Method and device for constructing vehicle operation schedule, processor and electronic equipment

Abstract

The application provides a vehicle operation schedule construction method and device, a processor and an electronic device. The method comprises the following steps: determining work face scheduling information of a target vehicle in a preset working area corresponding to a preset time period; controlling the work face scheduling information to be input into a preset learning model, and obtaining target data output by the preset learning model, wherein the target data is used to represent the stopping condition of the target vehicle at each station in each scheduling cycle in the preset time period, and the stopping condition is any one of the following: stopping or not stopping; and constructing a vehicle operation schedule corresponding to the target vehicle in the preset time period according to the target data. The technical problem that the calculation time of the most efficient path of the transport vehicle is relatively long in the related art is solved, and the technical effects of improving the efficiency and reliability of vehicle scheduling are achieved.

Description

Technical Field

[0001] This application relates to the field of vehicle transportation, and more specifically, to a method, apparatus, processor, and electronic equipment for constructing a vehicle timetable. Background Technology

[0002] The allocation of transport vehicles in coal mines has a direct impact on daily production needs such as shift handover, miners' schedules, vehicle passenger efficiency, and operating costs, making it a crucial aspect of coal mine production. Currently, due to the dispersed nature of mining operations and inconsistent work schedules, mines frequently need to dispatch vehicles to transport workers. In the current daily operations of coal mines, the allocation of transport vehicles is done manually. Waiting personnel call the vehicle allocation staff in the work area, who then manually register the information and issue the transport task to the driver. This method of vehicle allocation is not only time-consuming and labor-intensive but also prone to errors, low vehicle passenger efficiency, and excessively long waiting times for workers.

[0003] In related technologies, operations research optimization methods are mainly used for vehicle transportation, including heuristic algorithms, genetic algorithms, and ant colony algorithms. Heuristic algorithms, which primarily solve vehicle routing optimization problems, determine the vehicle's route between customers with known customer demand locations, aiming to minimize the transportation route or cost. Specifically, the algorithm calculates the optimal cost of traversing all target points from a given vehicle location and multiple target point locations, providing the cost and path description. Ant colony optimization (ACO) is a biomimetic algorithm designed based on the behavior of ants searching for food and finding the shortest path. A group of cooperating ants can find the shortest path between food and their nest, while a single ant cannot. Through extensive and meticulous observation and research, biologists have discovered that the behavior of individual ants interacts and influences each other. During their movement, ants leave behind a substance called pheromones along their paths. This substance is the carrier of information exchange between individual ants. Ants can sense this substance and habitually follow it while crawling. They also release pheromones during crawling. The denser the pheromone trail on a path, the higher the probability that other ants will follow it, thus strengthening the pheromone trail. Therefore, the collective behavior of a large ant colony exhibits a positive feedback phenomenon. The more ants that have traversed a path, the more likely later ants are to choose that path. Individual ants achieve the goal of collaboratively searching for the shortest path through this indirect communication mechanism. The idea behind genetic algorithms is survival of the fittest. Through crossover and mutation operations, diverse solutions are obtained. The optimal solution of the objective function is retained in the next generation, and the next generation of individuals is selected by roulette wheel selection. This process is repeated continuously to optimize the solution to the problem.

[0004] However, current mainstream heuristic algorithms require traversing all target points to find the optimal cost and traversal order, and to calculate the optimal path for each target point, which often results in excessive computation time. Secondly, ant colony algorithms and genetic algorithms are prone to premature convergence in path planning. When obstacles are densely distributed in the working environment, they may fail to plan a reasonable path. In sparse, continuous environments, they are prone to detours and fail to find the optimal path. They are also susceptible to local optima and struggle to guarantee global convergence.

[0005] No better solution has yet been proposed to address the aforementioned problems. Summary of the Invention

[0006] The main objective of this application is to provide a method, apparatus, processor, and electronic device for constructing a vehicle timetable, in order to solve the technical problem that the calculation time for determining the optimal route for transport vehicles is too long in the daily work of coal mines.

[0007] According to one aspect of the present invention, a method for constructing a vehicle timetable is provided, comprising: determining workface scheduling information of a target vehicle in a preset work area within a preset time period, wherein the workface scheduling information includes at least work point information corresponding to multiple work points included in the work area and station information corresponding to multiple stations included in the work area; controlling the input of the workface scheduling information into a preset learning model and obtaining target data output by the preset learning model, wherein the target data is used to characterize the stopping status of the target vehicle at each station in each scheduling cycle within the preset time period, and the stopping status is any one of the following: stopping, not stopping; and constructing a vehicle timetable corresponding to the target vehicle within the preset time period based on the target data.

[0008] Furthermore, before the control face scheduling information is input into the preset learning model and the target data output by the preset learning model is obtained, the method further includes: constructing an initial preset learning model, wherein the initial preset learning model includes a self-learning mask layer; determining multiple training samples, and training the initial preset learning model based on the multiple training samples to obtain the preset learning model.

[0009] Furthermore, multiple training samples are determined, including: acquiring multiple historical workface scheduling information, wherein the historical workface scheduling information is the scheduling information corresponding to the target vehicle within a historical time period in a preset work area; and determining multiple historical workface scheduling information as multiple training samples.

[0010] Furthermore, an initial preset learning model is trained based on multiple training samples, including: determining the data features corresponding to each training sample based on the mask layer to obtain multiple different data features, wherein the training samples and data features correspond one-to-one, and the multiple different data features constitute all the data features corresponding to the training samples; and training the initial preset learning model based on the multiple different data features.

[0011] Furthermore, the control of inputting working face scheduling information into a preset learning model includes: encoding the control of working face scheduling information and obtaining encoded data; and inputting the encoded data into the preset learning model.

[0012] Furthermore, the control work surface scheduling information is input into the preset learning model, and the target data output by the preset learning model is obtained, including: if the target data is that the target vehicle stops at the target station within the target scheduling cycle, the target data is determined as the first identifier, wherein the target station is any station set in the preset work area; if the target data is that the target vehicle stops at the target station within the target scheduling cycle, the target data is determined as the second identifier.

[0013] Furthermore, the scheduling information of the target vehicle in the preset work area within the preset time period is determined, including: determining multiple work points within the preset work area within the preset time period, and determining the number of workers at each work point, the completion time for each work point, the average work speed per worker at each work point, and the planned completion time for each work point; determining the number of workers, completion time, average work speed per worker, and planned completion time for each work point as work point information; determining multiple stations set up within the preset work area, and determining the matching status between each station and multiple work points, the station number, multiple distances between each station and multiple work points, and multiple walking times for each station and multiple work points; determining the matching status, station number, multiple distances, and multiple walking times for each station as station information.

[0014] According to another aspect of the present invention, a vehicle timetable construction apparatus is also provided, comprising: a first determining unit, configured to determine work surface scheduling information of a target vehicle in a preset work area within a preset time period, wherein the work surface scheduling information includes at least work point information corresponding to multiple work points included in the work area and station information corresponding to multiple stations included in the work area; a first controlling unit, configured to control the input of the work surface scheduling information into a preset learning model and obtain target data output by the preset learning model, wherein the target data is used to characterize the stopping status of the target vehicle at each station in each scheduling cycle within the preset time period, and the stopping status is any one of the following: stopping, not stopping; and a first construction unit, configured to construct a vehicle timetable corresponding to the target vehicle within the preset time period based on the target data.

[0015] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein the program executes a method for constructing a vehicle timetable.

[0016] According to another aspect of the present invention, a processor is also provided, which is used to run a program, wherein the program executes a method for constructing a vehicle timetable.

[0017] According to another aspect of the present invention, an electronic device is also provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include a method for performing a vehicle timetable construction method.

[0018] In this embodiment of the invention, by determining the workface scheduling information of the target vehicle in the corresponding work area within a preset time period, wherein the workface scheduling information includes at least the work point information corresponding to multiple work points included in the work area and the station information corresponding to multiple stations included in the work area; controlling the input of the workface scheduling information into a preset learning model, and obtaining the target data output by the preset learning model, wherein the target data is used to characterize the stopping status of the target vehicle at each station in each scheduling cycle within the preset time period, and the stopping status is any one of the following: stopping, not stopping; based on the target data, constructing the vehicle operation timetable corresponding to the target vehicle within the preset time period, solving the technical problem in the related technology that the calculation time for determining the optimal route of transport vehicles in the daily work of coal mines is too long, thereby achieving the technical effect of improving the efficiency and reliability of vehicle scheduling. Attached Figure Description

[0019] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0020] Figure 1 This is a flowchart illustrating a method for constructing a vehicle timetable according to an embodiment of this application; and

[0021] Figure 2 This is a schematic diagram of a vehicle timetable construction device according to an embodiment of this application. Detailed Implementation

[0022] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

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

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] It should be understood that when an element (such as a layer, film, region, or substrate) is described as being "on" another element, the element may be directly on the other element, or there may be an intermediate element present. Furthermore, in the specification and claims, when an element is described as being "connected" to another element, the element may be "directly connected" to the other element, or "connected" to the other element via a third element.

[0026] As mentioned in the background section, the existing technology has the problem that the calculation time for determining the optimal route for transport vehicles in daily coal mine operations is relatively long. In order to solve the above problem, a typical embodiment of this application provides a method for constructing a vehicle timetable.

[0027] According to an embodiment of this application, a method for constructing a vehicle operating timetable is provided.

[0028] Figure 1 This is a flowchart illustrating a method for constructing a vehicle timetable according to an embodiment of this application. Figure 1 As shown, the method includes the following steps:

[0029] Step S101: Determine the work surface scheduling information of the target vehicle in the preset work area within the preset time period. The work surface scheduling information includes at least the work point information corresponding to multiple work points included in the work area and the station information corresponding to multiple stations included in the work area.

[0030] The above-mentioned preset time period is a future time period.

[0031] Step S102: Control the input of workface scheduling information into the preset learning model and obtain the target data output by the preset learning model. The target data is used to characterize the stopping status of the target vehicle at each station in each scheduling cycle within a preset time period. The stopping status is any one of the following: stopping, not stopping.

[0032] Step S103: Based on the target data, construct the vehicle operation timetable corresponding to the target vehicle within the preset time period.

[0033] As mentioned above, due to the different distribution of operations in different work areas, each work area has at least one work point. At the same time, there are also multiple station stations distributed in the work area. By determining the work point information and station information corresponding to the work surface in the future time period, and inputting the work point information and station information into the learning model, the stopping situation of the vehicle at each station in the scheduling cycle is obtained, and then the vehicle operation timetable is constructed.

[0034] The method for constructing vehicle timetables provided in this application transforms the path optimization problem into a time series prediction problem. By finding historical vehicle operation patterns, a learning model is constructed, and information within the working area is input into the learning model. This allows for the rapid generation of vehicle scheduling plans for future time periods, forming a vehicle scheduling table to achieve vehicle arrangement and thus fulfilling the purpose of station allocation.

[0035] In an optional embodiment, before the control workface scheduling information is input into the preset learning model and the target data output by the preset learning model is obtained, the method further includes: constructing an initial preset learning model, wherein the initial preset learning model includes a self-learning mask layer; determining multiple training samples, and training the initial preset learning model based on the multiple training samples to obtain the preset learning model. Training the initial preset learning model based on multiple training samples includes: determining the data features corresponding to each training sample based on the mask layer to obtain multiple different data features, wherein the training samples and data features correspond one-to-one, and the multiple different data features constitute all the data features corresponding to the training samples; and training the initial preset learning model based on the multiple different data features.

[0036] As mentioned above, in this application, model training is performed using a Transformer. Its advantage is that the computation of time slice t no longer depends on the computation result at time t-1, thus increasing the model's parallelism. Furthermore, its ability to capture information that is particularly dependent on long-term computations is better than that of RNN-based models. One difference from the traditional Transformer is the addition of a learnable mask layer to the attention mechanism in the encoding part, performing feature selection for each batch of samples. The formula for the mask is:

[0037] M i =SPARSEMAX(P i-1 *H i (A i-1 )), M i ∈R B*N Each sample in each batch has an M. i And M of all samples in each batch i The sum of M equals 1, therefore, M i It is an instance-based mask, Hi is a learnable function, that is, in the learning model provided in this application, a fully connected layer plus a regularization term, and P is the prior probability. During model training, the inputs to the encoding and decoding parts are the same. The decoding part finally uses softmax for binary classification to predict whether each station will be used in the scheduling loop.

[0038] Furthermore, multiple training samples are determined, including: acquiring multiple historical workface scheduling information, wherein the historical workface scheduling information is the scheduling information of the target vehicle within a preset work area during a historical time period; and determining multiple historical workface scheduling information as multiple training samples. This application aims to organize the historical operation data of transport vehicles to construct a learning model; therefore, it is necessary to acquire the scheduling information of vehicles corresponding to multiple historical time periods within a preset work area, specifically including the following steps:

[0039] Step 1: Obtain historical vehicle running time, route, and station information, historical passenger boarding information at each station, and historical operation distribution map. The historical operation distribution map contains the distribution information of operation points and the corresponding operation information for each operation point.

[0040] Step 2: Based on the acquired historical data, calculate the average interval between boarding and alighting at each station in the historical data to determine the bus departure interval. For example, if statistics show that more than 50% of the stations with high utilization rates call for a bus once every hour on average, then use 1 hour as a time interval.

[0041] Step 3: Using the site information and work distribution map, calculate the distance and travel time between each site and the work location;

[0042] Step 4: Analyze the matching information of work locations and stations through historical data, that is, to find the corresponding ride-hailing stations after the work is completed. For example, a worker at work location 1 can take a ride at station 1 or station 2.

[0043] Step 5: Calculate the number of miners working on each working face, the type of working face, and the time taken to complete the working face each day;

[0044] Step 6: Finally, the number of workers on each work surface and the time spent on each work surface are counted in chronological order for each hour of each day. The average work speed is calculated by dividing the two by the number of workers on each work surface.

[0045] Step 7: Obtain a statistical table through steps 2 to 6 above, sort it in chronological order, and resample it at an hourly frequency to obtain historical workface scheduling information.

[0046] In one optional embodiment, the control workface scheduling information is input into a preset learning model, and the target data output by the preset learning model is obtained. This includes: if the target data indicates that the target vehicle is stopping at a target station within the target scheduling cycle, the target data is determined as a first identifier, where the target station is any station set within a preset work area; if the target data indicates that the target vehicle is not stopping at a target station within the target scheduling cycle, the target data is determined as a second identifier. After inputting the workface scheduling information corresponding to a preset time period into the learning model, the learning model outputs the target data. The stopping status at each station within the scheduling cycle is determined based on the output data. In a specific embodiment, if the vehicle stops at the station within the scheduling cycle, the number 1 is output; if it does not stop, the number 0 is output. The output numbers determine whether the vehicle passes through the station without stopping within each scheduling cycle.

[0047] In one optional embodiment, determining the work surface scheduling information of the target vehicle in the preset work area within a preset time period includes: determining multiple work points within the preset work area within the preset time period, and determining the number of workers corresponding to each work point, the work completion time for each work point, the average work speed per person at each work point, and the planned completion time for each work point; determining the number of workers, work completion time, average work speed per person, and planned completion time for each work point as work point information; determining multiple stations set up within the preset work area, and determining the matching status of each station with multiple work points, the station number, multiple distances between each station and multiple work points, and multiple walking times for each station and multiple work points; determining the matching status, station number, multiple distances, and multiple walking times for each station as station information.

[0048] It should be noted that in actual use, the required change cycle is relatively fixed. Therefore, if the production schedule is on a weekly basis, the information corresponding to the daily production plan is used as input data for the learning model, which is then fed into the pre-trained model to predict the station usage for each day. This prediction of the week's usage is then used to calculate the average station usage for each day of the week, which forms the train timetable for that week.

[0049] In the vehicle operation timetable construction method provided in this application, after training a better vehicle configuration learning model through artificial intelligence technology, once a version of the learning model has been trained, if new data (i.e., when a new work order appears), the trained model can be fine-tuned, which greatly speeds up the model solution speed and achieves the technical effect of improving the efficiency and reliability of vehicle scheduling.

[0050] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0051] This application also provides an apparatus for constructing a vehicle timetable. It should be noted that this apparatus can be used to execute the method for constructing a vehicle timetable provided in this application. The following describes the apparatus for constructing a vehicle timetable provided in this application.

[0052] Figure 2 This is a schematic diagram of a vehicle timetable construction apparatus according to an embodiment of this application. Figure 2As shown, the device includes: a first determining unit 201, used to determine the work surface scheduling information of the target vehicle in the preset work area within a preset time period, wherein the work surface scheduling information includes at least the work point information corresponding to multiple work points included in the work area and the station information corresponding to multiple stations included in the work area; a first controlling unit 202, used to control the input of the work surface scheduling information into the preset learning model and obtain the target data output by the preset learning model, wherein the target data is used to characterize the stopping status of the target vehicle at each station in each scheduling cycle within the preset time period, and the stopping status is any one of the following: stopping, not stopping; and a first constructing unit 203, used to construct the vehicle operation timetable corresponding to the target vehicle within the preset time period based on the target data.

[0053] In an optional embodiment, the apparatus further includes: a second construction unit, configured to construct an initial preset learning model before the control work surface scheduling information is input into the preset learning model and the target data output by the preset learning model is obtained, wherein the initial preset learning model includes a self-learning mask layer; and a second determination unit, configured to determine multiple training samples and train the initial preset learning model based on the multiple training samples to obtain the preset learning model.

[0054] In an optional embodiment, the second determining unit includes: an acquisition subunit, configured to acquire multiple historical workface scheduling information, wherein the historical workface scheduling information is the scheduling information corresponding to a target vehicle within a preset work area in a historical time period; and a first determining subunit, configured to determine that the multiple historical workface scheduling information are multiple training samples.

[0055] In an optional embodiment, the second determining unit includes: a second determining subunit, configured to determine the data features corresponding to each training sample based on the mask layer to obtain multiple different data features, wherein the training samples and data features correspond one-to-one, and the multiple different data features constitute all the data features corresponding to the training samples; and a training subunit, configured to train an initial preset learning model based on the multiple different data features.

[0056] In one optional embodiment, the first control unit 202 includes: a first control subunit for controlling the encoding of work surface scheduling information and obtaining encoded data; and a second control subunit for controlling the input of encoded data into a preset learning model.

[0057] In an optional embodiment, the first control unit 202 includes: a third determining subunit, configured to, when the target data is that the target vehicle is stopped at the target station within the target scheduling cycle, the target data is a first identifier, wherein the target station is any station set within a preset working area; and a fourth determining subunit, configured to, when the target data is that the target vehicle is not stopped at the target station within the target scheduling cycle, the target data is a second identifier.

[0058] In an optional embodiment, the first determining unit 201 includes: a fifth determining subunit, configured to determine multiple work points within a preset work area and within a preset time period, and determine the number of workers corresponding to each work point, the work completion time for each work point, the average work speed per worker at each work point, and the planned completion time for each work point; a sixth determining subunit, configured to determine the number of workers, work completion time, average work speed per worker, and planned completion time for each work point as work point information; and a seventh determining subunit, configured to determine multiple stations set within the preset work area, and determine the matching status between each station and multiple work points, the station number, multiple distances between each station and multiple work points, and multiple walking times for each station and multiple work points; and to determine the matching status, station number, multiple distances, and multiple walking times for each station as station information.

[0059] An apparatus for constructing a vehicle timetable includes a processor and a memory. The aforementioned first determining unit 201 and others are stored in the memory as program units, and the processor executes the aforementioned program units stored in the memory to realize the corresponding functions.

[0060] The processor contains a kernel, which retrieves the corresponding program units from memory. One or more kernels can be configured, and adjusting kernel parameters can address the technical problem of long computation times in determining the optimal route for transport vehicles during daily operations in coal mines.

[0061] The memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0062] This invention provides a computer-readable storage medium storing a program that, when executed by a processor, implements a method for constructing a vehicle timetable.

[0063] This invention provides a processor for running a program, wherein the program executes a method for constructing a vehicle timetable.

[0064] This invention provides a device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it performs at least the following steps: determining work surface scheduling information for a target vehicle within a preset time period in a preset work area, wherein the work surface scheduling information includes at least work point information corresponding to multiple work points within the work area and station information corresponding to multiple stations within the work area; controlling the input of the work surface scheduling information into a preset learning model and obtaining target data output by the preset learning model, wherein the target data characterizes the stopping status of the target vehicle at each station in each scheduling cycle within the preset time period, and the stopping status is any one of the following: stopping, not stopping; and constructing a vehicle operation timetable corresponding to the target vehicle within the preset time period based on the target data.

[0065] Furthermore, before the control face scheduling information is input into the preset learning model and the target data output by the preset learning model is obtained, the method further includes: constructing an initial preset learning model, wherein the initial preset learning model includes a self-learning mask layer; determining multiple training samples, and training the initial preset learning model based on the multiple training samples to obtain the preset learning model.

[0066] Furthermore, multiple training samples are determined, including: acquiring multiple historical workface scheduling information, wherein the historical workface scheduling information is the scheduling information corresponding to the target vehicle within a historical time period in a preset work area; and determining multiple historical workface scheduling information as multiple training samples.

[0067] Furthermore, an initial preset learning model is trained based on multiple training samples, including: determining the data features corresponding to each training sample based on the mask layer to obtain multiple different data features, wherein the training samples and data features correspond one-to-one, and the multiple different data features constitute all the data features corresponding to the training samples; and training the initial preset learning model based on the multiple different data features.

[0068] Furthermore, the control of inputting working face scheduling information into a preset learning model includes: encoding the control of working face scheduling information and obtaining encoded data; and inputting the encoded data into the preset learning model.

[0069] Furthermore, the control work surface scheduling information is input into the preset learning model, and the target data output by the preset learning model is obtained, including: if the target data is that the target vehicle stops at the target station within the target scheduling cycle, the target data is determined as the first identifier, wherein the target station is any station set in the preset work area; if the target data is that the target vehicle stops at the target station within the target scheduling cycle, the target data is determined as the second identifier.

[0070] Furthermore, the scheduling information of the target vehicle in the preset work area within the preset time period is determined, including: identifying multiple work points within the preset work area within the preset time period, and determining the number of workers at each work point, the completion time for each work point, the average work speed per worker at each work point, and the planned completion time for each work point; determining the number of workers, completion time, average work speed per worker, and planned completion time for each work point as work point information; identifying multiple stations set up within the preset work area, and determining the matching status of each station with multiple work points, the station number, multiple distances between each station and multiple work points, and multiple walking times for each station and multiple work points; determining the matching status, station number, multiple distances, and multiple walking times for each station as station information. The devices mentioned in this document can be servers, PCs, tablets, mobile phones, etc.

[0071] This application also provides a computer program product, which, when executed on a data processing device, is suitable for executing an initialization program having at least the following method steps: determining the work surface scheduling information of the target vehicle in a preset work area within a preset time period, wherein the work surface scheduling information includes at least the work point information corresponding to multiple work points included in the work area and the station information corresponding to multiple stations included in the work area; controlling the input of the work surface scheduling information into a preset learning model and obtaining the target data output by the preset learning model, wherein the target data is used to characterize the stopping status of the target vehicle at each station in each scheduling cycle within the preset time period, and the stopping status is any one of the following: stopping, not stopping; constructing a vehicle operation timetable corresponding to the target vehicle within the preset time period based on the target data.

[0072] Furthermore, before the control face scheduling information is input into the preset learning model and the target data output by the preset learning model is obtained, the method further includes: constructing an initial preset learning model, wherein the initial preset learning model includes a self-learning mask layer; determining multiple training samples, and training the initial preset learning model based on the multiple training samples to obtain the preset learning model.

[0073] Furthermore, multiple training samples are determined, including: acquiring multiple historical workface scheduling information, wherein the historical workface scheduling information is the scheduling information corresponding to the target vehicle within a historical time period in a preset work area; and determining multiple historical workface scheduling information as multiple training samples.

[0074] Furthermore, an initial preset learning model is trained based on multiple training samples, including: determining the data features corresponding to each training sample based on the mask layer to obtain multiple different data features, wherein the training samples and data features correspond one-to-one, and the multiple different data features constitute all the data features corresponding to the training samples; and training the initial preset learning model based on the multiple different data features.

[0075] Furthermore, the control of inputting working face scheduling information into a preset learning model includes: encoding the control of working face scheduling information and obtaining encoded data; and inputting the encoded data into the preset learning model.

[0076] Furthermore, the control work surface scheduling information is input into the preset learning model, and the target data output by the preset learning model is obtained, including: if the target data is that the target vehicle stops at the target station within the target scheduling cycle, the target data is determined as the first identifier, wherein the target station is any station set in the preset work area; if the target data is that the target vehicle stops at the target station within the target scheduling cycle, the target data is determined as the second identifier.

[0077] Furthermore, the scheduling information of the target vehicle in the preset work area within the preset time period is determined, including: determining multiple work points within the preset work area within the preset time period, and determining the number of workers at each work point, the completion time for each work point, the average work speed per worker at each work point, and the planned completion time for each work point; determining the number of workers, completion time, average work speed per worker, and planned completion time for each work point as work point information; determining multiple stations set up within the preset work area, and determining the matching status between each station and multiple work points, the station number, multiple distances between each station and multiple work points, and multiple walking times for each station and multiple work points; determining the matching status, station number, multiple distances, and multiple walking times for each station as station information.

[0078] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0079] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0080] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0081] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0082] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0083] As can be seen from the above description, the embodiments of this application achieve the following technical effects:

[0084] 1) In this application, a better vehicle configuration learning model is trained by artificial intelligence technology. After a version of the model is trained, if new data (i.e., when a new work order appears) is available, the trained model can be fine-tuned, which greatly speeds up the model solving speed.

[0085] 2) Add a learnable mask layer to the encoding part of the learning model, so that each training batch has an independent mask. The advantage of this is that the features corresponding to different targets are relatively independent.

[0086] The above are merely preferred embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for constructing a vehicle timetable, characterized in that, include: Determine the work surface scheduling information of the target vehicle in the preset work area within a preset time period, wherein the work surface scheduling information includes at least the work point information corresponding to multiple work points included in the work area and the station information corresponding to multiple stations included in the work area. The workface scheduling information is input into a preset learning model, and the target data output by the preset learning model is obtained. The target data is used to characterize the stopping status of the target vehicle at each station in each scheduling cycle within the preset time period. The stopping status is any one of the following: stopping, not stopping. Based on the target data, construct a vehicle operation schedule for the target vehicle within the preset time period; Determining the work surface scheduling information of the target vehicle in the preset work area within a preset time period includes: determining multiple work points within the preset work area and within the preset time period, and determining the number of workers corresponding to each work point, the work completion time for each work point, the average work speed per person at each work point, and the planned completion time for each work point; determining the number of workers, the work completion time, the average work speed per person, and the planned completion time for each work point as the work point information; determining multiple stations set up within the preset work area, and determining the matching status of each station with the multiple work points, the station number, multiple distances between each station and the multiple work points, and multiple walking times for each station and the multiple work points; determining the matching status, the station number, the multiple distances, and the multiple walking times for each person as the station information.

2. The method according to claim 1, characterized in that, Before controlling the input of the workface scheduling information into the preset learning model and obtaining the target data output by the preset learning model, the method further includes: Construct an initial preset learning model, wherein the initial preset learning model includes a self-learning mask layer; Multiple training samples are identified, and the initial preset learning model is trained based on the multiple training samples to obtain the preset learning model.

3. The method according to claim 2, characterized in that, Multiple training samples were identified, including: Obtain multiple historical workface scheduling information, wherein the historical workface scheduling information is the scheduling information of the target vehicle within a historical time period in the preset work area; Multiple historical working face scheduling information are identified as multiple training samples.

4. The method according to claim 2, characterized in that, Training the initial preset learning model based on multiple training samples includes: Based on the mask layer, the data features corresponding to each training sample are determined to obtain multiple different data features, wherein the training sample and the data features correspond one-to-one, and the multiple different data features constitute all the data features corresponding to the training sample; The initial preset learning model is trained based on multiple different data features.

5. The method according to claim 1, characterized in that, Controlling the input of the workface scheduling information into a preset learning model includes: The control panel scheduling information is encoded to obtain encoded data; The encoded data is controlled to be input into the preset learning model.

6. The method according to claim 1, characterized in that, The control of inputting the workface scheduling information into a preset learning model and obtaining the target data output by the preset learning model includes: If the target data is the parking status of the target vehicle at the target station within the target scheduling cycle, the target data is determined as the first identifier, wherein the target station is any one of the stations set in the preset working area; If the target data indicates that the target vehicle does not stop at the target station within the target scheduling cycle, then the target data is determined to be the second identifier.

7. A device for constructing a vehicle timetable, characterized in that, include: The first determining unit is used to determine the work surface scheduling information of the target vehicle in the preset work area within a preset time period, wherein the work surface scheduling information includes at least the work point information corresponding to multiple work points included in the work area and the station information corresponding to multiple stations included in the work area. The first control unit is used to control the input of the work surface scheduling information into the preset learning model and obtain the target data output by the preset learning model. The target data is used to characterize the stopping status of the target vehicle at each station in each scheduling cycle within the preset time period. The stopping status is any one of the following: stopping, not stopping. The first construction unit is used to construct a vehicle operation timetable corresponding to the target vehicle within the preset time period based on the target data. The first determining unit includes: a fifth determining subunit, used to determine multiple work points corresponding to the preset work area and the preset time period, and to determine the number of workers corresponding to each work point, the work completion time for each work point, the average work speed per person at each work point, and the planned completion time for each work point; a sixth determining subunit, used to determine the number of workers, the work completion time, the average work speed per person, and the planned completion time for each work point as work point information; a seventh determining subunit, used to determine multiple stations set up in the preset work area, and to determine the matching status of each station with multiple work points, the number of each station, multiple distances between each station and multiple work points, and multiple walking times for each station and multiple work points; and an eighth determining subunit, used to determine the matching status, the number, the multiple distances, and the multiple walking times for each person as station information.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program executes a method for constructing a vehicle timetable as described in any one of claims 1 to 6.

9. A processor, characterized in that, The processor is used to run a program, wherein the program executes a method for constructing a vehicle timetable as described in any one of claims 1 to 6.

10. An electronic device, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including a method for performing a vehicle timetable construction method according to any one of claims 1 to 6.

Images