A vehicle dispatching method and device

Abstract

The application discloses a vehicle scheduling method and device, wherein the method comprises the following steps: obtaining pick-up hot spots and drop-off hot spots of a target area in multiple preset time periods; constructing multiple travel tasks based on the pick-up hot spots and the drop-off hot spots; matching the multiple travel tasks with each other based on the start time period and the end time period of the travel tasks, so that the total travel time of the multiple travel tasks is the shortest, and obtaining a target path set; and scheduling vehicles in the target area based on the target path set. The method can ensure that each pick-up hot spot has available vehicles in the passenger pickup scene, and reduce the empty load rate of the vehicles.

Description

Technical Field

[0001] This invention relates to the field of vehicle dispatching technology, and in particular to a vehicle dispatching method and apparatus. Background Technology

[0002] With the rapid development of technology and the economy, people's travel needs are increasing daily. Due to the limitations of urban road resources, traffic congestion is becoming increasingly severe. Ride-hailing, such as taxis and ride-hailing services, is one of the main modes of urban transportation; in the foreseeable future, driverless taxis will gradually become more widespread. Therefore, optimizing vehicle dispatching during ride-hailing can improve road resource utilization and alleviate traffic congestion to a certain extent. However, in the current ride-hailing process, drivers mostly rely on their judgment of the passenger's likely location when picking up passengers, and there is a game-theoretic relationship between drivers and passengers during pick-up and drop-off, resulting in low matching efficiency. Although some existing dispatch systems can issue passenger demand prompts, picking up passengers based on these prompts still leads to over-concentration of vehicles in some areas, resulting in high empty load rates, while other areas are unavailable for ride-hailing, leading to a poor user experience.

[0003] Therefore, the current vehicle dispatching for ride-hailing services still suffers from imbalances in dispatching across different regions, resulting in a high empty load rate. Summary of the Invention

[0004] In view of the above problems, the present invention proposes a vehicle dispatching method and device, which can reduce the vehicle empty load rate while ensuring that passengers at each pick-up hotspot have available vehicles in the passenger ride-hailing scenario.

[0005] In a first aspect, this application provides the following technical solution through an embodiment:

[0006] A vehicle dispatching method includes: acquiring multiple pre-set pick-up hotspot locations and drop-off hotspot locations in a target area during multiple time periods; constructing multiple travel tasks based on the pick-up hotspot locations and drop-off hotspot locations; matching the multiple travel tasks with each other based on the start and end times of the travel tasks, with the shortest total travel time to complete the multiple travel tasks as the matching target, to obtain a target path set; and dispatching vehicles within the target area based on the target path set.

[0007] Optionally, the step of matching the multiple trip tasks based on their start and end times, with the goal of minimizing the total travel time to complete the multiple trip tasks, to obtain a target path set includes:

[0008] The trip tasks where the drop-off hotspot is located within the target time period are identified as prerequisite tasks, thus obtaining a prerequisite task set; wherein, the target time period is any one of the plurality of preset time periods; the pick-up hotspot locations located within the next preset time period adjacent to the target time period are identified as pick-up locations to be matched, thus obtaining a set of pick-up locations to be matched; with the shortest total travel time as the matching target, the prerequisite task set is matched with the set of pick-up locations to be matched, thus obtaining a set of continuing routes; each continuing route includes the prerequisite task and the corresponding pick-up location to be matched; based on the set of continuing routes for each preset time period, the target route set is obtained.

[0009] Optionally, the step of matching the set of preceding tasks with the set of boarding locations to be matched, using the shortest total travel time as the matching target, to obtain the set of continuing routes includes:

[0010] The set of preceding tasks is simulated and matched with the set of boarding locations to be matched to obtain the travel time of each matched virtual path; based on the travel time of the virtual path, the set of preceding tasks is matched with the set of boarding locations to be matched with the shortest total travel time as the matching target to obtain the set of continuing routes.

[0011] Optionally, the step of simulating the matching of the set of preceding tasks with the set of boarding locations to be matched to obtain the travel time of each virtual path includes:

[0012] The set of pre-selected tasks is simulated and matched with the set of boarding locations to obtain candidate routes; virtual routes are selected from the candidate routes based on the filtering condition that the travel time of the candidate route is less than the time period of the candidate route, and the travel time of each virtual route is obtained; wherein, the time period is the duration between the start time period and the end time period.

[0013] Optionally, the step of matching the set of preceding tasks with the set of boarding locations to be matched based on the travel time of the virtual route, with the shortest total travel time as the matching target, to obtain the set of continuing routes includes:

[0014] Based on the travel time of the virtual path, the weight of the virtual path is determined to obtain a weight set; based on the weight set, the Hungarian algorithm is used to match the preceding task with the boarding location to be matched to obtain the set of continuing routes.

[0015] Optionally, determining the weight of the virtual path based on the travel time of the virtual path, and obtaining the weight set, includes:

[0016] Based on the travel time of the virtual path, an original weight set is determined; using the travel time of the virtual path being less than the duration of the virtual path's time period as a filtering condition, target weights are filtered out from the original weight set to obtain the weight set; wherein, the duration of the time period is the duration between the start time period and the end time period.

[0017] Optionally, obtaining the boarding and alighting hotspot locations for multiple preset time periods in the target area includes:

[0018] Obtain historical vehicle usage data for multiple preset time periods in the target area; cluster the historical vehicle usage data for each preset time period to obtain the pick-up hotspot location and drop-off hotspot location for each preset time period.

[0019] Optionally, the step of clustering the historical vehicle usage data for each preset time period to obtain the pick-up hotspot locations and drop-off hotspot locations for each preset time period includes:

[0020] The historical vehicle usage data for each preset time period is clustered to obtain multiple cluster centers; a heat label is generated for each cluster center based on the amount of data corresponding to each cluster center; and the boarding hotspot location and the alighting hotspot location are obtained based on the cluster center and the corresponding heat label.

[0021] Optionally, based on the boarding hotspot location and the alighting hotspot location, multiple trip tasks are constructed, including:

[0022] Obtain historical trips from historical vehicle usage data; based on the historical trips, match the pick-up hotspot locations and the drop-off hotspot locations to obtain the multiple trip tasks.

[0023] Optionally, dispatching vehicles within the target area based on the target path set includes:

[0024] For each preset time period, vehicles are dispatched to the starting boarding hotspot location of each target route.

[0025] Optionally, dispatching vehicles to the starting pick-up hotspot location of each target route for each preset time period includes:

[0026] For each preset time period, the parking locations of multiple vehicles to be dispatched are obtained; the parking locations are matched with the starting pick-up hotspot locations, with the shortest total dispatch time as the matching target, to obtain the target vehicle matched for each starting pick-up hotspot location; wherein, the total dispatch time is the sum of the time for completing vehicle dispatch for each starting pick-up hotspot location; the target vehicle is dispatched to the matched starting pick-up hotspot location.

[0027] Secondly, based on the same inventive concept, this application provides the following technical solution through an embodiment:

[0028] A vehicle dispatching device includes: a hotspot location acquisition module for acquiring multiple pre-set time-slot pick-up and drop-off hotspot locations in a target area; a construction module for constructing multiple travel tasks based on the pick-up and drop-off hotspot locations; a matching module for matching the multiple travel tasks with each other based on the start and end times of the travel tasks, with the shortest total travel time for completing the multiple travel tasks as the matching target, to obtain a target path set; and a dispatching module for dispatching vehicles within the target area based on the target path set.

[0029] Thirdly, based on the same inventive concept, this application provides the following technical solution through an embodiment:

[0030] An electronic device includes a processor and a memory coupled to the processor, the memory storing instructions that, when executed by the processor, cause the electronic device to perform the steps of the method described in any of the first aspects above.

[0031] Fourthly, based on the same inventive concept, this application provides the following technical solution through an embodiment:

[0032] A readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the first aspects above.

[0033] This invention provides a vehicle dispatching method and apparatus that constructs multiple travel tasks by obtaining pick-up and drop-off hotspot locations, indicating that these hotspot locations may continue to have the same vehicle demand in the future. Then, by matching the start and end times of the travel tasks, multiple travel tasks are linked together, thus binding them together; that is, after a vehicle completes a travel task, it can be dispatched to the next pick-up hotspot location, forming a binding between the drop-off hotspot location and the next pick-up hotspot location. Furthermore, the matching process aims to minimize the total travel time for completing multiple travel tasks, ensuring that the final target path set allows for the completion of multiple travel tasks using the fewest vehicles in the shortest time. By using this target path set to guide vehicle dispatching within the target area, the vehicle vacancy rate is reduced while ensuring that passengers at each pick-up hotspot location have available vehicles.

[0034] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0036] Figure 1 A flowchart of a vehicle dispatching method provided in an embodiment of the present invention is shown;

[0037] Figure 2 The spatiotemporal distribution map of hotspot locations obtained by clustering in an embodiment of the present invention is shown;

[0038] Figure 3 An exemplary simulation matching principle diagram is shown in an embodiment of the present invention;

[0039] Figure 4 A schematic diagram illustrating the principle of scheduling target vehicles in an embodiment of the present invention is shown;

[0040] Figure 5 A schematic diagram of a vehicle dispatching device provided in an embodiment of the present invention is shown. Detailed Implementation

[0041] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0042] This invention provides a vehicle dispatching method and apparatus applicable to various fields, such as urban taxi dispatching, driverless taxi dispatching, driverless ride-hailing dispatching, urban public transport station and route planning, logistics center site selection planning, etc. When applied, the various steps of this vehicle dispatching method and apparatus can be stored in a storage medium as software code and executed by a computer host or server with computing power. The concept of the vehicle dispatching method and apparatus of this invention will be described in detail below through specific examples, with this embodiment mainly using the dispatching of driverless taxis / driverless ride-hailing vehicles as an example.

[0043] Please see Figure 1 The flowchart illustrates a vehicle dispatching method provided in one embodiment of the present invention, including:

[0044] Step S10: Obtain the boarding and alighting hotspot locations for multiple preset time periods in the target area;

[0045] Step S20: Based on the boarding hotspot location and the alighting hotspot location, construct multiple trip tasks;

[0046] Step S30: Based on the start and end times of the trip tasks, with the shortest total travel time to complete the multiple trip tasks as the matching target, the multiple trip tasks are matched with each other to obtain a target path set;

[0047] Step S40: Based on the target path set, dispatch vehicles within the target area.

[0048] In this embodiment, steps S10 to S40 construct multiple travel tasks based on the obtained pick-up and drop-off hotspot locations, indicating that these hotspot locations may have similar future travel needs. Next, the multiple travel tasks are matched based on their start and end times, thus binding them together. That is, after a vehicle completes a travel task, it can be dispatched to the next pick-up hotspot location, forming a binding between the drop-off hotspot location and the next pick-up hotspot location. The matching process aims to minimize the total travel time for completing multiple travel tasks, ensuring that the final target path set allows for the completion of multiple travel tasks using the fewest vehicles in the shortest time. Using this target path set to guide vehicle dispatch within the target area reduces the number of dispatched vehicles and lowers the vehicle vacancy rate while ensuring that passengers at each hotspot location have available transportation. The implementation principles and methods of the above steps are explained in detail below with examples.

[0049] Step S10: Obtain the boarding hotspot locations and alighting hotspot locations for multiple preset time periods in the target area.

[0050] In step S10, the target area represents the region where vehicle dispatching is required. This target area can be a custom-defined region. It is understandable that the target area can be obtained by dividing the city by its administrative divisions; the target area can be obtained by dividing the city by its road network, such as dividing the area within the Second Ring Road as the target area, dividing the area within the Third Ring Road as the target area, and so on; the target area can be obtained by dividing the city's urban and suburban areas, such as using the urban area of ​​the city as the target area; and so on.

[0051] Another method is to determine the target area through data statistics. For example, by statistically analyzing historical ride-hailing data for a city, the distribution of pick-up locations can be identified. Then, noisy pick-up locations can be removed from the distribution data. The outermost pick-up locations on a map can then be connected (e.g., by connecting line segments or fitting curves) to form the target area. Alternatively, the distribution of population and businesses in a city can be statistically analyzed to identify areas where the population and business density exceeds a certain threshold, which can then be used as the target area. Noisy pick-up locations can be understood as locations where ride-hailing demand is unlikely to occur only once in a relatively long period, such as pick-up locations where ride-hailing demand occurs only once a week.

[0052] The preset time period is the time period obtained by dividing the total duration of the scheduling plan (hereinafter referred to as the total scheduling duration for ease of description). For example, if the duration of a scheduling plan is 24 hours, then the 24 hours can be divided into various preset time periods. In addition, the duration of the scheduling plan can also be 2 hours, 4 hours, 8 hours, 12 hours, 16 hours, etc., without limitation. There is no limit to the number of preset time periods, for example, it can be 3, 4, 10, 12, etc.

[0053] Each preset time period can be of equal duration, for example, each preset time period can be 20 minutes, 30 minutes, 40 minutes, 1 hour, 1 hour and 20 minutes, etc.

[0054] Each preset time period can be of different lengths. Understandably, the total scheduling time is divided into first-frequency time periods, second-frequency time periods, third-frequency time periods, and so on, based on travel frequency. Then, based on travel frequency, the time periods corresponding to different frequencies are further subdivided to obtain preset time periods. When dividing preset time periods, higher travel frequency indicates higher demand for transportation, and the preset time periods can be shorter to ensure more accurate hotspot locations within the target area and preset time periods, thus improving the passenger experience.

[0055] For example, based on travel frequency, the periods between 7:00 AM and 9:30 AM, and between 5:00 PM and 7:00 PM are identified as the first frequency periods with higher travel frequency; the periods between 9:30 AM and 5:00 PM, and between 7:00 PM and 10:30 PM are identified as the second frequency periods with the next highest travel frequency; and the remaining times are identified as the third frequency periods with lower travel frequency. Furthermore, the first frequency periods can be divided into preset time slots in 20-minute intervals, the second frequency periods into preset time slots in 40-minute intervals, and the third frequency periods into preset time slots in 1-hour intervals. These different preset time slots result in different preset time slots; shorter preset time slots correspond to higher frequency travel, improving the accuracy of hotspot location; longer preset time slots correspond to lower frequency travel, reducing the complexity of subsequent planning.

[0056] In addition, the impact of weekdays and holidays can be considered to divide the preset time periods differently.

[0057] The determination of the boarding and alighting hot spots for each preset time period can be achieved in multiple ways.

[0058] Pick-up and drop-off hotspots can be empirically marked on the map. For example, one or more convenient pick-up and drop-off locations near residential areas can be marked as pick-up and / or drop-off hotspots; one or more convenient pick-up and drop-off locations near factory areas, commercial office buildings, etc., can be marked as pick-up and / or drop-off hotspots.

[0059] In some implementations, the pick-up and drop-off hot spots for each preset time period can be determined based on historical vehicle usage data. One specific implementation process includes: first, acquiring historical vehicle usage data for multiple preset time periods in the target area; then, clustering the historical vehicle usage data for each preset time period to obtain the pick-up and drop-off hot spots for each preset time period.

[0060] Understandably, this historical vehicle usage data may include multiple usage points within the target area. Each usage point may include the following attributes: the location coordinates of the usage point, the usage time of the usage point, and the status label of the usage point (boarding label or alighting label). This historical vehicle usage data may be usage data from 1 day, 2 days, 3 days, 4 days, etc., before the scheduling time; it may also be usage data from the same day of the week preceding the scheduling time; it may also be all usage data from the week preceding the scheduling time; or it may be usage data from one or more days of the month preceding the scheduling time.

[0061] Furthermore, these pick-up points are clustered into preset time periods, and termination conditions are set for each cluster. For example, the termination condition could be a lower limit on the number of clusters required for clustering; a smaller number of clusters indicates greater demand for transportation at each cluster center, potentially increasing the walking distance for passengers to reach the pick-up point. Alternatively, it could be the maximum radius of the region corresponding to each cluster after clustering; a larger radius indicates greater demand for transportation at each cluster center, potentially increasing the walking distance for passengers to reach the pick-up point. Since each pick-up point carries a status label, pick-up and drop-off points do not affect each other during clustering. Pick-up points with pick-up labels can obtain pick-up hotspot locations after clustering, and drop-off points with drop-off labels can obtain drop-off hotspot locations after clustering.

[0062] The ward linkage method is used as an example for illustration. This method treats each usage point in historical vehicle usage data as a cluster, enumerates all binomial clusters (N clusters are N*(N-1) / 2 binomial sets), calculates the total ESS value after merging these two clusters, and selects the two clusters with the smallest increase in total ESS value to merge. This merging process is repeated item by item until the number of clusters reaches a preset lower limit, or the actual maximum radius of the cluster reaches a preset radius, or the ESS value reaches a preset upper limit, at which point the process terminates. The resulting cluster centers can serve as multiple hotspot locations. During the merging calculation, the ESS (Error Sum of Squares criterion) formula can be used to calculate the merging target, merging the two clusters with the smallest ESS values. The principle is as follows:

[0063]

[0064]

[0065] In formula (1), x i Let x represent the distance between any element in a cluster and the cluster center (mean); in formula (2), x represents all elements in two clusters c1 and c2. This represents the cluster center after the two clusters have merged. This is the ESS value after merging the two clusters.

[0066] This hierarchical agglomerative clustering method merges historical vehicle usage data into a single cluster, allowing noisy or outlier points to each occupy their own cluster. This more accurately reflects the actual usage situation, including individual scattered points, and improves the reliability of pick-up and drop-off hotspot locations. In addition, other clustering methods can be used, such as Mean-shift clustering, K-Means clustering, BIRCH clustering, etc.

[0067] The spatiotemporal structure of hotspot locations obtained by clustering across multiple preset time periods can be as follows: Figure 2As shown in the diagram. The coordinate axis t represents time, and each layer represents the hotspot locations clustered within a preset time period. Coordinate axes a and b represent the longitude and latitude of the hotspot locations, respectively. For example, boarding hotspot location 1 (a1, b1, t1) indicates that in the preset time period t1, location (a1, b1) is a boarding hotspot location; boarding hotspot location 2 (a2, b2, t2) indicates that in the preset time period t2, location (a2, b2) is a boarding hotspot location; and so on. Similarly, disembarking hotspot location 1 (a′1, b′1, t′1) indicates that in the preset time period t′1, location (a′1, b′1) is a boarding hotspot location; disembarking hotspot location 2 (a′2, b′2, t′2) indicates that in the preset time period t′2, location (a′2, b′2) is a boarding hotspot location; and so on. When recording, the difference between the midpoint values ​​of two preset time periods can be used as the duration of the interval between the two preset time periods. For example, the duration between the preset time periods 7:00 to 7:30 and 7:30 to 8:00 is 30 minutes.

[0068] In some implementations, heat maps can be introduced to correct the clustering results, adjusting for areas with high vehicle usage and areas with low usage. For example, a cluster center A (hotspot location A) is obtained by clustering 5n usage points; a cluster center B (hotspot location B) is obtained by clustering n usage points. Clearly, the actual usage at hotspot location A is much greater than at hotspot location B. That is, the heat map of hotspot location A is higher than that of hotspot location B. If the number of vehicles dispatched to hotspot locations A and B is the same, then vehicles at hotspot location A may not meet the demand, or vehicles at hotspot location B may be empty.

[0069] The process of introducing popularity tags to correct clustering results and adjust the hotspot positions is as follows:

[0070] First, the historical vehicle usage data for each preset time period is clustered to obtain multiple cluster centers. Then, based on the amount of data corresponding to each cluster center, a heat label is generated for each cluster center; the amount of data is the total amount of historical vehicle usage data used to obtain that cluster center; for example, the amount of data could be 20, 200, 300, etc. The heat label is a parameter that measures the size of this data amount; for example, 20, 200, and 300 data points could correspond to heat labels 1, 10, and 15, respectively. Finally, based on the cluster centers and their corresponding heat labels, the hot spots for boarding and alighting are obtained.

[0071] It is understandable that the popularity tag can be carried by the corresponding hot spot location; when executing steps S20 and S30, the hot spot location is entered multiple times based on the popularity tag, and then the trip task construction and trip task matching are performed.

[0072] Alternatively, based on the popularity tag, one or more identical hotspot locations can be generated from a cluster center corresponding to the popularity tag. For example, if the popularity tag of a cluster center C (boarding tag) is 1, then one boarding hotspot location can be generated based on cluster center C; if the popularity tag of a cluster center D (boarding tag) is 4, then four identical boarding hotspot locations can be generated based on cluster center D.

[0073] In this implementation, the use of heat tags can further improve the reliability of hotspot locations, ensuring that the subsequently determined target routes are more reliable and the number of vehicles dispatched is more accurate.

[0074] Step S20: Based on the boarding hotspot location and the alighting hotspot location, construct multiple trip tasks.

[0075] In step S20, the boarding hotspot and alighting hotspot can be associated based on experience to form a travel task.

[0076] In some implementations, hotspot locations can be associated based on statistical results of historical vehicle usage data. Specifically, historical vehicle usage data may also include historical trips. First, historical trips are retrieved from the historical usage data; then, based on the historical trips, pick-up and drop-off hotspot locations are matched to obtain multiple trip tasks.

[0077] Understandably, if a historical trip exists between an area corresponding to a pick-up hotspot and an area corresponding to a drop-off hotspot, then it is assumed that there is a demand for transportation at the pick-up hotspot, and the destination is the drop-off hotspot. A trip task can then be constructed from the pick-up hotspot and the drop-off hotspot.

[0078] Additionally, a quantity threshold can be set to filter out noisy data in travel tasks. For example, if there are multiple historical trips in the area corresponding to a pick-up hotspot and the area corresponding to a drop-off hotspot, and the number of historical trips exceeds the quantity threshold, then it is considered that there is a demand for transportation at the pick-up hotspot and that the destination is the drop-off hotspot. A travel task can then be constructed based on the pick-up and drop-off hotspots. Otherwise, no travel task is constructed. This can eliminate a small amount of noisy data in travel tasks; for example, if there is only one historical trip between these two hotspots, it can be inferred that the historical trip was relatively sudden and has a low probability of recurring in the future. Excluding such travel tasks allows travel tasks to more accurately reflect the actual situation.

[0079] Once the travel task is constructed, the starting point for each task is the pick-up hotspot, and the ending point is the drop-off hotspot. The travel route and duration to complete the task can be planned using the starting and ending points. For example, based on the traffic conditions between the two hotspots and during the preset time period covered by the task, the route with the shortest travel time, shortest distance, fewest traffic lights, or most main roads can be determined. The required travel time to complete the task can then be determined based on the speed limits, congestion conditions, and number of traffic lights along the route.

[0080] In some implementations, the travel time can be determined based on the historical trips corresponding to the trip task. For example, firstly, all historical trips corresponding to the trip task are obtained; then, the average or median of the travel time required to complete each historical trip is calculated, and the result of the average calculation or the median is determined as the travel time to complete the trip task. This can more accurately reflect the actual time consumed and can take into account the influence of different historical trips.

[0081] Step S30: Based on the start and end times of the trip tasks, and with the shortest total travel time to complete the multiple trip tasks as the matching target, the multiple trip tasks are matched with each other to obtain a target path set.

[0082] In step S30, the start time of the trip task is a preset time period at the starting point of the trip task's travel path; the end time of the trip task is a preset time period at the ending point of the trip task's travel path.

[0083] It is understandable that different trip tasks may fall within different preset time periods. Trip tasks with different preset time periods can be matched with each other. For example, if the end time of trip task A is before the start time of another trip task B, it means that the vehicle completing trip task A has the opportunity to travel to the pick-up hotspot of trip task B and continue to pick up trip task B. Therefore, in this embodiment, after matching these trip tasks, the preceding trip task can be bound to the following trip task, avoiding the need to dispatch vehicles separately to the pick-up hotspot of trip task B, thereby reducing vehicle usage. The matching process aims to minimize the total travel time for completing multiple trip tasks, thus ensuring the shortest vehicle idle time. Therefore, the target path set obtained after matching can represent the operation plan with the shortest total running time and the fewest vehicles required.

[0084] There are no restrictions on the specific methods for matching multiple travel tasks. For example, based on the time sequence of the travel tasks within a preset time period, each travel task can be traversed to generate multiple path combinations (or all possible path combinations can be enumerated). Each path does not contain branches (understandably, for multiple identical hotspot locations that are repeatedly entered, each identical hotspot location belongs to a different target path). Furthermore, the total travel time of each of these multiple path combinations is calculated, and the path combination with the shortest total travel time is determined as the target path set.

[0085] Furthermore, in some implementations, step S30 may include the following sub-steps:

[0086] Step S31: Identify the trip tasks where the drop-off hotspot is within the target time period as the prerequisite tasks, and obtain the prerequisite task set; wherein, the target time period is any one of multiple preset time periods;

[0087] Step S32: Determine the boarding hotspot locations in the next preset time period adjacent to the target time period as the boarding locations to be matched, and obtain the set of boarding locations to be matched.

[0088] In steps S31 and S32, it is understood that any trip task can be used as a preceding task. The set of boarding locations to be matched corresponding to the preceding task is located in the next preset time period adjacent to the preceding task. Furthermore, to avoid invalid data in subsequent calculations, the latest preset time period can be removed when determining the set of preceding tasks, that is, the target time period is not the latest preset time period. Similarly, when determining the boarding locations to be matched, the earliest preset time period can be removed, that is, the target time period is not the earliest preset time period.

[0089] For example, if the total scheduling duration is from 6:00 AM to 11:00 PM, and the preset time slot is 30 minutes, then the earliest preset time slot is 6:00 AM to 6:30 AM, and the latest preset time slot is 10:30 PM to 11:00 PM. If the target time slot is 6:00 AM to 6:30 AM, then the next preset time slot adjacent to the target time slot is 6:30 AM to 7:00 AM.

[0090] It should also be noted that the order in which steps S31 and S32 are executed is not restricted.

[0091] Step S33: Using the shortest total travel time as the matching target, match the set of preceding tasks with the set of pick-up locations to be matched to obtain the set of continuing routes; each continuing route includes the preceding tasks and the corresponding pick-up locations to be matched.

[0092] The total travel time in step S33 represents the time required to complete all paths in the obtained continuing route set. Each continuing route in the continuing route set is a path that involves traveling from the pick-up hotspot location corresponding to the preceding task to the drop-off hotspot location of the preceding task, and then to the pick-up location to be matched with the preceding task. This continuing route set is the matching combination with the shortest total travel time among all possible matching combinations.

[0093] The matching process in step S33 can be implemented as follows:

[0094] 1. The set of preceding tasks can be simulated and matched with the set of boarding locations to be matched, so as to obtain the travel time of each matched virtual path.

[0095] Understandably, during simulated matching, each pre-task in the pre-task set is traversed and combined with all possible pick-up locations in the matchable pick-up location set to form multiple virtual paths. A virtual path includes the travel path to complete a pre-task and the travel path to the matched pick-up location. The method for obtaining the travel time to complete a pre-task has been explained earlier (travel task travel time), and can be denoted as T1. Since both the pre-task drop-off hotspot and the matched pick-up location are known, the travel time from the pre-task drop-off hotspot to the matched pick-up location can be estimated based on parameters such as the preset time period of the two hotspots and the traffic conditions between them, and can be denoted as T2. Therefore, the travel time to complete a virtual path is T = T1 + T2.

[0096] Let's take an example to illustrate the principle:

[0097] Please see Figure 3 The set of prerequisite tasks A1 includes three prerequisite tasks, and the set of pick-up locations B1 includes four pick-up locations. During the simulated matching process, each prerequisite task a1 is matched with one of the four pick-up locations, resulting in four virtual paths. The virtual path formed by prerequisite task a1 and pick-up location b1 has a travel time of T = T1 + T2.

[0098] Since the travel time of the virtual path includes the travel time of the preceding tasks, it can be guaranteed that the final set of target paths is the set with the shortest total travel time and the fewest number of vehicles. If only the travel time T2 is considered, it is difficult to guarantee the number of vehicles used. For example, if two vehicles with the same travel time T2 appear, it will be difficult to guarantee the optimal match.

[0099] Since hotspot locations represent the location of vehicle demand in a pre-defined time period, rather than the actual pick-up location for passengers, the feasibility of virtual routes can be further considered in some implementations. For example, in some cases, although the preliminary task and the pick-up location to be matched are in different time periods, the vehicle may still fail to arrive at the corresponding pick-up location on time after completing the preliminary task. This situation can be excluded to improve the reliability of the continuing route.

[0100] Understandably, the first step is to simulate a match between the set of pre-selected tasks and the set of pick-up locations to obtain candidate routes. Then, using the condition that the travel time of a candidate route is less than the duration of its time slot as a filtering criterion, virtual routes are selected from the candidate routes, and the travel time of each virtual route is obtained. The time slot duration refers to the duration between the start and end of the time slot.

[0101] Understandably, during simulated matching, each preceding task in the preceding task set is traversed and combined with all possible pick-up locations in the set to be matched, forming multiple candidate paths. A candidate path includes the travel path to complete a preceding task and the travel path to the matched pick-up location. For an understanding of candidate paths, please refer to the explanation of virtual paths in the previous implementation method; it will not be repeated here. If the travel time of a candidate path is less than the time period of the candidate path, it means that the vehicle is likely not to arrive at the corresponding pick-up location early or on time; otherwise, it means that the vehicle is likely to arrive at the corresponding pick-up location on time or early.

[0102] For example, in a candidate route, the preset time slots for the corresponding preceding tasks are 6:00-6:30 and 6:30-7:00; the time slot for the corresponding pick-up location is 7:00-7:30. If the estimated travel time based on the actual start and end points of the candidate route is 1 hour and 20 minutes, and there is a 1-hour interval between the preset time slots of 6:00-6:30 and 7:00-7:30, then there is a high probability that the vehicle will not arrive at the corresponding pick-up location on time, and the candidate route can be excluded. If the estimated travel time based on the actual start and end points of the candidate route is 40 minutes, then there is a high probability that the vehicle will arrive at the corresponding pick-up location on time, and the candidate route can be retained as a virtual route. This makes the final vehicle dispatch more punctual and improves the passenger experience.

[0103] After obtaining the travel time of the virtual path, proceed to the next matching step.

[0104] 2. Based on the travel time of the virtual path, with the shortest total travel time as the matching target, the set of preceding tasks is matched with the set of boarding locations to be matched to obtain the set of continuing routes.

[0105] Since the travel time of the virtual routes has already been obtained in step 1, we can iterate through the travel times of all virtual routes and, without repeating any hotspots covered by the virtual routes, find a combination of virtual routes that minimizes the total travel time to complete these virtual routes. This combination of virtual routes is the set of continuing routes.

[0106] In some implementations, the set of continuing routes can also be determined based on the Hungarian algorithm for matching. The implementation process is as follows:

[0107] First, based on the travel time of the virtual route, the weights of the virtual routes are determined, resulting in a weight set. For example, the travel time of the virtual route can be used as the weight, as can the reciprocal of the travel time, or a compensated and corrected value of the travel time, and so on. Taking the travel time of the virtual route as an example, this weight set can be represented by the following correlation matrix:

[0108]

[0109] Among them, l1~l n Indicates the prerequisite tasks, p1~p m Indicates the location to be matched for boarding, T 11 ~T mn This represents the travel time of the virtual path corresponding to the preceding task and the boarding location to be matched, which is also known as the weight.

[0110] Then, based on the weight set, the Hungarian algorithm is used to match the preceding tasks with the boarding locations to be matched, obtaining the set of continuing routes. At this point, since it is necessary to find the set of continuing routes with the shortest total travel time, the Hungarian algorithm can be used for minimum weight coverage matching. In some examples, if a larger determined weight corresponds to a shorter travel time, the Hungarian algorithm can be used for maximum weight coverage matching; in other examples, if the association matrix is ​​not a square matrix, virtual elements can be used to fill in the gaps in the association matrix to form a square matrix. The process of solving the above association matrix can be referred to the following steps:

[0111] 1) Subtract the minimum value of each row of the correlation matrix to obtain the first new matrix, and proceed to step 2).

[0112] 2) Subtract the minimum value of each column of the first new matrix to obtain the second new matrix, and proceed to step 3).

[0113] 3) Cover all the 0 elements in the second new matrix with the fewest row and column lines, and check if this is the optimal allocation. If the row and column lines do not cover all the elements of the matrix, proceed to step 4; otherwise, confirm that it is the optimal allocation and proceed to step 5.

[0114] 4) Find the smallest element among the elements where the row and column lines do not intersect, subtract the smallest element from the remaining elements, and add the smallest element to the elements at the intersection of the row and column lines.

[0115] 5) Find the zero element corresponding to each row and the zero element corresponding to each column. Start matching from the row or column with the fewest zero elements until all preceding tasks and the waiting boarding positions have been matched, thus obtaining the set of continuing routes. This set of continuing routes is the optimal solution. Since T mn This represents the travel time of the virtual route. The comprehensive set of continuing routes represents the matching scheme with the shortest total travel time and the fewest required vehicles.

[0116] It should be noted that if the number of preceding tasks and the number of pick-up locations to be matched are not the same, then either a preceding task or a pick-up location to be matched can match a virtual element. Understandably, if a preceding task matches a virtual element, then the drop-off hotspot of that preceding task is the endpoint of the corresponding target path; if a pick-up location to be matched matches a virtual element, then that pick-up location is the starting point of the corresponding target path.

[0117] In some implementations, the travel time of the virtual path can be filtered to eliminate matching results corresponding to pick-up locations where vehicles are unlikely to arrive early or on time. This filtering method can replace the filtering process described above in the simulation matching phase. When the travel time of a certain virtual path is filtered out, a replacement virtual time can be used to facilitate the calculation of the Hungarian algorithm. The specific implementation process is as follows: First, based on the travel time of the virtual path, the original weight set is determined; then, using the fact that the travel time of the virtual path is less than the duration of the virtual path's time segment as the filtering condition, the target weights are filtered out from the original weight set to obtain the weight set; where the time segment duration is the duration between the start and end time segments. The correlation matrix of the weight set in this implementation can be represented as follows:

[0118]

[0119] Among them, l1~l n Indicates the prerequisite tasks, p1~p m Indicates the location to be matched for boarding, X 11 ~X mn This represents the travel time of the virtual path corresponding to the preceding task and the boarding location to be matched, which is also the original weight; k 11 ~k mn This indicates the filtering criteria.

[0120] When the virtual path obtained by matching the i-th preceding task and the j-th waiting boarding location is actually difficult to arrive on time (i.e., the travel time of the virtual path is longer than the time period of the virtual path), k is used to...ji For the driving time X ji Invalidate or remove the marker; for example, let k ji Make X ji k ji = Virtual duration. When the virtual path obtained by matching the i-th preceding task and the j-th waiting boarding location can arrive on time (i.e., the travel time of the virtual path is less than the time period of the virtual path), then let k ji =1. After labeling and filtering all the original weights, the remaining original weights constitute a weight set. Similarly, based on this weight set, the Hungarian algorithm is used to match the previous tasks with the boarding positions to be matched, thus obtaining the set of continuing routes.

[0121] By using the various filtering methods described above, a more accurate travel time for the virtual route can be determined, ensuring that the subsequent matching results are more reasonable and realistic, and avoiding excessively long waiting times for passengers.

[0122] Step S34: Obtain the target route set based on the continuing route set for each preset time period.

[0123] In step S34, since the drop-off hotspot and the next pick-up hotspot within each pair of adjacent time periods have been bound through a matching process, connecting each set of continuing routes through common hotspots yields the target route set. A target route includes one or more continuing routes as sub-routes. Each continuing route set is obtained with the shortest total travel time as the matching objective; therefore, the total travel time corresponding to the target route set is also the shortest. Simultaneously, the travel time of the virtual route, which includes the travel time of the actual trip, is also considered during the matching process, further ensuring that the number of dispatched vehicles is minimized. Therefore, the target routes accurately reflect passengers' travel needs at various spatiotemporal locations. Dispatching vehicles based on the target route set can effectively reduce the empty load rate of driverless taxis and improve the passenger experience.

[0124] Step S40: Based on the target path set, dispatch vehicles within the target area.

[0125] In step S40, the target route includes various hotspot locations. When a hotspot location is located in the middle section of the target route, it indicates that the hotspot location has a corresponding prerequisite task; a vehicle that has completed the prerequisite task can pick up passengers at the hotspot location without the need for a separate vehicle dispatch. Therefore, for each preset time period, vehicles can be dispatched to the starting hotspot location of each target route.

[0126] Additionally, some implementation methods consider locating driverless taxis near popular passenger pick-up points, thereby further optimizing vehicle dispatching. Specific solutions are as follows:

[0127] First, for each preset time period, the parking locations of multiple vehicles to be dispatched are obtained. A parking location is a fixed spot where a vehicle to be dispatched will be parked when not in operation. When a vehicle to be dispatched parks, a preset electronic device at the parking location can report parameters such as the vehicle's number to the server; alternatively, the vehicle to be dispatched can report its own parking location to the server. The server continuously stores and updates this data. When the vehicle dispatching method needs to be executed, the processor or electronic device executing the method can retrieve and call the data.

[0128] Next, the parking locations are matched with the starting pick-up hotspot locations, with the shortest total dispatch time as the matching objective, to obtain the target vehicle matched for each starting pick-up hotspot location. The total dispatch time is the sum of the time required to dispatch a vehicle to each starting pick-up hotspot location. Furthermore, the matching process can also employ the matching method described earlier for obtaining the continuing route. For example, the travel time from each parking location to each starting pick-up hotspot location can be used as a weight, and the Hungarian algorithm can be used for minimum weight matching to obtain the target vehicle matched for each starting pick-up hotspot location. Other details and processing methods in the matching process can be found in the previous section on matching to obtain the continuing route, and will not be repeated here.

[0129] Understandably, a parking location A may be occupied by multiple vehicles awaiting dispatch. During the dispatch matching process, parking location A can first be read multiple times (or read once and then copied multiple times) to create multiple identical parking locations A, and then the matching calculation is performed. For example, if parking location A has 3 vehicles awaiting dispatch, then parking location A can be read 3 times, creating three identical parking locations A; after matching is completed, all three vehicles awaiting dispatch can be dispatched away from that parking location.

[0130] Finally, the target vehicle will be dispatched to the matching starting pick-up hotspot location.

[0131] Understandably, after a vehicle completes all its journey tasks, it can be matched with a parking location to ensure that the vehicle can drive to the optimal parking position, avoiding detours or uneven vehicle allocation to each parking location. Specifically, the end drop-off hotspot and parking location of each target route can be matched. The matching process can refer to the matching process between parking location and starting pick-up hotspot described above, and will not be repeated here.

[0132] To make the concept of the embodiments of the present invention easier to understand, please refer to Figure 4 Example shown:

[0133] by Figure 4The following example illustrates the principle. Historical vehicle usage data within a given scheduling duration is clustered to obtain hotspot locations for each preset time period. After generating trip tasks, these trip tasks are matched against each other to obtain a target path set, which includes target path ①, target path ②, target path ③, target path ④, and target path ⑤. Since the boarding hotspot locations S1 and S2 have high popularity, both locations are entered twice during the method execution. Specifically, boarding hotspot location S1 is used twice when constructing the trip task; and boarding hotspot location S2 is matched twice during the acquisition of the continuing route.

[0134] Next, the vehicles to be dispatched at parking locations P1, P2, and P3 are matched to each starting pick-up hotspot location. Since parking location P3 has at least 3 vehicles, parking location P3 was entered 3 times and matched 3 times when matching target vehicles for each starting pick-up hotspot location.

[0135] After matching is completed, the vehicle to be matched at parking location P1 will be dispatched to the starting pick-up hotspot of target route ① during the preset time period TI3; the vehicle to be matched at parking location P2 will be dispatched to the starting pick-up hotspot of target route ② during the preset time period TI1; the two vehicles to be matched at parking location P3 will be dispatched to the starting pick-up hotspots of target routes ③ and ⑤ during the preset time period TI1; and the one vehicle to be matched at parking location P3 will be dispatched to the starting pick-up hotspot of target route ④ during the preset time period TI2.

[0136] Finally, after the ride ends, the drop-off hotspot and parking location at the end of each target route can be matched. After matching, vehicles that have completed target routes ①, ②, and ③ can be parked at parking location P3, vehicles that have completed target route ④ can be parked at parking location P2, and vehicles that have completed target route ⑤ can be parked at parking location P4. This allows the current vehicle dispatch to be completed within the total dispatch time.

[0137] In summary, the vehicle scheduling method provided in this embodiment obtains a target path set by constructing travel tasks and matching these tasks together. Based on this target path set, multiple travel tasks can be completed in the shortest time using the fewest vehicles. Therefore, by using this target path set to guide vehicle scheduling within the target area, it is possible to reduce the number of vehicles to be scheduled and lower the vehicle idle rate while ensuring that vehicles are available at all hotspot locations.

[0138] Please see Figure 5 In another embodiment of the present invention, a vehicle dispatching device 300 is also provided, comprising:

[0139] The hotspot location acquisition module 301 is used to acquire multiple pre-set time period pick-up and drop-off hotspot locations in the target area; the construction module 302 is used to construct multiple trip tasks based on the pick-up and drop-off hotspot locations; the matching module 303 is used to match the multiple trip tasks with each other based on the start and end time periods of the trip tasks, with the shortest total travel time to complete the multiple trip tasks as the matching target, to obtain a target path set; and the scheduling module 304 is used to schedule vehicles in the target area based on the target path set.

[0140] As an optional implementation, the matching module 303 is specifically used for:

[0141] The trip tasks where the drop-off hotspot is located within the target time period are identified as prerequisite tasks, thus obtaining a prerequisite task set; wherein, the target time period is any one of the plurality of preset time periods; the pick-up hotspot locations located within the next preset time period adjacent to the target time period are identified as pick-up locations to be matched, thus obtaining a set of pick-up locations to be matched; with the shortest total travel time as the matching target, the prerequisite task set is matched with the set of pick-up locations to be matched, thus obtaining a set of continuing routes; each continuing route includes the prerequisite task and the corresponding pick-up location to be matched; based on the set of continuing routes for each preset time period, the target route set is obtained.

[0142] As an optional implementation, the matching module 303 is further specifically used for:

[0143] The set of preceding tasks is simulated and matched with the set of boarding locations to be matched to obtain the travel time of each matched virtual path; based on the travel time of the virtual path, the set of preceding tasks is matched with the set of boarding locations to be matched with the shortest total travel time as the matching target to obtain the set of continuing routes.

[0144] As an optional implementation, the matching module 303 is further specifically used for:

[0145] The set of pre-selected tasks is simulated and matched with the set of boarding locations to obtain candidate routes; virtual routes are selected from the candidate routes based on the filtering condition that the travel time of the candidate route is less than the time period of the candidate route, and the travel time of each virtual route is obtained; wherein, the time period is the duration between the start time period and the end time period.

[0146] As an optional implementation, the matching module 303 is further specifically used for:

[0147] Based on the travel time of the virtual path, the weight of the virtual path is determined to obtain a weight set; based on the weight set, the Hungarian algorithm is used to match the preceding task with the boarding location to be matched to obtain the set of continuing routes.

[0148] As an optional implementation, the matching module 303 is further specifically used for:

[0149] Based on the travel time of the virtual path, an original weight set is determined; using the travel time of the virtual path being less than the duration of the virtual path's time period as a filtering condition, target weights are filtered out from the original weight set to obtain the weight set; wherein, the duration of the time period is the duration between the start time period and the end time period.

[0150] As an optional implementation, the hotspot location acquisition module 301 is specifically used for:

[0151] Obtain historical vehicle usage data for multiple preset time periods in the target area; cluster the historical vehicle usage data for each preset time period to obtain the pick-up hotspot location and drop-off hotspot location for each preset time period.

[0152] As an optional implementation, the hotspot location acquisition module 301 is further specifically used for:

[0153] The historical vehicle usage data for each preset time period is clustered to obtain multiple cluster centers; a heat label is generated for each cluster center based on the amount of data corresponding to each cluster center; and the boarding hotspot location and the alighting hotspot location are obtained based on the cluster center and the corresponding heat label.

[0154] As an optional implementation, the construction module 302 is specifically used for:

[0155] Obtain historical trips from historical vehicle usage data; based on the historical trips, match the pick-up hotspot locations and the drop-off hotspot locations to obtain the multiple trip tasks.

[0156] As an optional implementation, the scheduling module 304 is specifically used for:

[0157] For each preset time period, vehicles are dispatched to the starting boarding hotspot location of each target route.

[0158] As an optional implementation, the scheduling module 304 is further specifically used for:

[0159] For each preset time period, the parking locations of multiple vehicles to be dispatched are obtained; the parking locations are matched with the starting pick-up hotspot locations, with the shortest total dispatch time as the matching target, to obtain the target vehicle matched for each starting pick-up hotspot location; wherein, the total dispatch time is the sum of the time for completing vehicle dispatch for each starting pick-up hotspot location; the target vehicle is dispatched to the matched starting pick-up hotspot location.

[0160] It should be noted that the vehicle dispatching device 300 provided in this embodiment of the invention has the same specific implementation and technical effects as the aforementioned method embodiment. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0161] Based on the same inventive concept, another embodiment of the present invention provides an electronic device, including a processor and a memory, wherein the memory is coupled to the processor and stores instructions. When the instructions are executed by the processor, the electronic device performs the steps of any of the methods described in the foregoing method embodiments. It should be noted that in the electronic device provided by the embodiments of the present invention, the specific implementation of each step and the resulting technical effects are the same as in the foregoing method embodiments when the instructions are executed by the processor. For the sake of brevity, any parts not mentioned in the device embodiments can be referred to the corresponding content in the foregoing method embodiments.

[0162] Based on the same inventive concept, another embodiment of the present invention provides a readable storage medium storing a computer program, characterized in that, when executed by a processor, the program implements the steps of any of the methods described in the foregoing method embodiments. It should be noted that, in the readable storage medium provided in the embodiments of the present invention, when the program is executed by a processor, the specific implementation of each step and the resulting technical effects are the same as in the foregoing method embodiments. For the sake of brevity, any parts not mentioned in the device embodiments can be referred to the corresponding content in the foregoing method embodiments.

[0163] The term "and / or" as used herein is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects are in an "or" relationship; the word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of multiple such elements. This invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims listing several means, several of these means can be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0164] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0165] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0166] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0167] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0168] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0169] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A vehicle dispatching method, characterized in that, include: Obtain the pick-up and drop-off hotspot locations for multiple preset time periods in the target area; Based on the boarding hotspot location and the alighting hotspot location, multiple trip tasks are constructed, including: Retrieve historical trips from historical vehicle usage data; Based on the historical trips, the boarding hotspot location and the alighting hotspot location are matched to obtain the multiple trip tasks, including: if the number of historical trips between the boarding hotspot location and the alighting hotspot location is greater than the number threshold, then a trip task is constructed from the boarding hotspot location and the alighting hotspot location. Based on the start and end times of the trip tasks, and with the shortest total travel time to complete the multiple trip tasks as the matching target, the multiple trip tasks are matched with each other to obtain a target path set; Based on the target path set, vehicles are dispatched within the target area.

2. The method according to claim 1, characterized in that, Based on the start and end times of the trip tasks, and with the shortest total travel time for completing the multiple trip tasks as the matching target, the multiple trip tasks are matched against each other to obtain a target path set, including: The trip tasks where the drop-off hotspot is within the target time period are identified as prerequisite tasks, thus obtaining a set of prerequisite tasks; wherein, the target time period is any one of the plurality of preset time periods; The boarding hotspot locations in the next preset time period adjacent to the target time period are determined as boarding locations to be matched, thus obtaining a set of boarding locations to be matched; Using the shortest total travel time as the matching target, the set of preceding tasks is matched with the set of pick-up locations to be matched to obtain a set of continuing routes; each continuing route includes the preceding task and the corresponding pick-up location to be matched. The target route set is obtained based on the set of continuing routes for each preset time period.

3. The method according to claim 2, characterized in that, The process of matching the set of preceding tasks with the set of boarding locations to be matched, using the shortest total travel time as the matching target, to obtain a set of continuing routes includes: The set of pre-tasks is simulated and matched with the set of boarding locations to be matched, and the travel time of each matched virtual path is obtained. Based on the travel time of the virtual route, with the shortest total travel time as the matching target, the set of preceding tasks is matched with the set of boarding locations to be matched to obtain the set of continuing routes.

4. The method according to claim 3, characterized in that, The step of simulating the matching of the pre-task set with the set of boarding locations to be matched to obtain the travel time of each virtual path includes: The set of preceding tasks is simulated and matched with the set of boarding locations to be matched to obtain candidate paths; Using the condition that the travel time of the candidate route is less than the time period of the candidate route as a filtering condition, virtual routes are filtered from the candidate routes, and the travel time of each virtual route is obtained. The duration of the time period is the duration between the start time period and the end time period.

5. The method according to claim 3, characterized in that, The method of matching the preceding task set with the set of boarding locations to be matched based on the travel time of the virtual route, with the shortest total travel time as the matching target, to obtain the set of continuing routes, includes: Based on the travel time of the virtual path, the weight of the virtual path is determined, and a weight set is obtained; Based on the weight set, the Hungarian algorithm is used to match the preceding task with the boarding location to be matched, thereby obtaining the set of continuing routes.

6. The method according to claim 5, characterized in that, The process of determining the weight of the virtual path based on its travel time, and obtaining the weight set, includes: The original weight set is determined based on the travel time of the virtual path; Using the fact that the travel time of the virtual path is less than the duration of the time segment of the virtual path as a filtering condition, target weights are filtered out from the original weight set to obtain the weight set; The duration of the time period is the duration between the start time period and the end time period.

7. The method according to claim 1, characterized in that, The acquisition of boarding and alighting hotspot locations in the target area for multiple preset time periods includes: Acquire historical vehicle usage data for multiple preset time periods in the target area; Cluster the historical vehicle usage data for each preset time period to obtain the pick-up hotspot locations and drop-off hotspot locations for each preset time period.

8. The method according to claim 7, characterized in that, The step of clustering the historical vehicle usage data for each preset time period to obtain the pick-up hotspot locations and drop-off hotspot locations for each preset time period includes: The historical vehicle usage data for each preset time period are clustered to obtain multiple cluster centers; Based on the amount of data corresponding to each cluster center, a heat label is generated for each cluster center; Based on the cluster center and the corresponding heat tag, the boarding hotspot location and the alighting hotspot location are obtained.

9. The method according to claim 1, characterized in that, The step of dispatching vehicles within the target area based on the target path set includes: For each preset time period, vehicles are dispatched to the starting boarding hotspot location of each target route.

10. The method according to claim 9, characterized in that, The step of dispatching vehicles to the starting pick-up hotspot location of each target route for each preset time period includes: For each preset time period, obtain the parking locations of multiple vehicles to be dispatched; The parking locations and the starting pick-up hotspot locations are matched, with the shortest total dispatch time as the matching target, to obtain the target vehicle matched for each starting pick-up hotspot location; wherein, the total dispatch time is the sum of the time required to complete vehicle dispatch for each starting pick-up hotspot location; The target vehicle is dispatched to the matching starting boarding hotspot location.

11. A vehicle dispatching device, characterized in that, include: The hotspot location acquisition module is used to acquire the boarding and alighting hotspot locations in the target area for multiple preset time periods; A construction module is used to construct multiple trip tasks based on the boarding hotspot location and the alighting hotspot location; The matching module is used to match the multiple trip tasks with each other based on the start and end times of the trip tasks, with the shortest total travel time to complete the multiple trip tasks as the matching target, to obtain a target path set; The scheduling module is used to schedule vehicles within the target area based on the target path set; The construction module is specifically used to obtain historical trips from historical vehicle usage data; based on the historical trips, the pick-up hotspot locations and the drop-off hotspot locations are matched to obtain the multiple trip tasks, including: if the number of historical trips between the pick-up hotspot location and the drop-off hotspot location is greater than a number threshold, then a trip task is constructed from the pick-up hotspot location and the drop-off hotspot location.

12. An electronic device, characterized in that, The device includes a processor and a memory, the memory being coupled to the processor, the memory storing instructions that, when executed by the processor, cause the electronic device to perform the steps of the method according to any one of claims 1-10.

13. A readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method described in any one of claims 1-10.

Images