invention provides a method and system for demonstrating a vehicle maintenance simulation model based on dynamic scheduling, belonging to the field of simulation optimization technology. The method includes defining the entire process flow and specific rules for vehicle maintenance; constructing a vehicle maintenance simulation model based on the entire process flow and specific rules; acquiring the current operation behavior of the vehicle maintenance and predicting high-probability subsequent operations based on historical maintenance data; combining a dynamic resource scheduling strategy and a frame rate adaptive strategy, and mobilizing the simulation model to perform a simulation demonstration based on the current operation behavior and the high-probability subsequent operations; collecting feedback data during the simulation demonstration process, and optimizing the dynamic resource scheduling strategy and frame rate adaptive strategy of the simulation model based on the feedback data. This invention solves the problems of poor simulation stability, smoothness, and demonstration effect, making it difficult to meet the needs of full-process, multi-scenario, and high-concurrency automotive maintenance simulation training.

Description

Technical Field

[0001] This invention relates to the field of simulation optimization technology, and more specifically, to a method and system for demonstrating a vehicle maintenance simulation model based on dynamic scheduling. Background Technology

[0002] With the rapid development of virtual reality technology, thanks to its advantages of immersive interaction, high-fidelity scene simulation and realistic visual presentation, it has been widely used in the physical simulation of on-site working environments and in various equipment operation teaching, skills training and fault maintenance training.

[0003] Vehicle maintenance simulation training systems are characterized by covering many procedures, complex component structures, high operational technical requirements, and strict safety regulations. They can realistically display the entire process of vehicle operation, intuitively demonstrate the standardized use of various maintenance tools, and the vehicle malfunctions and safety risks that may be caused by improper operation. They can provide a highly realistic and immersive training experience for vehicle maintenance skills training.

[0004] However, existing vehicle maintenance simulation training systems still have significant shortcomings: On the one hand, the simulation models are not differentiated and layered according to the precision and process requirements of maintenance operations. In multi-component, full-process simulation scenarios, the large amount of data and redundant loading can easily cause system lag, large frame rate fluctuations, and process switching delays, seriously affecting the smoothness and user experience of the training. On the other hand, the systems mostly operate in a passive response manner, unable to predict subsequent high-probability processes based on historical maintenance data and preload resources in advance. They also lack a dynamic resource scheduling mechanism that combines operational precision, operational complexity, and real-time system status, resulting in unreasonable allocation of computing resources and difficulty in balancing the simulation requirements of high-precision maintenance operations with system operating efficiency. In addition, existing systems cover limited processes, making it difficult to achieve integrated simulation of the entire process, and lack a strategy iteration and optimization mechanism based on training feedback data. This makes it difficult to continuously improve the simulation stability, adaptability, and demonstration effect, hindering the efficient implementation of full-process vehicle maintenance simulation training. Summary of the Invention

[0005] The purpose of this invention is to provide a vehicle maintenance simulation model demonstration method and system based on dynamic scheduling, so as to improve the above-mentioned problems. To achieve the above objective, the technical solution adopted by this invention is as follows:

[0006] Firstly, this application provides a method for demonstrating a vehicle maintenance simulation model based on dynamic scheduling, including:

[0007] Define the entire process flow and specific rules for vehicle maintenance. The entire process flow includes at least disassembly and inspection, fault diagnosis, component repair, assembly and acceptance procedures. The specific rules include operation accuracy rules, fault association rules and process coordination rules.

[0008] A simulation model for vehicle maintenance is constructed based on the entire process flow and the specific rules.

[0009] Acquire the current operation behavior of vehicle maintenance and predict high-probability subsequent procedures based on historical maintenance data;

[0010] A simulation demonstration is conducted based on the current operation behavior and the high-probability subsequent process mobilization simulation model, combining dynamic resource scheduling strategy and frame rate adaptive strategy.

[0011] The simulation demonstration process collects feedback data, and the dynamic resource scheduling strategy and frame rate adaptive strategy of the simulation model are optimized based on the feedback data.

[0012] Secondly, this application also provides a vehicle maintenance simulation model demonstration system based on dynamic scheduling, including:

[0013] The definition module is used to define the entire process flow and specific rules for vehicle maintenance. The entire process flow includes at least disassembly and inspection, fault diagnosis, component repair, assembly and acceptance procedures. The specific rules include operation accuracy rules, fault association rules and process coordination rules.

[0014] A construction module is used to build a simulation model of vehicle maintenance based on the full process flow and the exclusive rules;

[0015] The acquisition module is used to acquire the current operation behavior of vehicle maintenance and predict high-probability subsequent procedures based on historical maintenance data.

[0016] The simulation module is used to combine dynamic resource scheduling strategy and frame rate adaptive strategy to perform simulation demonstration based on the current operation behavior and the high probability subsequent process mobilization simulation model;

[0017] The feedback module is used to collect feedback data during the simulation demonstration process and optimize the dynamic resource scheduling strategy and frame rate adaptive strategy of the simulation model based on the feedback data.

[0018] The beneficial effects of this invention are as follows: This invention achieves efficient, stable, and accurate vehicle maintenance simulation through a comprehensive design encompassing full-process flow definition, simulation model construction, process prediction, dynamic resource scheduling, frame rate adaptation, and feedback optimization. On one hand, by layering simulation data according to operational precision and splitting process into independent modules, combined with targeted lightweight optimization, the invention effectively solves the problems of data redundancy and loading lag in traditional simulation models. Simultaneously, based on component-level process prediction using historical maintenance data, it achieves high-probability preloading of core resources for subsequent processes, significantly reducing process switching latency and improving the smoothness of simulation demonstrations and the training experience. On the other hand, it constructs a dynamic resource scheduling and frame rate adaptation strategy that combines operational precision weights, operational complexity, and process correlation, supporting differentiated resource allocation at the module and operational levels in multi-operation parallel scenarios. This achieves efficient utilization of system computing resources, balancing the simulation requirements of high-precision maintenance operations with overall system operating efficiency.

[0019] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram illustrating the vehicle maintenance simulation model demonstration method based on dynamic scheduling as described in this embodiment of the invention.

[0022] Figure 2 This is a schematic diagram of the vehicle maintenance simulation model demonstration equipment based on dynamic scheduling as described in this embodiment of the invention.

[0023] The diagram is labeled as follows: 800, Vehicle maintenance simulation model demonstration equipment based on dynamic scheduling; 801, Processor; 802, Memory; 803, Multimedia component; 804, I / O interface; 805, Communication component. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0025] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0026] Example 1:

[0027] This embodiment provides a method for demonstrating a vehicle maintenance simulation model based on dynamic scheduling.

[0028] It is understood that the demonstration method of this embodiment can be applied to scenarios such as vehicle maintenance VR / virtual simulation training systems, vehicle maintenance teaching and assessment platforms, vehicle fault simulation demonstration systems, and intelligent maintenance training cabins.

[0029] In practical applications such as large-scale vehicle repair training bases, vehicle repair professional education, and pre-employment skills training for new employees, virtual maintenance scenarios encompassing the entire process of disassembly, fault diagnosis, component repair, assembly, and acceptance can be constructed to achieve high-precision simulation operations and process demonstrations of key components such as engines, chassis, and electrical systems. The simulation system maintains stable and efficient operation even under long-term, high-intensity, and multi-student parallel use, making it suitable for various application needs such as standardized vehicle repair teaching, skills training, operational assessments, safety demonstrations, and fault simulations, effectively improving the realism, safety, and teaching efficiency of vehicle repair training.

[0030] See Figure 1 The figure shows that the method includes steps S1, S2, S3, S4 and S5.

[0031] Step S1: Define the entire process flow and specific rules for vehicle maintenance. The entire process flow includes at least disassembly and inspection, fault diagnosis, component repair, assembly and acceptance procedures. The specific rules include operation accuracy rules, fault association rules and process coordination rules.

[0032] In this step, based on the actual vehicle repair process, the repair process is divided into five standard procedures: disassembly and inspection, fault diagnosis, component repair, assembly, and acceptance. The execution sequence, work content, operation objects, and process requirements of each procedure are clearly defined, forming a structured full-process flow.

[0033] Based on the process requirements of different maintenance operations, corresponding accuracy levels, allowable error ranges, and simulation display accuracy standards are set for each process and operation type, forming operation accuracy rules. Among them, high-precision operations correspond to high simulation accuracy and high image rendering requirements, while ordinary operations correspond to basic simulation accuracy and lightweight rendering requirements.

[0034] Secondly, establish the correspondence between vehicle components, fault phenomena, fault causes, and repair methods, clarify the triggering conditions, manifestations, and processing logic of different faults in the disassembly, diagnosis, repair, and assembly stages, and form fault association rules.

[0035] Then, define the sequential constraints, dependencies, switching conditions, and collaborative execution logic between each process, clarify the impact of the completion status of the preceding process on the start of the subsequent process, and the collaborative constraints when multiple operations are performed in parallel, thus forming process collaboration rules.

[0036] Step S2: Construct a simulation model for vehicle maintenance based on the entire process flow and the specific rules;

[0037] Step S2 includes:

[0038] Step S21: Perform three-layer layering processing on the simulation basic data of the simulation model according to the accuracy of vehicle maintenance operation to obtain multiple data layers, including a high-precision detail layer, a dynamic interaction layer and a static basic layer.

[0039] It should be noted that the simulation model in this embodiment is actually a vehicle repair workshop model. When constructing the simulation model, it is necessary to clearly define the scene characteristics and simulation objectives of the vehicle repair workshop, comprehensively covering the types of vehicles to be repaired, repair processes, training time arrangements, and human resource allocation. Based on this, simulation parameters matching the vehicle repair workshop are set, and standardized repair processes are developed for the actual operating mode of the vehicle repair workshop. This allows the simulation demonstration to be presented step-by-step in the order of vehicle entry, disassembly and inspection, fault diagnosis, component repair, assembly, and acceptance. Based on the actual job configuration of a repair workshop, the personnel configuration before the vehicle enters for repair is simulated, including the number of repair personnel, job responsibilities, operating permissions, and collaborative working relationships.

[0040] Before building the simulation model, it is necessary to collect and organize the data required for the simulation model demonstration, including vehicle structure data, component size data, equipment operating parameters, maintenance tool information, process constraint information, fault characteristic information, and historical maintenance data, so as to provide data support for the subsequent construction of the simulation model, process flow definition, rule formulation, and implementation of dynamic scheduling strategies.

[0041] In this step, the model meshes, materials, textures, animations, and interactive data of vehicle components, repair tools, and repair environment are classified and organized. According to the precision level of vehicle repair operations, the simulation basic data is divided into a three-layer data structure. The high-precision detail layer contains high-polygon models, fine textures, dimensional parameters, and high-precision assembly constraint data of key precision components, which are used for repair procedures that require precision operation. The dynamic interaction layer contains motion logic, collision bodies, disassembly and assembly paths, and state switching data of detachable, movable, and detectable components, which are used to realize real-time interaction during the repair process. The static basic layer contains the basic models and environmental data of non-interactive and non-precision components such as vehicle body, repair station, and fixed tooling, which are used to maintain the basic display of the simulation scene.

[0042] Step S22: Bind operation precision rules to each data layer separately, and perform targeted lightweight processing on the data layers;

[0043] In this step, the original precision of the high-precision detail layer is preserved, and only redundant vertices and invalid faces are simplified; for the dynamic interaction layer, the number of model faces and rendering overhead are simplified while ensuring the accuracy of interaction; for the static base layer, multi-level LOD optimization, texture compression and model merging are used to reduce system resource consumption.

[0044] Step S23: According to the full process flow, the simulation model is divided into independent modules corresponding to each process;

[0045] In this step, each process module has its own independent loading, unloading, scheduling, and running mechanism.

[0046] Step S24: Adapt each independent module of the process to the data layer corresponding to the accuracy requirements;

[0047] In this step, for high-requirement processes such as precision repair and high-precision assembly, the high-precision detail layer, dynamic interaction layer, and static base layer are loaded and enabled; for medium-requirement processes such as routine disassembly and diagnosis, the dynamic interaction layer and static base layer are loaded; for low-requirement processes such as appearance inspection and workstation switching, only the static base layer is loaded.

[0048] Step S25: Build corresponding 3D simulation logic for each independent module of each process, and bind the fault association rules and the process collaboration rules.

[0049] In this step, fault association rules are bound, enabling the simulation model to trigger corresponding fault displays, fault locations, and fault prompts based on operational behaviors. Simultaneously, process collaboration rules are bound, ensuring that independent modules of each process execute simulation demonstrations according to preset sequences, dependencies, and switching conditions, guaranteeing that the maintenance process conforms to actual process specifications.

[0050] It should be noted that the simulation model includes corresponding fully 3D virtual scenes for vehicle maintenance, covering multiple locations such as the sides, interior, undercarriage, and roof of the vehicle. Each fully 3D virtual scene has maintenance points, each configured with standard fault types, which can demonstrate faults and provide typical faults and maintenance points. The maintenance process is carried out in a virtual workshop, and all inspection content is completely consistent with the actual vehicle. The complete undercarriage pipeline structure, side bogie structure, and various internal structures can be seen. Corresponding repair methods are formulated based on faults and maintenance points, and the manpower and time costs are referenced to the requirements of an actual maintenance workshop. The simulation demonstration results for different vehicles are not completely the same, and in the event of user error, fault simulation feedback is immediately triggered, and the correct steps are prompted.

[0051] Step S3: Obtain the current operation behavior of vehicle maintenance and predict high-probability subsequent procedures based on historical maintenance data;

[0052] Step S3 includes:

[0053] Step S31: Collect the user's vehicle maintenance operation instructions, operation objects, and operation execution nodes during the simulation demonstration, as the current operation behavior of vehicle maintenance;

[0054] In this step, vehicle maintenance operation commands issued by the user via VR controller, keyboard, or touch device are collected, including but not limited to removing engine spark plugs, calibrating fuel injectors, and testing the braking system, and represented by command codes. The vehicle components and maintenance tools corresponding to the current operation are collected as operation objects and represented by object codes. Simultaneously, the progress and current stage of the operation are collected and represented by node codes. The collected operation commands, operation objects, and operation execution nodes are integrated into the current operation behavior.

[0055] Step S32: Obtain historical maintenance data for vehicle maintenance, wherein the historical maintenance data includes the operation execution sequence of each process, the probability of collaborative execution between processes, and the probability of process association caused by operational errors;

[0056] In this step, the historical maintenance data comes from the historical training records of the simulation system, the work logs of real vehicle maintenance workshops, and publicly available vehicle maintenance case data in the industry. After deduplication and standardization, the data is stored in the system database.

[0057] This involves recording the operational sequence of maintenance personnel completing the entire maintenance process for different vehicle models and fault types, and compiling the corresponding operation execution sequences. The probability of collaborative execution between processes is the probability of collaborative execution between any two adjacent processes, i.e., the probability of continuing to execute a subsequent process after completing a preceding process. The probability of process association caused by operational errors is the probability that an additional process needs to be executed due to an operational error within a given process, such as an operational error in the disassembly and inspection process leading to the need to perform a component replacement process.

[0058] Step S33: Construct a process prediction probability matrix based on the historical maintenance data. The process prediction probability matrix includes the subsequent processes corresponding to the current process and the trigger probability value of each subsequent process.

[0059] It should be noted that the vehicle maintenance process in this embodiment consists of 5 categories (disassembly and inspection, fault diagnosis, component repair, assembly, and acceptance). However, the vehicle interior contains a large number of maintenance components such as the engine, chassis, braking system, and electrical system. Therefore, when constructing the process prediction probability matrix, it is not based solely on the process, but rather on the current process and the currently operated component as a joint state, performing probability statistics and matrix construction to make the prediction results more closely match the actual maintenance behavior.

[0060] Specifically, the process and operating component involved in the current operation are combined into a joint state. , ,in, Indicates the process index. Indicates the index of the operating component.

[0061] Statistics from historical maintenance data and ,in, Indicates a joint state Next, the process will be transferred to the next step. Number of times, Indicates a joint state Total number of occurrences.

[0062] Calculate the component-level process transfer probability at the component level. :

[0063] ;

[0064] In the formula, This represents the component-level process transfer probability, i.e., the probability of process transfer. and operating components Next, we will proceed to the subsequent processes. The trigger probability, Indicates a joint state Next, the process will be transferred to the next step. Number of times, Indicates a joint state Total number of occurrences.

[0065] Then through Build a joint state as the row, and subsequent processes Given a probability matrix of columns, normalize the probability matrix to obtain the process prediction probability matrix.

[0066] Step S34: Match the current operation behavior with the process prediction probability matrix, and extract the subsequent processes with a trigger probability higher than a preset threshold from the matching results as high-probability subsequent processes.

[0067] In this step, the corresponding joint state is obtained based on the current operation behavior. The component-level process transition probability corresponding to the joint state is found through the process prediction probability matrix. The found component-level process transition probability (i.e. trigger probability) is compared with a preset threshold, and subsequent processes with trigger probabilities greater than the preset threshold are selected as high-probability subsequent processes.

[0068] Step S4: Combining dynamic resource scheduling strategy and frame rate adaptive strategy, a simulation demonstration is performed based on the current operation behavior and the simulation model for high-probability subsequent process mobilization;

[0069] In step S4, the simulation demonstration based on the current operation behavior and the high-probability subsequent process mobilization simulation model includes:

[0070] Step S41: Real-time acquisition of system operating state parameters and automotive repair scenario feature parameters of the simulation model to construct a high-dimensional state vector;

[0071] In this step, system operating status parameters include CPU utilization, remaining CPU cores, GPU memory usage, idle GPU memory, and overall memory usage. Automotive repair scenario characteristic parameters include the current process, operating component, operation precision level, model facet count, and number of interaction points.

[0072] Step S42: Determine the operation precision weight of the current process based on the current operation behavior, and preload core resources based on the high probability of subsequent processes;

[0073] In this step, based on the operation accuracy rules of step S1, operation accuracy weights are assigned to the current process and component. Simultaneously, based on the high-probability subsequent processes obtained in step S3, the models, textures, and interaction logic of the corresponding independent modules are preloaded to complete the preloading of core resources.

[0074] Step S43: Calculate the complexity of the current operation;

[0075] In this step, the complexity is:

[0076] ;

[0077] In the formula, Indicates the complexity of the current operation. Indicates the operation response delay time. This indicates the precision of resource utilization during operation execution.

[0078] Step S44: Combining the high-dimensional state vector, operation precision weights, and pre-loaded core resources, allocate resources for the current operation and high-probability subsequent processes through a dynamic resource scheduling strategy;

[0079] In step S44, the allocation of resources for the current operation and high-probability subsequent processes through a dynamic resource scheduling strategy includes:

[0080] Step S441: Determine the resource requirement level of the current operation based on the operation accuracy weight and complexity;

[0081] In this step, the operation precision weights are... With complexity Weighted, the resource requirement score is obtained. :

[0082] ;

[0083] In the formula, Indicates resource demand score, and All of these represent weighting coefficients. This indicates the precision weight of the current operation. This indicates the complexity of the current operation.

[0084] according to Given the given range, obtain the corresponding resource requirement level.

[0085] Step S442: Extract the number of remaining CPU cores in the current system using a high-dimensional state vector.

[0086] Free GPU memory capacity;

[0087] Step S443: Based on the current number of remaining CPU cores, the capacity of idle GPU memory, and the system resources occupied by preloaded core resources, calculate the total amount of actually allocable idle resources;

[0088] In this step, the total amount of available free resources includes available CPU resources and available GPU resources. The available CPU resources are obtained by subtracting the CPU resources occupied by preloaded core resources from the current remaining CPU cores, and the available GPU resources are obtained by subtracting the GPU resources occupied by preloaded core resources from the available GPU memory capacity.

[0089] It should be noted that the system resources occupied by the preloaded core resources are the video memory and memory resources used to preload data such as models, textures, animations, and interaction logic in order to ensure that subsequent processes can quickly access the required data. These are static data resources.

[0090] Step S444: Allocate matching computing resources for the current operation from the total available resources according to the basic allocation ratio corresponding to the resource demand level of the current operation.

[0091] Understandably, when there are multiple parallel current operations in the system, corresponding to multiple independent modules of the process, in order to achieve differentiated and orderly resource scheduling, the priority of module resource allocation is calculated for each current operation and its associated module:

[0092]

[0093] In the formula, Indicates the first Priority of module resource allocation for each independent module of each process. , and All represent weighting coefficients. Indicates the first The average operation precision weight of all current operation behaviors within an independent module of each process. Indicates the first The average complexity of all current operations within an independent module of each process. Indicates the first The execution frequency of each independent module within a preset time period. Indicates the first Is each process's independent module a high-probability subsequent module?

[0094] After allocating corresponding computing resources to each independent module of the process according to the module resource allocation priority, the computing resources for each current operation behavior in the corresponding independent module of the process are calculated based on the resource requirement level, operation precision weight and complexity of the current operation behavior, and then differentiated allocation is performed.

[0095] Step S445: Based on the high probability of subsequent processes, reserve a preset proportion of idle resources from the remaining idle system resources after allocation;

[0096] In this step, the corresponding preset ratio is obtained by using the component-level process transfer probability corresponding to the high-probability subsequent process. The preset ratio is positively correlated with the component-level process transfer probability, which means that the component-level process transfer probability reflects the correlation between processes.

[0097] Here, reserving a preset proportion of idle resources refers to locking a portion of computing power resources from the remaining CPU computing resources, GPU rendering computing power, and other dynamic operating resources of the system. This portion is specifically reserved for use when there is a high probability of subsequent process switching. It is a dynamic computing power resource used to ensure that there is no competition for computing power, frame rate drop, or running lag when switching processes.

[0098] Step S446: Directly schedule the allocated computing resources to the process-independent module corresponding to the current operation, and mark the reserved idle resources as dedicated pre-allocated resources.

[0099] In this step, once a resource is marked as a dedicated pre-allocated resource, it is prohibited from being used by other modules.

[0100] Step S45: Based on the operation accuracy weight and complexity, adjust the demonstration frame rate of the simulation model through a frame rate adaptive strategy;

[0101] In this step, the formula for calculating the frame rate is demonstrated as follows:

[0102] ;

[0103] In the formula, The display frame rate representing the current operation. Indicates the base frame rate. , and All represent weighting coefficients. This indicates the precision weight of the current operation. Indicates the complexity of the current operation. This indicates the degree of correlation between the overall processes.

[0104] The overall process correlation is calculated by summing the component-level process transition probabilities of all high-probability subsequent processes corresponding to the current operation. It can be understood that the overall process correlation characterizes the overall probability of a process switch occurring after the current operation. A higher overall process correlation indicates a higher likelihood of subsequent maintenance processes following the current operation, placing higher demands on the smoothness of process switching and rendering continuity. Therefore, increasing the demonstration frame rate ensures a smooth, lag-free switching process. A lower overall process correlation indicates a higher likelihood of the current operation continuing, eliminating the need for additional frame rate increases and conserving system resources. When there are no high-probability subsequent processes, the overall process correlation is 0, eliminating the need for additional frame rate increases for process switching. The system determines the demonstration frame rate solely based on the operation's precision weight and complexity, conserving system resources while maintaining simulation effectiveness. Therefore, by introducing the overall process correlation, the frame rate adaptive strategy not only responds to the current operation but also anticipates future process switching needs.

[0105] Step S46: Based on the allocated resources, preloaded core resources, and adjusted demonstration frame rate, invoke the simulation model to perform the simulation demonstration of the current operation behavior, and synchronously respond to the dynamic resource adaptation for process switching.

[0106] In this step, a process-independent module is considered active if it has a current operation; otherwise, it is considered inactive. To prevent inactive process-independent modules from consuming excessive system resources and to ensure efficient resource utilization, a resource release strategy is implemented for inactive process-independent modules. The resource release ratio formula is as follows:

[0107] ;

[0108] In the formula, Indicates the first The resource release ratio of each independent module in each process. This represents the resource release coefficient, used to adjust the release intensity. This indicates the module's activation status; 1 indicates activation, and 0 indicates inactivation. Indicates the first The current resource usage of each independent module in each process. Indicates the first The maximum allowable resource usage for each independent module of a process.

[0109] Understandably, adopting a proportional release strategy, which releases only the excess resources while retaining the resources required for the basic operation of the module, can improve the system's resource utilization and ensure that the module can be quickly activated and respond instantly when it is called again, thus balancing resource conservation and simulation smoothness.

[0110] Step S5: Collect feedback data during the simulation demonstration process, and optimize the dynamic resource scheduling strategy and frame rate adaptive strategy of the simulation model based on the feedback data.

[0111] Step S5 includes:

[0112] Step S51: Collect feedback data during the simulation demonstration process. The feedback data includes process switching delay time, high-precision operation accuracy compliance rate, demonstration frame rate fluctuation value, CPU / GPU resource utilization rate, and preloaded resource hit rate.

[0113] Step S52: Set the compliance threshold for each feedback data, compare the feedback data with the corresponding compliance threshold, and filter out the abnormal data items that do not meet the standard;

[0114] Step S53: Perform root cause analysis on the abnormal data items to obtain the root cause analysis results;

[0115] In this step, if the process switching delay is high, it may be due to insufficient preloading, insufficient reserved resources, or untimely release of resources from inactive modules. If the high-precision operation accuracy rate is low, it may be due to low operation accuracy weight configuration and insufficient resource allocation, resulting in insufficient demonstration frame rate. If the demonstration frame rate fluctuates greatly, it may be due to an unreasonable frame rate adaptation coefficient, causing severe resource contention and untimely release of resources from inactive modules. If the CPU / GPU utilization is too high, it may be due to excessive resource allocation and insufficient release, and an unreasonable module priority strategy. If the CPU / GPU utilization is too low, it may be due to a conservative resource allocation strategy, wasted computing power, or excessive reserved resources. If the preload resource hit rate is low, it may be due to an unreasonable threshold for determining the probability of process transfer at the component level, and the screening conditions for high-probability subsequent processes being too lenient or too strict.

[0116] Step S54: Optimize the dynamic resource scheduling strategy and frame rate adaptive strategy based on the root cause analysis results;

[0117] In this step, corresponding optimizations are performed on different anomaly types based on the root cause analysis results, and the optimized weight parameters, various thresholds, and scheduling rules are updated and obtained.

[0118] Step S55: Synchronize the optimized dynamic resource scheduling strategy and frame rate adaptive strategy to the simulation model.

[0119] This invention, through a multi-dimensional simulation feedback data acquisition and root cause analysis mechanism, can iteratively optimize parameters such as weights and thresholds for resource scheduling and frame rate adaptation based on feedback data such as process switching delay, accuracy achievement rate, and frame rate fluctuation. This allows the simulation model's scheduling strategy to continuously adapt to the needs of actual training scenarios, achieving dynamic improvement in simulation effects. Furthermore, this invention covers the entire process of vehicle inspection, disassembly, diagnosis, repair, assembly, and acceptance, and is adaptable to various application scenarios such as VR simulation training, teaching assessment, and fault simulation. It maintains stable operation even under long-term, high-intensity, and multi-student parallel training scenarios, effectively improving the realism, standardization, and teaching efficiency of vehicle maintenance skills training, and providing technical support for the large-scale implementation of vehicle maintenance simulation training.

[0120] Example 2:

[0121] This embodiment provides a vehicle maintenance simulation model demonstration system based on dynamic scheduling, the system comprising:

[0122] The definition module is used to define the entire process flow and specific rules for vehicle maintenance. The entire process flow includes at least disassembly and inspection, fault diagnosis, component repair, assembly and acceptance procedures. The specific rules include operation accuracy rules, fault association rules and process coordination rules.

[0123] A construction module is used to build a simulation model of vehicle maintenance based on the full process flow and the exclusive rules;

[0124] The acquisition module is used to acquire the current operation behavior of vehicle maintenance and predict high-probability subsequent procedures based on historical maintenance data.

[0125] The simulation module is used to combine dynamic resource scheduling strategy and frame rate adaptive strategy to perform simulation demonstration based on the current operation behavior and the high probability subsequent process mobilization simulation model;

[0126] The feedback module is used to collect feedback data during the simulation demonstration process and optimize the dynamic resource scheduling strategy and frame rate adaptive strategy of the simulation model based on the feedback data.

[0127] The building module includes:

[0128] The layered unit is used to perform three-layer layering processing on the simulation basic data of the simulation model according to the accuracy of vehicle maintenance operations, resulting in multiple data layers, including a high-precision detail layer, a dynamic interaction layer, and a static basic layer.

[0129] The first processing unit is used to bind operation precision rules to each data layer and perform targeted lightweight processing on the data layers.

[0130] The splitting unit is used to split the simulation model into independent modules corresponding to each process according to the entire process flow;

[0131] An adapter unit is used to adapt each independent module of a process to the data layer with corresponding accuracy requirements.

[0132] The second processing unit is used to build corresponding 3D simulation logic for each independent module of the process, and bind the fault association rules and the process collaboration rules.

[0133] The acquisition module includes:

[0134] The data acquisition unit is used to collect the user's vehicle maintenance operation instructions, operation objects, and operation execution nodes during the simulation demonstration, as the current operation behavior of vehicle maintenance;

[0135] The acquisition unit is used to acquire historical maintenance data of the vehicle, which includes the operation execution sequence of each process, the probability of collaborative execution between processes, and the probability of process association caused by operational errors.

[0136] The first construction unit is used to construct a process prediction probability matrix based on the historical maintenance data. The process prediction probability matrix includes the subsequent processes corresponding to the current process and the trigger probability value of each subsequent process.

[0137] The first calculation unit is used to match the current operation behavior with the process prediction probability matrix, and extract the subsequent processes with a trigger probability higher than a preset threshold from the matching results as high-probability subsequent processes.

[0138] The simulation module includes:

[0139] The second building unit is used to collect the system operation state parameters and auto repair scene feature parameters of the simulation model in real time, and to construct a high-dimensional state vector.

[0140] The determination unit is used to determine the operation precision weight of the current process based on the current operation behavior, and to preload core resources based on high-probability subsequent processes;

[0141] The second computational unit is used to calculate the complexity of the current operation.

[0142] The allocation unit is used to combine high-dimensional state vectors, operational precision weights, and pre-loaded core resources to allocate resources for the current operation and high-probability subsequent processes through a dynamic resource scheduling strategy.

[0143] The adjustment unit is used to adjust the demonstration frame rate of the simulation model based on the operation accuracy weight and complexity through a frame rate adaptive strategy.

[0144] The demonstration unit is used to simulate the current operation behavior of the simulation model based on the allocated resources, preloaded core resources and adjusted demonstration frame rate, and synchronously respond to the dynamic resource adaptation for process switching.

[0145] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0146] Example 3:

[0147] Corresponding to the above method embodiments, this embodiment also provides a vehicle maintenance simulation model demonstration device based on dynamic scheduling. The vehicle maintenance simulation model demonstration device based on dynamic scheduling described below and the vehicle maintenance simulation model demonstration method based on dynamic scheduling described above can be referred to in correspondence.

[0148] Figure 2 This is a block diagram illustrating a vehicle maintenance simulation model demonstration device 800 based on dynamic scheduling, according to an exemplary embodiment. Figure 2 As shown, the vehicle maintenance simulation model demonstration device 800 based on dynamic scheduling may include: a processor 801 and a memory 802. The vehicle maintenance simulation model demonstration device 800 based on dynamic scheduling may also include one or more of the following: a multimedia component 803, an I / O interface 804, and a communication component 805.

[0149] The processor 801 controls the overall operation of the vehicle maintenance simulation model demonstration device 800 based on dynamic scheduling to complete all or part of the steps in the aforementioned vehicle maintenance simulation model demonstration method based on dynamic scheduling. The memory 802 stores various types of data to support the operation of the vehicle maintenance simulation model demonstration device 800 based on dynamic scheduling. This data may include, for example, instructions for any application or method operating on the vehicle maintenance simulation model demonstration device 800 based on dynamic scheduling, as well as application-related data, such as contact data, sent and received messages, images, audio, video, etc. The memory 802 can be implemented using any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 802 or transmitted via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. I / O interface 804 provides an interface between processor 801 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 805 is used for wired or wireless communication between the dynamic scheduling-based vehicle maintenance simulation model demonstration device 800 and other devices. Wireless communication methods include Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 805 may include a Wi-Fi module, a Bluetooth module, and an NFC module.

[0150] In an exemplary embodiment, the vehicle maintenance simulation model demonstration device 800 based on dynamic scheduling can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the above-described vehicle maintenance simulation model demonstration method based on dynamic scheduling.

[0151] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided. When executed by a processor, these program instructions implement the steps of the above-described dynamic scheduling-based vehicle maintenance simulation model demonstration method. For example, the computer-readable storage medium may be the memory 802 including the program instructions, which may be executed by the processor 801 of the dynamic scheduling-based vehicle maintenance simulation model demonstration device 800 to complete the above-described dynamic scheduling-based vehicle maintenance simulation model demonstration method.

[0152] Example 4:

[0153] Corresponding to the above method embodiments, this embodiment also provides a readable storage medium. The readable storage medium described below can be referred to in conjunction with the vehicle maintenance simulation model demonstration method based on dynamic scheduling described above.

[0154] A readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the vehicle maintenance simulation model demonstration method based on dynamic scheduling as described in the above method embodiments.

[0155] Specifically, the readable storage medium can be a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or any other readable storage medium capable of storing program code.

[0156] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0157] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A dynamic scheduling-based vehicle inspection simulation model demonstration method, characterized in that, include: Define the entire process flow and specific rules for vehicle maintenance. The entire process flow includes at least disassembly and inspection, fault diagnosis, component repair, assembly and acceptance procedures. The specific rules include operation accuracy rules, fault association rules and process coordination rules. A simulation model for vehicle maintenance is constructed based on the entire process flow and the specific rules. Acquire the current operation behavior of vehicle maintenance and predict high-probability subsequent procedures based on historical maintenance data; A simulation demonstration is conducted based on the current operation behavior and the high-probability subsequent process mobilization simulation model, combining dynamic resource scheduling strategy and frame rate adaptive strategy. The simulation demonstration process collects feedback data, and the dynamic resource scheduling strategy and frame rate adaptive strategy of the simulation model are optimized based on the feedback data.

2. The dynamic scheduling based vehicle inspection simulation model demonstration method according to claim 1, wherein, The simulation model for vehicle maintenance, constructed based on the entire process flow and the specific rules, includes: The simulation model's basic data is processed in three layers according to the precision of vehicle maintenance operations, resulting in multiple data layers, including a high-precision detail layer, a dynamic interaction layer, and a static basic layer. Each data layer is bound to an operation precision rule, and the data layers are subjected to targeted lightweight processing. According to the entire process flow, the simulation model is divided into independent modules corresponding to each process. Each process module is adapted to the corresponding precision requirement of the data layer; Build corresponding 3D simulation logic for each process's independent module, and bind the fault association rules and the process collaboration rules.

3. The dynamic-scheduling-based vehicle inspection simulation model demonstration method according to claim 1, wherein, The process of acquiring the current operation of vehicle maintenance and predicting high-probability subsequent procedures based on historical maintenance data includes: The user's vehicle maintenance operation instructions, operation objects, and operation execution nodes are collected during the simulation demonstration and used as the current operation behavior of vehicle maintenance. Acquire historical maintenance data for vehicle maintenance, which includes the operation execution sequence of each process, the probability of collaborative execution between processes, and the probability of process association caused by operational errors; Based on the historical maintenance data, a process prediction probability matrix is ​​constructed. The process prediction probability matrix includes the subsequent processes corresponding to the current process and the trigger probability value of each subsequent process. The current operation behavior is matched with the process prediction probability matrix, and subsequent processes with a trigger probability higher than a preset threshold are extracted from the matching results and regarded as high-probability subsequent processes.

4. The dynamic-scheduling-based vehicle inspection simulation model demonstration method according to claim 1, wherein, The simulation demonstration based on the current operational behavior and the high-probability subsequent process mobilization simulation model includes: Real-time acquisition of system operation state parameters and automotive repair scenario feature parameters of the simulation model to construct a high-dimensional state vector; The operation precision weight of the current process is determined based on the current operation behavior, and core resources are preloaded based on the high probability of subsequent processes. Calculate the complexity of the current operation. By combining high-dimensional state vectors, operational precision weights, and pre-loaded core resources, a dynamic resource scheduling strategy is used to allocate resources for the current operation and high-probability subsequent processes. Based on operational accuracy weights and complexity, the demonstration frame rate of the simulation model is adjusted through a frame rate adaptive strategy. Based on the allocated resources, preloaded core resources, and adjusted demonstration frame rate, the simulation model is invoked to perform a simulation demonstration of the current operation behavior, and resources are dynamically adapted to respond to process switching.

5. The dynamic-scheduling-based vehicle inspection simulation model demonstration method according to claim 4, characterized in that, The method of allocating resources for the current operation and high-probability subsequent processes through a dynamic resource scheduling strategy includes: The resource requirement level for the current operation is determined based on the operation accuracy weight and complexity. The remaining number of CPU cores and the idle GPU memory capacity of the current system are extracted using a high-dimensional state vector. Based on the current number of remaining CPU cores, the available GPU memory capacity, and the system resources occupied by preloaded core resources, the total amount of available free resources can be calculated. Based on the basic allocation ratio corresponding to the resource demand level of the current operation, allocate matching computing resources from the total amount of idle resources to the current operation. Based on the high probability of subsequent processes, a preset proportion of idle resources are reserved from the remaining idle resources in the system after allocation; The allocated computing resources are scheduled to the independent module of the process corresponding to the current operation, and the reserved idle resources are marked as dedicated pre-allocated resources.

6. The dynamic-schedule-based vehicle inspection simulation model demonstration method according to claim 1, wherein, The process of collecting feedback data during the simulation demonstration and optimizing the dynamic resource scheduling strategy and frame rate adaptive strategy of the simulation model based on the feedback data includes: The simulation demonstration process collects feedback data, including process switching delay time, high-precision operation accuracy compliance rate, demonstration frame rate fluctuation value, CPU / GPU resource utilization rate, and preloaded resource hit rate. Set the compliance threshold for each feedback data, compare the feedback data with the corresponding compliance threshold, and filter out abnormal data items that do not meet the standard; Perform root cause analysis on the outlier data items to obtain the root cause analysis results; Optimize dynamic resource scheduling strategy and frame rate adaptive strategy based on root cause analysis results; The optimized dynamic resource scheduling strategy and frame rate adaptive strategy are synchronized to the simulation model.

7. A dynamic scheduling based vehicle inspection simulation model demonstration system, characterized in that, include: The definition module is used to define the entire process flow and specific rules for vehicle maintenance. The entire process flow includes at least disassembly and inspection, fault diagnosis, component repair, assembly and acceptance procedures. The specific rules include operation accuracy rules, fault association rules and process coordination rules. A construction module is used to build a simulation model of vehicle maintenance based on the full process flow and the exclusive rules; The acquisition module is used to acquire the current operation behavior of vehicle maintenance and predict high-probability subsequent procedures based on historical maintenance data. The simulation module is used to combine dynamic resource scheduling strategy and frame rate adaptive strategy to perform simulation demonstration based on the current operation behavior and the high probability subsequent process mobilization simulation model; The feedback module is used to collect feedback data during the simulation demonstration process and optimize the dynamic resource scheduling strategy and frame rate adaptive strategy of the simulation model based on the feedback data.

8. The dynamic scheduling based vehicle service simulation model demonstration system according to claim 7, characterized in that, The building module includes: The layered unit is used to perform three-layer layering processing on the simulation basic data of the simulation model according to the accuracy of vehicle maintenance operations, resulting in multiple data layers, including a high-precision detail layer, a dynamic interaction layer, and a static basic layer. The first processing unit is used to bind operation precision rules to each data layer and perform targeted lightweight processing on the data layers. The splitting unit is used to split the simulation model into independent modules corresponding to each process according to the entire process flow; An adapter unit is used to adapt each independent module of a process to the data layer with corresponding accuracy requirements. The second processing unit is used to build corresponding 3D simulation logic for each independent module of the process, and bind the fault association rules and the process collaboration rules.

9. The dynamic scheduling based vehicle service simulation model demonstration system according to claim 7, wherein, The acquisition module includes: The data acquisition unit is used to collect the user's vehicle maintenance operation instructions, operation objects, and operation execution nodes during the simulation demonstration, as the current operation behavior of vehicle maintenance; The acquisition unit is used to acquire historical maintenance data of the vehicle, which includes the operation execution sequence of each process, the probability of collaborative execution between processes, and the probability of process association caused by operational errors. The first construction unit is used to construct a process prediction probability matrix based on the historical maintenance data. The process prediction probability matrix includes the subsequent processes corresponding to the current process and the trigger probability value of each subsequent process. The first calculation unit is used to match the current operation behavior with the process prediction probability matrix, and extract the subsequent processes with a trigger probability higher than a preset threshold from the matching results as high-probability subsequent processes.

10. The dynamic scheduling based vehicle service simulation model demonstration system according to claim 7, wherein, The simulation module includes: The second building unit is used to collect the system operation state parameters and auto repair scene feature parameters of the simulation model in real time, and to construct a high-dimensional state vector. The determination unit is used to determine the operation precision weight of the current process based on the current operation behavior, and to preload core resources based on high-probability subsequent processes; The second computational unit is used to calculate the complexity of the current operation. The allocation unit is used to combine high-dimensional state vectors, operational precision weights, and pre-loaded core resources to allocate resources for the current operation and high-probability subsequent processes through a dynamic resource scheduling strategy. The adjustment unit is used to adjust the demonstration frame rate of the simulation model based on the operation accuracy weight and complexity through a frame rate adaptive strategy. The demonstration unit is used to simulate the current operation behavior of the simulation model based on the allocated resources, preloaded core resources and adjusted demonstration frame rate, and synchronously respond to the dynamic resource adaptation for process switching.

Images