Automatic driving public transport intelligent scheduling method, device and equipment and storage medium

Abstract

The invention provides an automatic driving public transport intelligent scheduling method, device and equipment and a storage medium, and is applied to the technical field of intelligent public transport, and the method comprises the steps: obtaining the traffic data of a target one-way route, the target one-way route being an uplink route or a downlink route of a bus driving route; according to the traffic data, constructing description text data with a preset text structure; according to the description text data, obtaining whether a multi-classification result of the vehicle needs to be sent; and according to the multi-classification result, scheduling the to-be-departed shift vehicles in the departure station corresponding to the target one-way route. According to the method, the traffic data of the one-way line is acquired, and the text description and the multi-classification judgment are performed based on the traffic data to obtain the multi-classification result, so that the intelligent scheduling of the automatic driving public transport is realized based on the multi-classification result, the efficiency of the automatic driving public transport is improved, and the robustness of the scheduling method is improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent public transportation technology, and in particular to an autonomous driving public transportation intelligent scheduling method, device, equipment and storage medium. Background Art

[0002] With the acceleration of urbanization, public transportation systems play an increasingly important role in urban transportation. However, traditional public transportation scheduling methods often rely on manual experience and manual scheduling, lacking the ability to respond to real-time traffic conditions, resulting in inefficient public transportation scheduling, unreasonable resource allocation, and poor passenger travel experience, which is particularly unsuitable for autonomous public transportation systems. Summary of the invention

[0003] The technical purpose to be achieved by the embodiments of the present application is to provide an autonomous driving public transportation intelligent scheduling method, device, equipment and storage medium to solve the problems of slow response, low scheduling efficiency and unreasonable resource allocation in the current scheduling method through manual scheduling.

[0004] In order to solve the above technical problems, the embodiment of the present application provides an intelligent dispatching method for public transportation by autonomous driving, including: Acquire traffic data of a target one-way route, where the target one-way route is an up route or a down route of a bus route; Constructing descriptive text data having a preset text structure according to the traffic data, wherein the preset text structure includes weather conditions and road condition reference information, and the road condition reference information includes at least one of station waiting conditions, road congestion information, and vehicle position information; According to the description text data, a multi-classification result of whether the vehicle needs to be dispatched is obtained; According to the multi-classification results, the vehicles to be departed from the departure station corresponding to the target one-way route are dispatched.

[0005] Optionally, in the method as described above, the step of dispatching the vehicles to be departed from the departure station corresponding to the target one-way route according to the multiple classification results includes at least one of the following: If the multi-classification result indicates that a vehicle needs to be dispatched, a first signal is output, wherein the first signal is used to control the vehicle to be dispatched to switch to a working state; If the multi-classification result indicates that the vehicle does not need to be dispatched, a second signal is output, wherein the second signal is used to control the vehicle to be dispatched to maintain a waiting state; If the multi-classification result indicates that human intervention is required, the traffic data and / or the description text data are fed back to the client, and based on the instruction input by the client, the first signal or the second signal is output.

[0006] Specifically, in the method as described above, the step of obtaining the traffic data of the target one-way route includes: Obtain the current first time and the number of bus stops and the number of vehicles in working state in the target one-way route; Determine a city congestion coefficient according to a time range in which the first time is located, wherein if the first time is within a preset time range, determine the city congestion coefficient to be a first threshold, otherwise, determine the city congestion coefficient to be a second threshold, and the second threshold is less than the first threshold; Determining a data collection interval corresponding to the target one-way route according to the number of bus stops, the number of vehicles, and the city congestion coefficient; The traffic data of the target one-way route is acquired according to the data collection interval.

[0007] Preferably, in the above method, obtaining a multi-classification result of whether a vehicle needs to be dispatched based on the description text data includes: Inputting the description text data into a pre-trained target semantic question-answering multi-classification model to obtain the multi-classification result of whether the vehicle needs to be dispatched; The method for obtaining the target semantic question-answering multi-classification model includes: Acquiring the historical traffic data corresponding to at least one one-way route; Constructing the historical traffic data into historical description text data having the preset text structure; Inputting the historical description text data into a Chinese word segmentation encoding model with frozen parameters to obtain first data after word segmentation encoding; Inputting the text data into a bidirectional long short-term memory network to obtain second data having a preset vector structure; Inputting the second data into a fully connected layer to obtain a historical multi-classification result; Based on the historical multi-classification results, adjusting the parameters of gradient descent; The target semantic question-answering multi-classification model is determined according to the parameter information at the end of the training.

[0008] Optionally, the method as described above further comprises: Receive information viewing request input by the client; According to the information viewing request, part or all of the information in the traffic data is fed back to the client.

[0009] Preferably, in the method as described above, the traffic data includes weather information, vehicle operating status information, time information and road condition information, wherein the road condition information includes at least one of vehicle speed information, vehicle position information, platform position information and the number of people waiting at the platform.

[0010] Another embodiment of the present application provides a control device for autonomous driving public transportation intelligent dispatching, comprising: A first data collection module is used to obtain traffic data of a target one-way route, where the target one-way route is an up route or a down route of a bus route; A first data processing module, configured to construct descriptive text data having a preset text structure according to the traffic data, wherein the preset text structure includes weather conditions and road condition reference information, and the road condition reference information includes at least one of station waiting conditions, road congestion information, and vehicle position information; A dispatch algorithm module, used to obtain a multi-classification result of whether a vehicle needs to be dispatched according to the description text data; The scheduling execution module is used to schedule the vehicles to be departed from the departure station corresponding to the target one-way route according to the multi-classification results.

[0011] Yet another embodiment of the present application provides a mobile terminal, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method for intelligent scheduling of public transportation by autonomous driving as described above.

[0012] Another embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for intelligent scheduling of public transportation by autonomous driving are implemented as described above.

[0013] Another embodiment of the present application provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the method for intelligent scheduling of autonomous driving public transportation as described above.

[0014] Compared with the prior art, the method, device, equipment and storage medium for intelligent dispatching of public transportation by autonomous driving provided in the embodiments of the present application have at least the following beneficial effects: The present application obtains traffic data of one-way routes, performs text description and multi-classification judgment based on the traffic data, obtains multi-classification results, and then performs scheduling based on the multi-classification results, thereby realizing intelligent scheduling of autonomous driving public transportation, improving the efficiency of autonomous driving public transportation, and increasing the robustness of the scheduling method. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is one of the flowcharts of the method for intelligent dispatching of public transportation by autonomous driving of the present application; Figure 2 This is the second flow chart of the method for intelligent dispatching of public transportation by autonomous driving of the present application; Figure 3 This is the third flow chart of the method for intelligent dispatching of public transportation by autonomous driving of the present application; Figure 4 This is the fourth flow chart of the method for intelligent dispatching of public transportation by autonomous driving of the present application; Figure 5 This is a schematic diagram of the structure of the control device for intelligent dispatching of autonomous driving public transportation in the present application. DETAILED DESCRIPTION

[0016] In order to make the technical problems, technical solutions and advantages to be solved by the application clearer, the following will be described in detail in conjunction with the accompanying drawings and specific embodiments. In the following description, specific details such as specific configurations and components are provided only to help fully understand the embodiments of the application. Therefore, it should be clear to those skilled in the art that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the application. In addition, for clarity and brevity, the description of known functions and structures has been omitted.

[0017] It should be understood that the references to "one embodiment" or "an embodiment" throughout the specification mean that the specific features, structures, or characteristics associated with the embodiment are included in at least one embodiment of the present application. Therefore, the references to "in one embodiment" or "in an embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0018] In the various embodiments of the present application, it should be understood that the size of the serial numbers of the following processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0019] It should be understood that the term "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0020] In the embodiments provided in this application, it should be understood that "B corresponding to A" means that B is associated with A and B can be determined according to A. However, it should also be understood that determining B according to A does not mean determining B only according to A, but B can also be determined according to A and / or other information.

[0021] See also Figure 1 An embodiment of the present application provides an autonomous driving public transportation intelligent scheduling method, comprising: Step S101, obtaining traffic data of a target one-way route, wherein the target one-way route is an up route or a down route of a bus route; Step S102, constructing description text data with a preset text structure according to the traffic data, wherein the preset text structure includes weather conditions and road condition reference information, and the road condition reference information includes at least one of station waiting conditions, road congestion information, and vehicle position information; Step S103, obtaining a multi-classification result of whether a vehicle needs to be dispatched based on the description text data; Step S104, dispatching the vehicles to be departed from the departure station corresponding to the target one-way route according to the multi-classification results.

[0022] In this embodiment, when performing automatic driving public transportation dispatch, the traffic data of the dispatch object, that is, the target one-way route, will be obtained first, wherein the traffic data is used to reflect the operation status of the buses on the target one-way route. In order to facilitate data processing and data-based judgment, a description text data with a preset text structure will be constructed based on the traffic data first. The description text data can simply and clearly express the operation status of the buses on the target one-way route through the preset text structure, and increase the robustness of the method. According to the description text data, it is possible to obtain a multi-classification result of whether a vehicle needs to be dispatched by judging, and then dispatch the vehicles to be dispatched in the departure station corresponding to the target one-way route according to the multi-classification results, so as to realize the intelligent dispatch and automatic dispatch of the buses.

[0023] In one embodiment, the preset text structure includes weather conditions and road condition reference information, and the road condition reference information includes at least one of the station waiting situation, road congestion information, and vehicle location information, wherein the weather conditions can be used to determine the reference thresholds of other parameters, for example: in one embodiment, when determining the classification results, if the weather conditions are good, the waiting personnel can wait for a long time, and the reference threshold can be appropriately increased (for example: increasing the threshold of the number of waiting people at the station); if the weather conditions are poor, the reference threshold can be appropriately lowered. The station waiting situation is used to indicate the number of waiting people at each station, or the average number of waiting people. In one embodiment, when determining the classification results, if the number of waiting people is large, the vehicle needs to be dispatched in advance, otherwise it is dispatched according to a predetermined departure schedule. Road congestion information and vehicle location information can be used to indicate the flow of buses on the route, and multiple classification results are determined with the goal of ensuring the mobility of buses when making judgments.

[0024] In one embodiment, the road congestion information and vehicle location information in the descriptive text data are expressed in the form of percentages, for example: the target one-way route is 60% unobstructed, 38% congested, and 2% severely congested, the weather is clear, vehicles in working condition are located at 5%, 25%, 47%, and 67% of the line, and the waiting situation at the station is 1.3 people / station.

[0025] In one embodiment, the multi-classification result can be represented by numerical values ​​or text, wherein when represented by numerical values, it is represented by the numerical values ​​corresponding to each classification, and the sum of the numerical values ​​of all classifications is 1, and the result with the largest numerical value is determined as the final result. When represented by text, the text can be represented by "yes", "no", "manual intervention", etc.

[0026] To summarize, the present application obtains traffic data of one-way routes, and performs text description and multi-classification judgment based on the traffic data to obtain multi-classification results, and then based on the multi-classification results, realizes intelligent scheduling of autonomous driving public transportation, improves the efficiency of autonomous driving public transportation, and increases the robustness of the scheduling method.

[0027] It should be noted that the above step of obtaining whether it is necessary to issue a multi-classification result of the vehicle based on the descriptive text data can be achieved through a pre-set judgment condition, or through at least one of mathematical modeling such as neural networks, reinforcement learning, etc.

[0028] Optionally, in the method as described above, the step of dispatching the vehicles to be departed from the departure station corresponding to the target one-way route according to the multiple classification results includes at least one of the following: If the multi-classification result indicates that a vehicle needs to be dispatched, a first signal is output, wherein the first signal is used to control the vehicle to be dispatched to switch to a working state, so that the vehicle is dispatched; If the multi-classification result indicates that the vehicle does not need to be dispatched, a second signal is output, wherein the second signal is used to control the vehicle to be dispatched to maintain a waiting state, so that the vehicle of the scheduled shift continues to stay in the departure station; If the multi-classification result indicates that human intervention is required, the traffic data and / or the description text data are fed back to the client, and based on the instruction input by the client, the first signal or the second signal is output.

[0029] In this embodiment, when scheduling, it can be determined whether a vehicle needs to be dispatched based on the multi-classification results, and the status of the vehicles to be dispatched in the departure station can be controlled by outputting different signals, thereby realizing automatic and intelligent scheduling of buses.

[0030] It should be noted that the second signal may be a null signal.

[0031] See also Figure 2 Specifically, in the method as described above, the step of obtaining the traffic data of the target one-way route includes: Step S201, obtaining the current first time and the number of bus stops and the number of vehicles in working state in the target one-way route; Step S202, determining a city congestion coefficient according to a time range in which the first time is located, wherein if the first time is within a preset time range, determining the city congestion coefficient to be a first threshold, otherwise, determining the city congestion coefficient to be a second threshold, and the second threshold is less than the first threshold; Step S203, determining the data collection interval corresponding to the target one-way route according to the number of bus stops, the number of vehicles and the city congestion coefficient; Step S204: acquiring the traffic data of the target one-way route according to the data collection interval.

[0032] In this embodiment, when acquiring traffic data, in order to ensure timely data collection, the data collection interval is determined according to the current city congestion coefficient, the number of bus stops and the number of vehicles in working state. Specifically, the data collection interval is obtained by dividing the quotient of the number of bus stops and the number of vehicles with a preset multiple by the city congestion coefficient. The calculation method can be expressed as: .

[0033] in, is the data collection interval, is the number of bus stops, is a preset multiple, which is preferably , is the number of vehicles in working condition, is the city congestion coefficient.

[0034] The city congestion coefficient is related to the time range of the current first time. For example, if the current first time is within a preset time range with a large number of vehicles (for example, rush hour), the city congestion coefficient is determined to be a larger first threshold. If the current first time is not within the preset time range, the city congestion coefficient is determined to be a larger second threshold. In a specific embodiment, the first threshold is set to 1.2 and the second threshold is set to 1.

[0035] To facilitate understanding by those skilled in the art, the steps of determining multi-classification results by mathematical modeling are exemplified below.

[0036] Preferably, in the above method, obtaining a multi-classification result of whether a vehicle needs to be dispatched based on the description text data includes: Inputting the description text data into a pre-trained target semantic question-answering multi-classification model to obtain the multi-classification result of whether the vehicle needs to be dispatched; See also Figure 3 , wherein the method for obtaining the target semantic question-answering multi-classification model includes: Step S302, obtaining the historical traffic data corresponding to at least one one-way route; Step S303, constructing the historical traffic data into historical description text data having the preset text structure; Step S304, inputting the history description text data into a Chinese word segmentation encoding model with frozen parameters to obtain first data after word segmentation encoding; Step S305, inputting the text data into a bidirectional long short-term memory network to obtain second data having a preset vector structure; Step S306, inputting the second data into a fully connected layer to obtain a historical multi-classification result; Step S307, adjusting the parameters of gradient descent based on the historical multi-classification results; Step S308, determining the target semantic question-answering multi-classification model according to the parameter information at the end of training.

[0037] In this embodiment, a semantic question-answering multi-classification model in mathematical modeling is used, that is, by performing semantic analysis on the input text data, and searching and judging based on the semantic analysis structure, a multi-classification result in the form of question and answer is obtained. Among them, when obtaining the target semantic question-answering multi-classification model used in this application, historical traffic data corresponding to at least one one-way route will be obtained, and the historical traffic data will be constructed as historical description text data with the above-mentioned preset text structure, and the historical description text data can constitute the training set for subsequent model training; and then training is performed based on the historical description text data to determine the required target semantic question-answering multi-classification model.

[0038] The following is an example of the model training process.

[0039] First, the historical description text data is input into a Chinese word segmentation encoding model with frozen parameters to obtain first data after word segmentation encoding, wherein the Chinese word segmentation encoding model is preferably a word segmenter in BERT (Bidirectional Encoder Representation from Transformer) -base-chinese, wherein the maximum length of the first data output by the word segmenter is a first preset value (for example: 300), and if the length after word segmentation encoding is less than the first preset value, it needs to be padded with "0" to meet the maximum length requirement.

[0040] Secondly, the first data obtained above is input into a bidirectional long short-term memory network (LSTM) to obtain second data with a preset vector structure, so as to facilitate subsequent judgment to obtain a multi-classification result. In one embodiment, the parameters in the bidirectional LSTM are: the number of features of each time step in the input sequence (input_size = 768), the size of the hidden state (hidden_size = 768), and the number of stacked LSTM layers (num_layers = 2).

[0041] Then, the second data is input into the fully connected layer for judgment to obtain the final historical multi-classification result of the current training.

[0042] Finally, the parameters of gradient descent are adjusted based on the historical multi-classification results. When the number of training times is less than the threshold, the next training is started. When the number of training times reaches the threshold, the training is determined to be over, and the target semantic question-answering multi-classification model is determined based on the parameter information at the end of the training.

[0043] In one embodiment, the model is trained with epoch=20, dropout coefficient=0.1, and number of classes class_num=3.

[0044] See also Figure 4 Optionally, the method as described above further comprises: Step S401, receiving an information viewing request input by a client; Step S402: Feedback part or all of the traffic data to the client according to the information viewing request.

[0045] In this embodiment, the method in this application also supports human-computer interaction, that is, receiving information viewing requests input by users through the client (such as bus management interface, bus viewing interface), and based on the specific information viewing request, feedback part or all of the information in the traffic data to the client, so that users can obtain the required information (for example: vehicle current location, schedule, time) based on the feedback information, thereby improving passengers' travel experience. At the same time, it is also convenient for staff to discover vehicle operation problems based on the feedback information and intervene in time, thereby ensuring the safety of system operation and improving the efficiency of autonomous driving public transportation.

[0046] Preferably, in the method as described above, the traffic data includes weather information, vehicle operating status information, time information and road condition information, wherein the road condition information includes at least one of vehicle speed information, vehicle position information, platform position information and the number of people waiting at the platform.

[0047] The weather information is used to determine the weather conditions in the preset text structure, including but not limited to the information feedback from the weather forecast system and the bus monitoring system; The time information includes the time corresponding to each data, such as the upload time of the current traffic data, the collection time of the number of people waiting at the platform, the collection time of the vehicle speed and / or vehicle position, etc.; Vehicle speed information, preferably the speed of all vehicles in working state between two traffic data acquisition intervals; The vehicle position information, preferably the position of all vehicles in working state between two traffic data acquisition intervals, can be determined by a vehicle positioning system, such as a Global Positioning System (GPS), a Beidou positioning system, etc.

[0048] See also Figure 5 Another embodiment of the present application provides a control device for intelligent dispatching of public transportation by autonomous driving, comprising: The first data collection module 501 is used to obtain traffic data of a target one-way route, where the target one-way route is an up route or a down route of a bus route; A first data processing module 502 is used to construct descriptive text data with a preset text structure according to the traffic data, wherein the preset text structure includes weather conditions and road condition reference information, and the road condition reference information includes at least one of station waiting conditions, road congestion information and vehicle position information; The dispatch algorithm module 503 is used to obtain a multi-classification result of whether a vehicle needs to be dispatched according to the description text data; The scheduling execution module 504 is used to schedule the vehicles to be departed from the departure station corresponding to the target one-way route according to the multi-classification results.

[0049] Optionally, in the control device as described above, the scheduling execution module includes at least one of the following: A first processing unit, configured to output a first signal if the multi-classification result indicates that a vehicle needs to be dispatched, wherein the first signal is used to control the vehicle to be dispatched to switch to a working state; A second processing unit is used to output a second signal if the multi-classification result indicates that the vehicle does not need to be dispatched, wherein the second signal is used to control the vehicle to be dispatched to maintain a waiting state; The third processing unit is configured to feed back the traffic data and / or the description text data to the client if the multi-classification result indicates that human intervention is required, and output the first signal or the second signal based on an instruction input by the client.

[0050] Specifically, in the control device as described above, the first data collection module includes: A fourth processing unit is used to obtain the current first time and the number of bus stops and the number of vehicles in working state in the target one-way route; a fifth processing unit, configured to determine a city congestion coefficient according to a time range in which the first time is located, wherein if the first time is within a preset time range, the city congestion coefficient is determined to be a first threshold; otherwise, the city congestion coefficient is determined to be a second threshold, and the second threshold is less than the first threshold; a sixth processing unit, configured to determine a data collection interval corresponding to the target one-way route according to the number of bus stops, the number of vehicles, and the city congestion coefficient; The seventh processing unit is used to obtain the traffic data of the target one-way route according to the data collection interval.

[0051] Preferably, in the control device as described above, the scheduling algorithm module comprises: An eighth processing unit, configured to input the description text data into a pre-trained target semantic question-answering multi-classification model to obtain the multi-classification result of whether the vehicle needs to be dispatched; The method for acquiring the target semantic question-answering multi-classification model in the control device includes: A second data collection module, used to obtain the historical traffic data corresponding to at least one one-way route; A second data processing module, used for constructing the historical traffic data into historical description text data having the preset text structure; A model upstream module, used for inputting the historical description text data into a Chinese word segmentation encoding model with frozen parameters to obtain first data after word segmentation encoding; A model downstream module, used for inputting the text data into a bidirectional long short-term memory network to obtain second data with a preset vector structure; A model output module, used for inputting the second data into a fully connected layer to obtain a historical multi-classification result; A model adjustment module, used for adjusting the parameters of gradient descent based on the historical multi-classification results; The model determination module is used to determine the target semantic question-answering multi-classification model according to the parameter information at the end of training.

[0052] Optionally, the control device as described above further includes: A first information interaction module, used to receive an information viewing request input by a client; The second information interaction module is used to feed back part or all of the information in the traffic data to the client according to the information viewing request.

[0053] Preferably, in the control device as described above, the traffic data includes weather information, vehicle operating status information, time information and road condition information, wherein the road condition information includes at least one of vehicle speed information, vehicle position information, platform position information and the number of people waiting at the platform.

[0054] The control device embodiment of the automatic driving public transportation intelligent dispatching of the present application is a device corresponding to the embodiment of the automatic driving public transportation intelligent dispatching method described above. All the implementation means in the above method embodiment are applicable to the embodiment of the device and can achieve the same technical effect. The above control device provided in the embodiment of the present application can implement all the method steps implemented in the above method embodiment and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as those in the method embodiment will not be described in detail here.

[0055] Yet another embodiment of the present application provides a mobile terminal, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, the steps of the method for intelligent scheduling of public transportation by autonomous driving are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be described here.

[0056] Another embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for intelligent scheduling of autonomous driving public transportation as described above are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0057] Another embodiment of the present application provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of the autonomous driving public transportation intelligent scheduling method as described above, and can achieve the same technical effect. To avoid repetition, they are not described here.

[0058] In addition, the present application may repeat reference numerals and / or letters in different examples. This repetition is for the purpose of simplicity and clarity, and does not in itself indicate the relationship between the various embodiments and / or settings discussed.

[0059] It should also be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusions.

[0060] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles described in the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. An intelligent dispatching method for autonomous driving public transportation, characterized in that: include: Acquire traffic data of a target one-way route, where the target one-way route is an up route or a down route of a bus route; Constructing descriptive text data having a preset text structure according to the traffic data, wherein the preset text structure includes weather conditions and road condition reference information, and the road condition reference information includes at least one of station waiting conditions, road congestion information, and vehicle position information; According to the description text data, a multi-classification result of whether the vehicle needs to be dispatched is obtained; According to the multi-classification results, the vehicles to be departed from the departure station corresponding to the target one-way route are dispatched.

2. The method according to claim 1, characterized in that The step of dispatching the vehicles to be departed from the departure station corresponding to the target one-way route according to the multiple classification results includes at least one of the following: If the multi-classification result indicates that a vehicle needs to be dispatched, a first signal is output, wherein the first signal is used to control the vehicle to be dispatched to switch to a working state; If the multi-classification result indicates that the vehicle does not need to be dispatched, a second signal is output, wherein the second signal is used to control the vehicle to be dispatched to maintain a waiting state; If the multi-classification result indicates that human intervention is required, the traffic data and / or the description text data are fed back to the client, and based on the instruction input by the client, the first signal or the second signal is output.

3. The method according to claim 1, characterized in that The acquiring of the traffic data of the target one-way route includes: Obtain the current first time and the number of bus stops and the number of vehicles in working state in the target one-way route; Determine a city congestion coefficient according to a time range in which the first time is located, wherein if the first time is within a preset time range, determine the city congestion coefficient to be a first threshold, otherwise, determine the city congestion coefficient to be a second threshold, and the second threshold is less than the first threshold; Determining a data collection interval corresponding to the target one-way route according to the number of bus stops, the number of vehicles, and the city congestion coefficient; The traffic data of the target one-way route is acquired according to the data collection interval.

4. The method according to claim 1, characterized in that: The method of obtaining a multi-classification result of whether a vehicle needs to be dispatched based on the description text data includes: Inputting the description text data into a pre-trained target semantic question-answering multi-classification model to obtain the multi-classification result of whether the vehicle needs to be dispatched; The method for obtaining the target semantic question-answering multi-classification model includes: Acquiring the historical traffic data corresponding to at least one one-way route; Constructing the historical traffic data into historical description text data having the preset text structure; Inputting the historical description text data into a Chinese word segmentation encoding model with frozen parameters to obtain first data after word segmentation encoding; Inputting the text data into a bidirectional long short-term memory network to obtain second data having a preset vector structure; Inputting the second data into a fully connected layer to obtain a historical multi-classification result; Based on the historical multi-classification results, adjusting the parameters of gradient descent; The target semantic question-answering multi-classification model is determined according to the parameter information at the end of the training.

5. The method according to claim 1, characterized in that Also includes: Receive information viewing request input by the client; According to the information viewing request, part or all of the information in the traffic data is fed back to the client.

6. The method according to claim 1, 4 or 5, characterized in that: The traffic data includes weather information, vehicle operating status information, time information and road condition information, wherein the road condition information includes at least one of vehicle speed information, vehicle position information, platform position information and the number of people waiting at the platform.

7. A control device for intelligent dispatching of public transportation by automatic driving, characterized in that: include: A first data collection module is used to obtain traffic data of a target one-way route, where the target one-way route is an up route or a down route of a bus route; A first data processing module, configured to construct descriptive text data having a preset text structure according to the traffic data, wherein the preset text structure includes weather conditions and road condition reference information, and the road condition reference information includes at least one of station waiting conditions, road congestion information, and vehicle position information; A dispatch algorithm module, used to obtain a multi-classification result of whether a vehicle needs to be dispatched according to the description text data; The scheduling execution module is used to schedule the vehicles to be departed from the departure station corresponding to the target one-way route according to the multi-classification results.

8. A mobile terminal, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method for intelligent dispatching of public transportation by autonomous driving as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for intelligent scheduling of autonomous driving public transportation as described in any one of claims 1 to 6 are implemented.

10. A computer program product, characterized in that The method comprises computer instructions which, when executed by a processor, implement the steps of the method for intelligent scheduling of autonomous driving public transportation as described in any one of claims 1 to 6.

Images