Parking scheduling method and device for autonomous vehicle, and medium

Abstract

The invention relates to the technical field of automatic driving, in particular to a parking scheduling method and device for an automatic driving vehicle and a medium, and aims to solve the problem of low response efficiency of a vehicle parking scheduling scheme in related technologies. The method comprises the following steps: acquiring temporary parking areas of each target area before m vehicle demand prediction values in a future time period in a preset area, wherein each temporary parking area is used for parking an automatic driving vehicle which is planned to serve the vehicle demand in the corresponding target area in the future time period; based on the vehicle demand prediction value of each target area and the road network distance from the fixed parking area and each temporary parking area to each target area, evaluating response values of the vehicle parking in different parking areas; and based on each response value and the current position of each to-be-dispatched autonomous vehicle, the matching combination of each autonomous vehicle and different parking areas is evaluated, and each autonomous vehicle is determined and dispatched to a corresponding target parking area, so that the vehicle response efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of autonomous driving technology, and in particular to a parking scheduling method, device and medium for autonomous vehicles. Background Technology

[0002] With the rapid development of technology, autonomous vehicles, such as Robotaxi vehicles, are gradually appearing in people's daily lives. Typically, during operation, when demand is low, especially at night when demand is low, autonomous vehicles are automatically dispatched back to parking lots. During peak demand periods, they are then dispatched to areas with higher demand to provide service. However, due to cost considerations, these parking lots are often located in relatively remote, non-core areas. Therefore, this traditional parking dispatching scheme inevitably results in low vehicle response efficiency, and the empty-run dispatching also incurs time and operating costs. Summary of the Invention

[0003] This application provides a parking scheduling method, device, and medium for autonomous vehicles, which can park on-site or be pre-scheduled to temporary parking areas in areas with predicted peak future vehicle demand, thereby improving vehicle response efficiency and reducing vehicle scheduling time and driving costs.

[0004] The specific technical solutions provided in this application are as follows:

[0005] In a first aspect, embodiments of this application provide a parking scheduling method for autonomous vehicles, comprising:

[0006] Get the temporary parking areas of each target area in the preset area with the top m predicted values ​​of vehicle demand in the future time period. Each temporary parking area is used to park autonomous vehicles that are planned to serve the vehicle demand in the corresponding target area in the future time period. m is a positive integer not less than 1.

[0007] Based on the predicted vehicle demand values ​​corresponding to each target area, and the road network distances from fixed parking areas and each temporary parking area to each target area, the response values ​​of vehicles parked in different parking areas are evaluated.

[0008] Based on each response value and the current location of each autonomous vehicle to be scheduled, the matching combination of each autonomous vehicle with the fixed parking area and the temporary parking area is evaluated to determine the target parking area corresponding to each autonomous vehicle, and the autonomous vehicles are scheduled to the corresponding target parking area.

[0009] In the above method, by acquiring temporary parking areas in each target region and evaluating the response of vehicles parking in different parking areas, response values ​​are obtained. Each response value reflects the value of the corresponding parking area and the spatial correlation between the parking area and areas with high user demand in the future (hot zones). Then, based on each response value and the current location of each autonomous vehicle to be dispatched, it is determined which autonomous vehicle will park in a fixed parking area and which will park in a temporary parking area. This allows autonomous vehicles to be parked nearby or dispatched in advance to target areas with high predicted demand, thereby improving vehicle response efficiency. At the same time, since autonomous vehicles are parked locally or dispatched in advance to temporary parking areas in target areas with high predicted demand, vehicle dispatching costs can be reduced, improving the overall operational efficiency of the platform.

[0010] In one possible implementation, before obtaining the temporary parking areas of each target area up to the top m of the predicted vehicle demand values ​​for a future time period in the preset region, the method further includes:

[0011] Determine that the parking dispatch triggering conditions are met, wherein the parking dispatch triggering conditions include:

[0012] The current demand for vehicles is less than the preset demand threshold; and / or,

[0013] The ratio of the number of autonomous vehicles in an empty state to the total number of vehicles is greater than a preset ratio.

[0014] The above method enables timely parking scheduling of autonomous vehicles by real-time detection of vehicle demand and / or the proportion of vehicles in an empty state.

[0015] In one possible implementation, after determining that the parking dispatch triggering condition is met, and before obtaining the temporary parking areas of each target area among the top m predicted vehicle demand values ​​for a future time period in the preset area, the method further includes:

[0016] The autonomous vehicles that meet the operating conditions are obtained, wherein the operating conditions include the vehicle's battery level being greater than a preset battery level, the vehicle's cleanliness being greater than a preset cleanliness level, and the vehicle having no abnormal conditions.

[0017] The vehicle demand in each region of the preset area is predicted during the future time period to obtain the predicted vehicle demand value for each region.

[0018] Based on the predicted vehicle demand values ​​for each region, the regions with the highest predicted vehicle demand values ​​are identified as the target regions.

[0019] The above method, by acquiring information on autonomous vehicles that meet operational conditions and target areas with high projected demand in the future, facilitates the subsequent dispatch of autonomous vehicles that meet operational conditions to high-demand areas for parking. This ensures that autonomous vehicles can quickly provide online services during peak demand periods in the future, reducing vehicle return (to fixed parking areas), dispatch time, and driving costs, thereby improving overall operational efficiency.

[0020] In one possible implementation, obtaining the temporary parking areas of each target area that are m ahead of the predicted vehicle demand values ​​for a future time period in a preset region includes:

[0021] Perform the following operations for each target region:

[0022] For any target area among the target areas, a temporary parking area within that target area is obtained based on road network information;

[0023] If the number of parking spaces in the obtained temporary parking area is less than the predicted demand for vehicles in any target area, then based on the road network information, a temporary parking area within a preset range outside any target area is obtained to obtain the number of parking spaces in any target area that is the predicted demand for vehicles.

[0024] The above method can obtain temporary parking areas near each target area through road network information, in order to prepare for subsequent evaluation of the response value of temporary parking areas and evaluation of matching combinations of different parking areas and vehicles.

[0025] In one possible implementation, evaluating the response value of vehicles parking in different parking areas based on the predicted vehicle demand values ​​corresponding to each target area and the road network distances from fixed parking areas and each temporary parking area to each target area includes:

[0026] Based on road network information, a first road network distance from the fixed parking area to each target area is determined, and a second road network distance from each temporary parking area to each target area is determined based on the road network information.

[0027] Based on the attenuation coefficient, the predicted vehicle demand for each target area, and their respective first road network distances, the response value of a vehicle parking in the fixed parking area is determined; and based on the attenuation coefficient, the predicted vehicle demand for each target area, and their respective second road network distances, the response value of a vehicle parking in each temporary parking area is determined.

[0028] The above method assesses the value of parking areas by considering their distance from the road network. This value depends on their spatial correlation with areas of high future demand for vehicles. The closer the distance, the shorter the response time for vehicles to depart from the parking area, thereby improving operational efficiency and user satisfaction.

[0029] In one possible implementation, before evaluating the matching combinations of each autonomous vehicle with the fixed parking area and the temporary parking areas based on each response value and the current location of each autonomous vehicle to be scheduled, and determining the target parking area corresponding to each autonomous vehicle, the method further includes:

[0030] The response values ​​of each temporary parking area are determined to be greater than the response values ​​of the fixed parking area.

[0031] The above method, after obtaining the response values ​​of each temporary parking area and fixed parking area, compares the response values ​​of each temporary parking area with those of the fixed parking area, and filters the obtained temporary parking areas to ensure that the value of the temporary parking areas is higher than that of the fixed parking areas, thereby ensuring that the subsequently allocated temporary parking areas are beneficial to the vehicle response efficiency.

[0032] In one possible implementation, the evaluation of matching combinations between each autonomous vehicle and the fixed parking area and the temporary parking areas based on each response value and the current location of each autonomous vehicle to be scheduled, to determine the target parking area corresponding to each autonomous vehicle, includes:

[0033] For each autonomous vehicle, the following operations are performed: For any one of the autonomous vehicles, the driving cost of the autonomous vehicle from the current location to the fixed parking area and each of the temporary parking areas is determined, as well as the parking cost of each autonomous vehicle parking in each of the temporary parking areas, wherein the parking cost of the fixed parking area is a preset value.

[0034] Based on the response values, driving costs, and parking costs, a preset vehicle scheduling optimization model is used to optimize and evaluate the matching combinations of each autonomous vehicle with the fixed parking area and the temporary parking area, thereby obtaining the target parking area corresponding to each autonomous vehicle.

[0035] The above method iterates through the matching combinations of each autonomous vehicle with fixed parking areas and temporary parking areas by using each response value and the driving cost and parking cost of each autonomous vehicle. This results in the matching combination of each autonomous vehicle with parking area (fixed parking area or any temporary parking area) that maximizes the objective function (representing the total net benefit) of the vehicle scheduling optimization model. The parking area in the matching combination is then used as the target parking area for the corresponding autonomous vehicle to maximize the overall net benefit.

[0036] In one possible implementation, determining the driving cost of any autonomous vehicle traveling from its current location to the fixed parking area and each of the temporary parking areas, and the parking cost of each autonomous vehicle parking in the fixed parking area and each of the temporary parking areas, respectively, includes:

[0037] Based on the current location and unit distance driving cost of any autonomous vehicle, determine the driving cost of any autonomous vehicle from the current location to the fixed parking area, and the driving cost from the current location to each of the temporary parking areas.

[0038] Based on the unit-time parking cost of each temporary parking area, the parking cost of each autonomous vehicle parked in each temporary parking area is obtained.

[0039] The above method determines the total scheduling cost of each autonomous vehicle based on the unit distance driving cost and the unit duration parking cost, thus preparing for the subsequent calculation of net revenue and the sum of net revenue.

[0040] Secondly, embodiments of this application provide a parking scheduling device for autonomous vehicles, comprising:

[0041] The acquisition module is used to acquire the temporary parking areas of each target area in the preset area with the top m predicted values ​​of vehicle demand in the future time period. Here, a single temporary parking is used to park autonomous vehicles that are planned to serve the vehicle demand in the corresponding target area in the future time period, and m is a positive integer not less than 1.

[0042] The response prediction module is used to evaluate the response value of vehicles parked in different parking areas based on the predicted vehicle demand value corresponding to each target area and the road network distance from the fixed parking area and each temporary parking area to each target area.

[0043] The parking scheduling module is used to evaluate the matching combination of each autonomous vehicle with the fixed parking area and the temporary parking area based on each response value and the current location of each autonomous vehicle to be scheduled, determine the target parking area corresponding to each autonomous vehicle, and schedule each autonomous vehicle to the corresponding target parking area.

[0044] In one possible implementation, before obtaining the temporary parking areas of each target area among the top m predicted vehicle demand values ​​for a future time period in the preset region, the acquisition module is further configured to:

[0045] Determine that the parking dispatch triggering conditions are met, wherein the parking dispatch triggering conditions include:

[0046] The current demand for vehicles is less than the preset demand threshold; and / or,

[0047] The ratio of the number of autonomous vehicles in an empty state to the total number of vehicles is greater than a preset ratio.

[0048] In one possible implementation, after determining that the parking dispatch triggering condition is met, and before obtaining the temporary parking areas of each target area that are the top m of the predicted vehicle demand values ​​for the future time period in the preset area, the acquisition module is further configured to:

[0049] The autonomous vehicles that meet the operating conditions are obtained, wherein the operating conditions include the vehicle's battery level being greater than a preset battery level, the vehicle's cleanliness being greater than a preset cleanliness level, and the vehicle having no abnormal conditions.

[0050] The vehicle demand in each region of the preset area is predicted during the future time period to obtain the predicted vehicle demand value for each region.

[0051] Based on the predicted vehicle demand values ​​for each region, the regions with the highest predicted vehicle demand values ​​are identified as the target regions.

[0052] In one possible implementation, the acquisition module is specifically used for:

[0053] Perform the following operations for each target region:

[0054] For any target area among the target areas, a temporary parking area within that target area is obtained based on road network information;

[0055] If the number of parking spaces in the obtained temporary parking area is less than the predicted demand for vehicles in any target area, then based on the road network information, a temporary parking area within a preset range outside any target area is obtained to obtain the number of parking spaces in any target area that is the predicted demand for vehicles.

[0056] In one possible implementation, the response prediction module is specifically used for:

[0057] Based on road network information, a first road network distance from the fixed parking area to each target area is determined, and a second road network distance from each temporary parking area to each target area is determined based on the road network information.

[0058] Based on the attenuation coefficient, the predicted vehicle demand for each target area, and their respective first road network distances, the response value of a vehicle parking in the fixed parking area is determined; and based on the attenuation coefficient, the predicted vehicle demand for each target area, and their respective second road network distances, the response value of a vehicle parking in each temporary parking area is determined.

[0059] In one possible implementation, before evaluating the matching combinations of each autonomous vehicle with the fixed parking area and the temporary parking areas based on each response value and the current location of each autonomous vehicle to be scheduled, and determining the target parking area corresponding to each autonomous vehicle, the response prediction module is further configured to:

[0060] The response values ​​of each temporary parking area are determined to be greater than the response values ​​of the fixed parking area.

[0061] In one possible implementation, the parking scheduling module is specifically used for:

[0062] For each autonomous vehicle, the following operations are performed: For any one of the autonomous vehicles, the driving cost of the autonomous vehicle from the current location to the fixed parking area and each of the temporary parking areas is determined, as well as the parking cost of each autonomous vehicle parking in each of the temporary parking areas, wherein the parking cost of the fixed parking area is a preset value.

[0063] Based on the response values, driving costs, and parking costs, a preset vehicle scheduling optimization model is used to optimize and evaluate the matching combinations of each autonomous vehicle with the fixed parking area and the temporary parking area, thereby obtaining the target parking area corresponding to each autonomous vehicle.

[0064] In one possible implementation, the parking scheduling module is specifically used for:

[0065] Based on the current location and unit distance driving cost of any autonomous vehicle, determine the driving cost of any autonomous vehicle from the current location to the fixed parking area, and the driving cost from the current location to each of the temporary parking areas.

[0066] Based on the unit-time parking cost of each temporary parking area, the parking cost of each autonomous vehicle parked in each temporary parking area is obtained.

[0067] Thirdly, embodiments of this application provide an electronic device, including:

[0068] Memory is used to store computer programs or instructions;

[0069] A processor for executing a computer program or instructions in the memory, such that the method described in any of the first aspects is performed.

[0070] Fourthly, embodiments of this application provide a computer-readable storage medium that, when instructions in the storage medium are executed by a processor, enables the processor to perform the method described in any one of the first aspects above.

[0071] Fifthly, embodiments of this application provide a computer program product comprising: computer program code, which, when executed on a computer, causes the computer to perform the method described in any one of the first aspects.

[0072] Furthermore, the technical effects of any of the implementation methods in the second to fifth aspects can be found in the technical effects of different implementation methods in the first aspect, and will not be repeated here.

[0073] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0074] Figure 1 This is a schematic diagram of an optional application scenario in the embodiments of this application;

[0075] Figure 2 This is a flowchart illustrating a parking scheduling method for an autonomous vehicle according to an embodiment of this application.

[0076] Figure 3 This is a schematic diagram of a process for obtaining each autonomous vehicle to be scheduled and each target area in an embodiment of this application;

[0077] Figure 4 This is a schematic diagram of a process for obtaining temporary parking areas for each target area in an embodiment of this application;

[0078] Figure 5 This is a schematic diagram illustrating how a temporary parking area for target area 1 is obtained in an embodiment of this application.

[0079] Figure 6 This is a schematic diagram of a process for determining the response value of any parking area in an embodiment of this application;

[0080] Figure 7 This is a schematic diagram of a vehicle dispatch optimization process in an embodiment of this application;

[0081] Figure 8 This is a schematic diagram of the overall process of parking scheduling for an autonomous vehicle according to an embodiment of this application;

[0082] Figure 9 This is a schematic diagram of the logical architecture of a parking scheduling device for an autonomous vehicle according to an embodiment of this application;

[0083] Figure 10 This is a schematic diagram of the physical architecture of an electronic device according to an embodiment of this application. Detailed Implementation

[0084] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0085] It should be noted that the terms "first," "second," "third," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein.

[0086] The preferred embodiments of this application will be further described in detail below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only for illustration and explanation of this application and are not intended to limit this application. Furthermore, the embodiments of this application and the features in the embodiments can be combined with each other without conflict.

[0087] See Figure 1The diagram shown illustrates an optional application scenario in this application embodiment. This application scenario may include a user terminal 10, an electronic device 20, and a vehicle-mounted terminal 30. The user terminal 10 can be connected to the electronic device 20 via a wired or wireless network, and the vehicle-mounted terminal 30 can also be connected to the electronic device 20 via a wired or wireless network.

[0088] User terminal 10 can be used as an order placement terminal. It is equipped with order placement software. Users can send their car usage requests to the server through the order placement software. The car usage requests include at least the device identifier, location information, and departure time of the order placement terminal.

[0089] The electronic device 20 can be a server or a terminal device. In some embodiments, after receiving a car-hailing request from any ordering end, the electronic device 20 can allocate a vehicle to the user based on the location information and departure time in the car-hailing request, and send a dispatch instruction to the order-receiving end so that the order-receiving end can provide the user with the corresponding service. The dispatch instruction includes order information such as the order-receiving location, delivery location, and contact person.

[0090] The vehicle terminal 30 can serve as the order receiving terminal mentioned above, used to receive the order dispatch instructions sent by the electronic device 20, and provide corresponding services according to the order information in the order dispatch instructions.

[0091] It should be noted that, Figure 1 The location of the vehicle-mounted terminal 30 in the figure is for illustrative purposes only and does not represent its actual fixed location within the vehicle. In other embodiments, the vehicle-mounted terminal 30 may be located in other parts of the vehicle, such as at the door or between the front seats.

[0092] In some embodiments, the aforementioned server may be an independent physical server, a server cluster composed of multiple physical servers, or a distributed system that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0093] In other embodiments, the aforementioned terminal device may be, for example, a smart terminal, a smart mobile terminal, a tablet computer, a laptop computer, a smart handheld device, a personal computer (PC), a computer, a smart screen, a display device, a personal digital assistant (PDA), etc.

[0094] It is understood that this application does not limit the specific type of the aforementioned electronic device.

[0095] With the rapid development of technology, autonomous vehicles are gradually appearing in people's daily lives, such as Robotaxi vehicles. The vehicle equipped with the on-board terminal 30 mentioned above can be an autonomous vehicle. Of course, the vehicle equipped with the on-board terminal 30 mentioned above can also be a traditional vehicle. After determining the target parking area corresponding to each autonomous vehicle, the method in this embodiment can push the information of the determined target parking area to the driver of the corresponding traditional vehicle to provide the driver with assisted parking suggestions.

[0096] After introducing one optional application scenario in the embodiments of this application, the design concept of the embodiments of this application will be briefly introduced below.

[0097] This application relates to the field of autonomous driving technology, and mainly to a parking scheduling method, device and medium for autonomous vehicles.

[0098] In related technologies, autonomous vehicles typically operate during periods of low demand, particularly at night. When service is slow, vehicles are usually centrally dispatched back to depots (designated parking areas for vehicle maintenance and cleaning, e.g., Robotaxi vehicles). During peak demand periods, vehicles are then reassigned to designated areas to provide service. However, to reduce parking costs, these designated parking areas are often located in relatively remote, non-core areas. This traditional parking dispatching scheme inevitably leads to lower vehicle response efficiency, and the empty-run dispatching also incurs time and travel costs.

[0099] To address the low vehicle response efficiency issue in the aforementioned vehicle parking scheduling schemes, this application provides a parking scheduling method for autonomous vehicles. In this application, the method involves obtaining temporary parking areas for each target area within a preset region, prior to the predicted vehicle demand over a future time period (value m). Each temporary parking area is used to park an autonomous vehicle scheduled to serve the vehicle demand within the corresponding target area in the future time period, and m is a positive integer not less than 1. Based on the predicted vehicle demand for each target area and the road network distances from the fixed parking area and each temporary parking area to each target area, the response values ​​for parking in different parking areas are evaluated. Then, based on each response value and the current location of each autonomous vehicle to be scheduled, the matching combination of each autonomous vehicle with the fixed parking area and each temporary parking area is evaluated to determine the target parking area corresponding to each autonomous vehicle, and each autonomous vehicle is scheduled to its corresponding target parking area.

[0100] In this embodiment, by predicting the vehicle demand of a preset area in the future time period, the top m target areas with the highest predicted vehicle demand in the future time period are determined. Then, temporary parking areas in the target areas are obtained, and the response of vehicles parking in different parking areas is evaluated to obtain response values. In this way, each response value can reflect the value of the parking area corresponding to the response value and reflect the spatial correlation between the parking area and each target area, i.e., the future peak user demand area (hot zone). Then, based on each response value and the current location of each autonomous vehicle to be dispatched, it is determined which autonomous vehicles will be parked in fixed parking areas and which autonomous vehicles will be parked in temporary parking areas. This achieves the goal of parking autonomous vehicles nearby or dispatching them in advance to target areas with high predicted vehicle demand, thereby improving vehicle response efficiency. At the same time, since autonomous vehicles are parked locally or dispatched in advance to temporary parking areas in target areas with high predicted vehicle demand, vehicle dispatching costs can be reduced and the overall operational efficiency of the platform can be improved.

[0101] The following refers to the application scenarios described in the embodiments of this application. Figure 2 As shown, the specific process of a parking scheduling method for an autonomous vehicle according to an embodiment of this application is described in detail below:

[0102] Step 200: Obtain the temporary parking areas of each target area that are the top m in the predicted vehicle demand value for the future time period in the preset area. Here, a single temporary parking area is used to park autonomous vehicles that are planned to serve the vehicle demand in the corresponding target area in the future time period, and m is a positive integer not less than 1.

[0103] In some embodiments of this application, before executing step 200, it is necessary to determine whether a preset parking dispatch triggering condition is met, wherein the parking dispatch triggering condition includes some or all of the following conditions:

[0104] Condition 1: The current demand for vehicles is less than the preset demand threshold.

[0105] Condition 2: The ratio of the number of autonomous vehicles in an empty state to the total number of vehicles is greater than a preset ratio.

[0106] In some embodiments of this application, combined with Figure 1 In the application scenario shown, the server can monitor the current demand for vehicles in real time, and / or the ratio of the number of autonomous vehicles in an empty state to the total number of vehicles. If it is determined that the current demand for vehicles is less than the preset demand threshold, and / or the above ratio is greater than the preset ratio, it means that the current demand for vehicles is in a low period, with fewer orders and a large number of empty vehicles, thereby determining that the parking scheduling trigger condition is met and triggering the execution step 200.

[0107] In some embodiments, after determining that the parking dispatch triggering conditions are met, before executing step 200, refer to Figure 3 As shown, the following steps also need to be performed:

[0108] Step 301: Obtain each autonomous vehicle that meets the operating conditions, wherein the operating conditions include the vehicle's battery level being greater than the preset battery level, the vehicle's cleanliness being greater than the preset cleanliness, and the vehicle having no abnormal conditions.

[0109] In this embodiment of the application, by executing step 301, the real situation of the currently operating autonomous vehicles can be obtained, thereby determining the vehicles that can operate normally, that is, the vehicles that meet the operating conditions, that is, the autonomous vehicles to be dispatched in the future, so as to make reasonable parking and dispatching of the vehicles that can operate normally in the future, which can improve the overall operating efficiency and user satisfaction.

[0110] Step 302: Predict the vehicle demand in each area of ​​the preset region in the future time period to obtain the predicted vehicle demand value for each area.

[0111] In specific implementation, when performing step 302, the prediction model in the relevant technology can be used to predict the vehicle demand of each area in the preset area in the future time period based on the historical data of the preset area, so as to obtain the predicted value of vehicle demand for each area.

[0112] In this process, a day can be pre-divided into multiple time periods, denoted as time period t, where t = 1, 2, ..., T, and T represents the total number of time periods. Similarly, a pre-defined region can be divided into multiple areas, denoted as areas d, where d = 1, 2, ..., M, and M represents the total number of areas within that region. Therefore, the predicted vehicle demand for area d in future time period t can be represented by D. d To express.

[0113] As one possible implementation, for example, based on historical order data, such as regional order volume / cancellation rate / average waiting time (which can be sliced ​​into 15-minute segments), time (e.g., hours, days of the week, holidays), weather (e.g., temperature / precipitation / wind speed / weather level (smog, heavy rain, etc.)), special events (e.g., concerts / sports events (which can be represented by 0 / 1 flags), event scale (e.g., graded values), and point of interest (POI) distributions (e.g., commercial / residential / transportation hub density (kernel density estimation), regional functional mixing), a spatiotemporal graph attention long short-term memory network (ST-GAT-LSTM) hybrid model can be used to predict the vehicle demand in each region d within a future time period t, thus obtaining the predicted vehicle demand value D for each region. d.

[0114] Step 303: Based on the predicted vehicle demand values ​​for each region, the regions with the highest predicted vehicle demand values ​​are identified as the target regions.

[0115] In practice, after obtaining the predicted vehicle demand values ​​for each region, step 303 is executed. Specifically, the regions can be sorted in descending order of the predicted vehicle demand values, and the top m regions are selected to obtain the target regions mentioned above.

[0116] In this embodiment of the application, when performing step 200, refer to... Figure 4 As shown, the following steps can be performed for each target area to obtain the temporary parking areas near each target area, such as roadside parking areas and related information, such as geographical location, capacity, parking time restrictions, and pricing:

[0117] Step 2001: For any target area among the above target areas, obtain the temporary parking area within that target area based on road network information;

[0118] Step 2002: If the number of parking spaces in the obtained temporary parking area is less than the predicted demand for vehicles in the target area, then based on the road network information, obtain temporary parking areas within a preset range outside the target area to obtain the number of parking spaces in the predicted demand for vehicles in the target area.

[0119] like Figure 5 As shown, the temporary parking area is the roadside parking area as an example.

[0120] For target area 1, when obtaining the roadside parking area of ​​target area 1, the roadside parking area within target area 1 is first obtained based on road network information. If the number of parking spaces in the obtained roadside parking area is less than the predicted value of the vehicle demand in target area 1, then the roadside parking area outside target area 1 within a preset range is obtained by expanding the range. The radius or specific shape of the preset range can be set according to actual business needs, and this application does not make specific limitations.

[0121] In this way, by performing the above steps 2001 and 2002, available temporary parking areas near each target area with a high predicted value of vehicle demand in the future time period can be collected. This will facilitate the subsequent quantification of the vehicle's response to vehicle demand when parking in different parking areas, i.e., fixed parking areas and temporary parking areas, i.e., the subsequent response value.

[0122] In other embodiments of this application, when collecting information on temporary parking areas, they can be compared with fixed parking areas. If the road network distance from a target area to a fixed parking area is less than the road network distance to a temporary parking area, since the fixed parking area is the aforementioned depot where vehicles are defaulted to be centrally dispatched after stopping operation, it is usually a short-term or long-term leased area. Therefore, it is a paid parking area, and parking a vehicle in the fixed parking area will not incur parking fees or parking costs. Therefore, in this case, vehicles can be prioritized to be parked in the fixed parking area, i.e., the temporary parking area can be abandoned.

[0123] Step 210: Based on the predicted vehicle demand values ​​corresponding to each of the above target areas, and the road network distances from the fixed parking areas and each temporary parking area to each of the above target areas, evaluate the response values ​​of vehicles parking in different parking areas.

[0124] Different parking areas can affect the vehicle's response to usage demand. For example, if a vehicle is parked in a distant parking area the previous day, the response time to usage demand the following day will be longer, resulting in a lower response value. Consequently, the scheduling cost and empty running cost of the autonomous vehicle will also increase. Therefore, in this embodiment, step 210 is performed to quantify the vehicle's response to usage demand in different parking areas, i.e., to obtain the response value. This response value reflects the value of different parking areas. The value of a parking area depends on its spatial correlation with each target area, i.e., the peak usage demand area. That is, the closer to the peak demand area (hot zone), the shorter the response time of the vehicle departing from that parking area, thereby improving response efficiency, operational efficiency, and user satisfaction.

[0125] In this embodiment, the response value of each parking area can be quantified using the following formula:

[0126]

[0127] Where, r j The response value of parking area j is represented. From an economic perspective, it can also be called the revenue score. The higher the value, the greater the value of the parking area to the operation and the greater the revenue. j≥0, where j=0 represents a fixed parking area and j>0 represents the temporary parking areas mentioned above.

[0128] D d (t) represents the predicted vehicle demand for region d in time period t;

[0129] distance(j,d) represents the road network distance from parking area j to target area d. In this embodiment, it can be obtained through the map application programming interface (API).

[0130] λ is the attenuation coefficient, λ>0, and in some embodiments of this application, it can be set to λ=0.1;

[0131] m is the total number of target regions.

[0132] As can be seen from the above formula, through e -λ·distance(j,d) This simulates the attenuation effect of road network distance on response values. The closer the road network distance (the smaller distance(j,d)), the closer the exponent term is to 1, and the larger the response value. Conversely, the farther the road network distance (the larger distance(j,d)), the closer the exponent term is to 0, and the smaller the response value. In this way, by obtaining the response value of each parking area j, the geographical location value of each parking area can be reflected.

[0133] In specific implementation, when performing step 210, refer to the above formula and Figure 6 As shown, the response values ​​of a vehicle parked in different parking areas can be evaluated by performing the following steps:

[0134] Step 2101: Based on road network information, determine the first road network distance from the fixed parking area to each of the above target areas, and based on road network information, determine the second road network distance from each of the above temporary parking areas to each of the above target areas.

[0135] Step 2102: Based on the attenuation coefficient, the predicted vehicle demand for each target area and their respective first road network distance, determine the response value of the vehicle parking in the fixed parking area; and based on the attenuation coefficient, the predicted vehicle demand for each target area and their respective second road network distance, determine the response value of the vehicle parking in each of the above-mentioned temporary parking areas respectively.

[0136] For example, suppose there are 3 target areas in the preset area, which are 3 hot spots for vehicle demand in the future time period. The predicted vehicle demand values ​​for each area are D1=200, D2=150, and D3=100, respectively. The road network distances from temporary parking areas and fixed parking areas to each target area are shown in the table below:

[0137] Table 1. Road network distances from different parking areas to various target areas.

[0138] Distance from the target area by road network 1 Distance to target area via two-way network Distance to target area via 3-way network Temporary parking area 1 2km 5km 8km Temporary parking area 2 5 km 8km 6km Fixed parking areas 6 km 3km 4km

[0139] Using the above formula, assuming λ = 0.1, taking temporary parking area 1 in the table above as an example, the response value r1 for vehicles parking in temporary parking area 1 can be calculated:

[0140] r1 = 200 × e -0.1×2 +150×e -0.1×5 +100×e -0.1×8

[0141] = 200 × 0.8187 + 150 × 0.6065 + 100 × 0.4493

[0142] ≈163.7 + 91.0 + 44.9 = 299.6;

[0143] Similarly, the response values ​​for temporary parking area 2 and fixed parking area in the table above can be calculated, and will not be listed here again.

[0144] In this embodiment of the application, after determining the response value of each temporary parking area and fixed parking area, in order to ensure that the value of the temporary parking area is higher than that of the fixed parking area and to ensure that the subsequent allocation of temporary parking areas is beneficial to the vehicle response efficiency, after executing step 210 and before executing step 220, it is also necessary to compare the response value of each temporary parking area with the response value of the fixed parking area to determine that the response value of each temporary parking area is greater than that of the fixed parking area.

[0145] In other embodiments of this application, if the response value of any of the temporary parking areas is not greater than the response value of the fixed parking area, the temporary parking area is deleted. In this way, the obtained temporary parking areas can be screened through the above operation, ensuring that the value of each temporary parking area participating in the subsequent evaluation is higher than the value of the fixed parking area, so as to maximize the overall operational efficiency.

[0146] Step 220: Based on each response value and the current location of each autonomous vehicle to be scheduled, evaluate the matching combination of each autonomous vehicle with the fixed parking area and each temporary parking area, determine the target parking area corresponding to each autonomous vehicle, and schedule each autonomous vehicle to the corresponding target parking area.

[0147] In this embodiment of the application, a vehicle scheduling optimization model is pre-set, which can be represented by the following formula:

[0148]

[0149] Where i represents an autonomous vehicle, and n represents the total number of autonomous vehicles to be dispatched, i = 1, 2, ..., n;

[0150] j represents a parking area, where j = 0 represents a fixed parking area, j = 1, 2, ..., k represent temporary parking areas, k is the total number of temporary parking areas, and k ≥ n;

[0151] r j This represents the response value of the evaluated parking area j, reflecting the value of its geographical location;

[0152] cij This represents the total cost of dispatching vehicle i to parking area j, which may include driving costs and parking costs (i.e., parking fees);

[0153] x ij is a decision variable, taking the value 0 or 1, indicating whether vehicle i is assigned to parking area j.

[0154] For example, if parking area j is close to a high-demand area, then its r j The parking area is relatively high, and the parking area j is far from the area with high demand for vehicles, then its r j Lower.

[0155] In this embodiment of the application, the above c ij It can be calculated using the following formula:

[0156] c ij =α·distance(i,j)+p j

[0157] Where α represents the cost per unit distance traveled, expressed in yuan per kilometer, and this value can cover energy, vehicle wear and tear, etc.; p j The parking cost for parking area j can be obtained from the charging standard, such as the municipal parking fee standard or real-time dynamic pricing data. For example, for parking areas that charge by the hour, the parking cost per unit time can be obtained.

[0158] For example, if the distance from vehicle i to parking area j is 10 kilometers, and assuming α = 0.5 yuan / km, p j =10 yuan, then c ij =0.5×10+10=15 yuan.

[0159] In this embodiment of the application, the above-mentioned x ij The constraints may include, but are not limited to, the following:

[0160] Condition 1: Each vehicle can only park in one location; that is, a vehicle cannot be assigned to multiple parking spaces. This can be expressed by the following formula:

[0161]

[0162] Condition 2, parking area capacity limit, C j The maximum capacity of parking area j can be expressed by the following formula:

[0163]

[0164] Where i represents an autonomous vehicle; n represents the total number of autonomous vehicles to be dispatched; j represents a parking area; k is the total number of temporary parking areas; x ijis a decision variable, taking the value 0 or 1, indicating whether vehicle i is assigned to parking area j.

[0165] It should be noted that the parking area in this embodiment can represent a parking space, such as a single standard roadside parking space, in which case C j =1; it can also represent a road segment where vehicles can park, in which case C j = The total number of parking spaces available for this section of road; it can also represent a parking lot that can accommodate multiple vehicles, in which case C j = The total number of parking spaces available in the parking lot.

[0166] From the above equation, we can see that (r) j -c ij ) represents the net benefit of vehicle i, characterizing the contribution value of a single dispatch, i.e., vehicle i to parking area j. If r j >c ij If r is positive, it indicates that the scheduling generates a positive benefit; otherwise, r is negative. j <c ij If the result is negative, it indicates that the scheduling resulted in a loss. Therefore, the objective function for this vehicle scheduling optimization model is set as follows:

[0167]

[0168] Where Maximize represents the maximum value; i represents autonomous vehicles, n represents the total number of autonomous vehicles to be dispatched; j represents parking areas, where j = 0 represents fixed parking areas, j = 1, 2, ..., k represent temporary parking areas, k is the total number of temporary parking areas, and k ≥ n; r j This represents the response value of the evaluated parking area j, reflecting its geographical location value; c ij x represents the total cost of dispatching vehicle i to parking area j, which may include driving costs and parking costs (i.e., parking fees); ij is a decision variable, taking the value 0 or 1, indicating whether vehicle i is assigned to parking area j.

[0169] In this way, by traversing the matching combinations of each autonomous vehicle i and each parking area j, the total net profit of all possible allocation schemes can be obtained. By selecting the matching combination with high profit and low cost, the overall net profit can be maximized. Accordingly, the matching combination that maximizes the value of the above objective function is the final target matching combination, thus obtaining the target parking area corresponding to each autonomous vehicle.

[0170] In specific implementation, when performing step 220, refer to... Figure 7 As shown, this can be achieved by performing the following steps:

[0171] Step 2201: Perform the following operations for each autonomous vehicle: For any of the above autonomous vehicles, determine the driving cost of the autonomous vehicle from its current location to the fixed parking area and the above temporary parking areas, and the parking cost of the autonomous vehicle parking in the above temporary parking areas, wherein the parking cost of the fixed parking area is a preset value.

[0172] In this embodiment, the fixed parking area is usually a paid parking area, therefore, the preset value is 0.

[0173] In specific implementation, when executing step 2201, the driving cost of the autonomous vehicle from its current location to the fixed parking area and the driving cost from its current location to each temporary parking area are determined based on the current location and the unit distance driving cost of the autonomous vehicle. Based on the unit duration parking cost of each temporary parking area, the parking cost of the autonomous vehicle in the fixed parking area and each temporary parking area is determined. The parking duration of the autonomous vehicle can be determined by estimation.

[0174] Step 2202: Based on the above response values, driving costs and parking costs, use a preset vehicle scheduling optimization model to optimize and evaluate the matching combination of the above autonomous vehicles with the fixed parking area and the above temporary parking areas, and obtain the target parking area corresponding to the above autonomous vehicles.

[0175] In this embodiment of the application, when performing step 2202, based on the response values ​​obtained from the evaluation, the driving costs and parking costs calculated in step 2201, and the parking costs of parking in fixed parking areas (i.e., the aforementioned preset values), the vehicle scheduling optimization model is used to traverse the matching combinations of each autonomous vehicle to be scheduled with fixed parking areas and each temporary parking area, thereby obtaining each target matching combination that maximizes the value of the aforementioned objective function, i.e., n matching combinations, where each autonomous vehicle is matched with a parking area; then, the parking area in each target matching combination is the target parking area for the corresponding autonomous vehicle, which can be a fixed parking area or any temporary parking area.

[0176] For example, suppose a Robotaxi fleet has two autonomous vehicles that meet the aforementioned operating conditions. They can choose to park in a fixed parking area, such as a Robotaxi parking lot (j=0), or a temporary parking area, such as a roadside parking space (j=1). The relevant parameters are shown in the table below:

[0177] Table 2 Parameter Overview

[0178]

[0179] Based on the table above, the net income (r) is calculated using the aforementioned formula. j -c ij The following are the details:

[0180] For vehicle 1:

[0181] 1) Park in the designated parking area:

[0182] c 10 =0.5×5+0=2.5 ​​yuan

[0183] r0-c 10 =30 - 2.5 = 27.5 yuan

[0184] 2) Park in a roadside parking area:

[0185] c11 = 0.5 × 2 + 15 = 16 yuan

[0186] r1-c11=50-16=34 yuan

[0187] For vehicle 2:

[0188] 1) Park in the designated parking area:

[0189] c 20 =0.5×8+0=4 yuan

[0190] r0-c 20 =30-4=26 yuan

[0191] 2) Park in a roadside parking area:

[0192] c 21 =0.5×4+15=17 yuan

[0193] r1-c 21 =50-17=33 yuan

[0194] Therefore, the optimal allocation is: Vehicle 1 → roadside parking area (net profit of 34 yuan), Vehicle 2 → roadside parking area (net profit of 33 yuan), total profit of 67 yuan.

[0195] If the capacity of the roadside parking area is unlimited, then the total benefit of both vehicle 1 and vehicle 2 being able to park in the roadside parking area is higher, therefore, this strategy is better.

[0196] If the capacity of the roadside parking area is limited (e.g., C1 = 1): Vehicle 1 → roadside parking area (net profit of 34 yuan), Vehicle 2 → fixed parking area (net profit of 26 yuan), total profit of 60 yuan; compared with Vehicle 1 and Vehicle 2 both returning to the fixed parking area, the total profit is 27.5 + 26 = 53.5 yuan, the above strategy is still better.

[0197] Although the fixed parking area is far from the target areas with high projected demand in the future time period, its r0 is lower than that of other temporary parking areas. j (j>0), but the total net benefit may be higher due to the lower cost of parking in fixed parking areas. Therefore, in this embodiment, by including fixed parking areas and temporary parking areas (roadside parking spaces) into the parking area set and reasonably setting the above-mentioned vehicle scheduling optimization model and parameters, the vehicle optimization scheduling model, such as the Integer Linear Programming (ILP) model, can automatically compare the strategies of centralized return to Robotaxi parking lots and dispersed roadside parking spaces, and select the scheme with the optimal total net benefit. In this way, by covering the cost trade-off between the two strategies in the optimization evaluation, this method can improve vehicle response efficiency, thereby ensuring the maximization of operational benefits.

[0198] The following specific embodiment illustrates the detailed process of a parking scheduling method for an autonomous vehicle according to an embodiment of this application.

[0199] Assuming we are currently in a period of low nighttime demand, meaning the current demand for vehicles is less than a preset demand threshold, thus meeting the conditions for triggering a parking dispatch, and planning to dispatch autonomous vehicles in operation for nighttime parking, then, refer to... Figure 8 As shown, the specific process is as follows:

[0200] Step 801: Obtain all autonomous vehicles that meet the operational requirements;

[0201] Step 802: Obtain the top-m of the predicted vehicle demand values ​​for the next time period in the preset regions;

[0202] Step 803: Obtain the temporary parking areas for each of the above target areas;

[0203] Step 804: Evaluate the response values ​​of vehicles parked in each temporary parking area and the response values ​​of vehicles parked in fixed temporary parking areas;

[0204] Step 805: Determine whether the response value of each temporary parking area is greater than the response value of the fixed parking area; if yes, proceed to step 806; otherwise, proceed to step 808.

[0205] Step 806: Based on the response values ​​of each temporary parking area that are greater than the response value of the fixed parking area, and the current location of each autonomous vehicle to be dispatched, evaluate the matching combination of each autonomous vehicle with the fixed parking area and the temporary parking area, and determine the target parking area corresponding to each autonomous vehicle.

[0206] Step 807: Dispatch each autonomous vehicle to its corresponding target parking area;

[0207] Step 808: Dispatch each autonomous vehicle to a designated parking area.

[0208] For a detailed description of step 802, please refer to the descriptions of steps 302 to 303 above; for a detailed description of step 803, please refer to the descriptions of steps 2001 to 2002 above; for a detailed description of step 804, please refer to the description of step 210 above; for a detailed description of step 806, please refer to the description of step 220 above, and will not be repeated here.

[0209] In this way, through the above process, autonomous vehicles that meet the operational conditions for the next day can be parked nearby in target areas with high predicted demand for vehicles the following day. This ensures that the vehicles can be quickly put into service the next day, reducing the time and driving costs of vehicle return (to fixed parking areas such as Robotaxi parking lots) and dispatching the next day, thereby improving vehicle response efficiency and overall operational efficiency.

[0210] In this embodiment, when low demand for vehicles or a large number of autonomous vehicles in an empty state are detected, the aforementioned parking scheduling is initiated. The demand for vehicles in each area over a future time period is predicted, thereby identifying target areas with high predicted demand. Then, temporary parking areas within these target areas are acquired, and the response values ​​of fixed parking areas and each temporary parking area are quantified, representing the value of different parking areas. Based on these response values ​​and the current locations of autonomous vehicles meeting operational conditions, the matching combinations of each autonomous vehicle with the fixed and temporary parking areas are evaluated to determine the target parking area for each autonomous vehicle. The autonomous vehicles are then scheduled to their corresponding target parking areas. This allows some or all autonomous vehicles meeting operational conditions to be parked locally or pre-scheduled to target areas with high predicted demand over a future time period, thereby improving vehicle response efficiency, reducing vehicle scheduling time and costs, and enhancing overall operational efficiency and user satisfaction.

[0211] Based on the same inventive concept, see [reference] Figure 9 As shown in the figure, this application provides a parking scheduling device for autonomous vehicles, including:

[0212] The acquisition module 910 is used to acquire the temporary parking areas of each target area in the preset area with the top m predicted values ​​of vehicle demand in the future time period. Here, a single temporary parking is used to park autonomous vehicles that are planned to serve the vehicle demand in the corresponding target area in the future time period, and m is a positive integer not less than 1.

[0213] The response prediction module 920 is used to evaluate the response value of vehicles parking in different parking areas based on the predicted vehicle demand value corresponding to each target area and the road network distance from the fixed parking area and each temporary parking area to each target area.

[0214] The parking scheduling module 930 is used to evaluate the matching combination of each autonomous vehicle with the fixed parking area and the temporary parking area based on each response value and the current location of each autonomous vehicle to be scheduled, determine the target parking area corresponding to each autonomous vehicle, and schedule each autonomous vehicle to the corresponding target parking area.

[0215] In one possible implementation, before obtaining the temporary parking areas of each target area up to the top m of the predicted vehicle demand values ​​for a future time period in the preset region, the acquisition module 910 is further configured to:

[0216] Determine that the parking dispatch triggering conditions are met, wherein the parking dispatch triggering conditions include:

[0217] The current demand for vehicles is less than the preset demand threshold; and / or,

[0218] The ratio of the number of autonomous vehicles in an empty state to the total number of vehicles is greater than a preset ratio.

[0219] In one possible implementation, after determining that the parking dispatch triggering condition is met, and before obtaining the temporary parking areas of each target area among the top m predicted vehicle demand values ​​for a future time period in the preset area, the acquisition module 910 is further configured to:

[0220] The autonomous vehicles that meet the operating conditions are obtained, wherein the operating conditions include the vehicle's battery level being greater than a preset battery level, the vehicle's cleanliness being greater than a preset cleanliness level, and the vehicle having no abnormal conditions.

[0221] The vehicle demand in each region of the preset area is predicted during the future time period to obtain the predicted vehicle demand value for each region.

[0222] Based on the predicted vehicle demand values ​​for each region, the regions with the highest predicted vehicle demand values ​​are identified as the target regions.

[0223] In one possible implementation, the acquisition module 910 is specifically used for:

[0224] Perform the following operations for each target region:

[0225] For any target area among the target areas, a temporary parking area within that target area is obtained based on road network information;

[0226] If the number of parking spaces in the obtained temporary parking area is less than the predicted demand for vehicles in any target area, then based on the road network information, a temporary parking area within a preset range outside any target area is obtained to obtain the number of parking spaces in any target area that is the predicted demand for vehicles.

[0227] In one possible implementation, the response prediction module 920 is specifically used for:

[0228] Based on road network information, a first road network distance from the fixed parking area to each target area is determined, and a second road network distance from each temporary parking area to each target area is determined based on the road network information.

[0229] Based on the attenuation coefficient, the predicted vehicle demand for each target area, and their respective first road network distances, the response value of a vehicle parking in the fixed parking area is determined; and based on the attenuation coefficient, the predicted vehicle demand for each target area, and their respective second road network distances, the response value of a vehicle parking in each temporary parking area is determined.

[0230] In one possible implementation, before evaluating the matching combinations of each autonomous vehicle with the fixed parking area and the temporary parking area based on each response value and the current position of each autonomous vehicle to be scheduled, and determining the target parking area corresponding to each autonomous vehicle, the response prediction module 920 is further configured to:

[0231] The response values ​​of each temporary parking area are determined to be greater than the response values ​​of the fixed parking area.

[0232] In one possible implementation, the parking scheduling module 930 is specifically used for:

[0233] For each autonomous vehicle, the following operations are performed: For any one of the autonomous vehicles, the driving cost of the autonomous vehicle from the current location to the fixed parking area and each of the temporary parking areas is determined, as well as the parking cost of each autonomous vehicle parking in each of the temporary parking areas, wherein the parking cost of the fixed parking area is a preset value.

[0234] Based on the response values, driving costs, and parking costs, a preset vehicle scheduling optimization model is used to optimize and evaluate the matching combinations of each autonomous vehicle with the fixed parking area and the temporary parking area, thereby obtaining the target parking area corresponding to each autonomous vehicle.

[0235] In one possible implementation, the parking scheduling module 930 is specifically used for:

[0236] Based on the current location and unit distance driving cost of any autonomous vehicle, determine the driving cost of any autonomous vehicle from the current location to the fixed parking area, and the driving cost from the current location to each of the temporary parking areas.

[0237] Based on the unit-time parking cost of each temporary parking area, the parking cost of each autonomous vehicle parked in each temporary parking area is obtained.

[0238] Based on the same inventive concept, this application provides an electronic device, including:

[0239] Memory 1010 is used to store computer programs or instructions;

[0240] Processor 1020 is configured to execute computer programs or instructions in memory 1010 such that any of the methods described in the above embodiments is performed.

[0241] The processor 1020 may include one or more central processing units (CPUs) or digital processing units, etc., for executing computer programs or instructions in the memory 1010, such that any of the methods described in the above embodiments are executed.

[0242] It should be noted that the specific connection medium between the memory 1010 and the processor 1020 is not limited in the embodiments of this application. The embodiments of this application... Figure 10 In this diagram, the memory 1010 and processor 1020 are connected via bus 1030. The connections between other components are merely illustrative and not intended to be limiting. Bus 1030 can be divided into address bus, data bus, control bus, etc. For ease of illustration, Figure 10 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0243] Based on the same inventive concept, embodiments of this application provide a computer-readable storage medium that, when executed by a processor, enables the processor to perform any of the methods described in the above embodiments. Since the principle by which the computer-readable storage medium solves the problem is similar to the aforementioned parking scheduling method for an autonomous vehicle, the implementation of the computer-readable storage medium can be found in the implementation of the method; repeated details will not be elaborated further.

[0244] Based on the same inventive concept, this application also provides a computer program product, which includes computer program code. When the computer program code is run on a computer, it causes the computer to execute the method described in any of the embodiments discussed above. Since the principle by which the above-described computer program product solves the problem is similar to the aforementioned parking scheduling method for an autonomous vehicle, the implementation of the above-described computer program product can refer to the implementation of the method, and repeated details will not be described again.

[0245] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application 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.

[0246] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should 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, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0247] 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 that implement the functions specified in one or more processes in a flowchart and / or one or more blocks in a block diagram.

[0248] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more blocks in the block diagram.

[0249] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A parking scheduling method for autonomous vehicles, characterized in that, include: Get the temporary parking areas of each target area in the preset area with the top m predicted values ​​of vehicle demand in the future time period. Each temporary parking area is used to park autonomous vehicles that are planned to serve the vehicle demand in the corresponding target area in the future time period. m is a positive integer not less than 1. Based on the predicted vehicle demand values ​​corresponding to each target area, and the road network distances from fixed parking areas and each temporary parking area to each target area, the response values ​​of vehicles parked in different parking areas are evaluated. Based on each response value and the current location of each autonomous vehicle to be scheduled, the matching combination of each autonomous vehicle with the fixed parking area and the temporary parking area is evaluated to determine the target parking area corresponding to each autonomous vehicle, and the autonomous vehicles are scheduled to the corresponding target parking area.

2. The method as described in claim 1, characterized in that, Before obtaining the temporary parking areas of each target area up to the top m of the predicted vehicle demand values ​​for a future time period in the preset area, the method further includes: Determine that the parking dispatch triggering conditions are met, wherein the parking dispatch triggering conditions include: The current demand for vehicles is less than the preset demand threshold; and / or, The ratio of the number of autonomous vehicles in an empty state to the total number of vehicles is greater than a preset ratio.

3. The method as described in claim 2, characterized in that, After determining that the parking dispatch triggering conditions are met, and before obtaining the temporary parking areas of each target area from the top m of the predicted vehicle demand values ​​for the future time period in the preset area, the process further includes: The autonomous vehicles that meet the operating conditions are obtained, wherein the operating conditions include the vehicle's battery level being greater than a preset battery level, the vehicle's cleanliness being greater than a preset cleanliness level, and the vehicle having no abnormal conditions. The vehicle demand in each region of the preset area is predicted during the future time period to obtain the predicted vehicle demand value for each region. Based on the predicted vehicle demand values ​​for each region, the regions with the highest predicted vehicle demand values ​​are identified as the target regions.

4. The method as described in claim 1, characterized in that, The temporary parking areas for each target area that are the top m in terms of predicted vehicle demand for a future time period within a preset region include: Perform the following operations for each target region: For any target area among the target areas, a temporary parking area within that target area is obtained based on road network information; If the number of parking spaces in the obtained temporary parking area is less than the predicted demand for vehicles in any target area, then based on the road network information, a temporary parking area within a preset range outside any target area is obtained to obtain the number of parking spaces in any target area that is the predicted demand for vehicles.

5. The method according to any one of claims 1-4, characterized in that, The evaluation of vehicle parking response values ​​in different parking areas, based on the predicted vehicle demand values ​​corresponding to each target area and the road network distances from fixed parking areas and each temporary parking area to each target area, includes: Based on road network information, a first road network distance from the fixed parking area to each target area is determined, and a second road network distance from each temporary parking area to each target area is determined based on the road network information. Based on the attenuation coefficient, the predicted vehicle demand for each target area, and their respective first road network distances, the response value of a vehicle parking in the fixed parking area is determined; and based on the attenuation coefficient, the predicted vehicle demand for each target area, and their respective second road network distances, the response value of a vehicle parking in each temporary parking area is determined.

6. The method as described in claim 5, characterized in that, Before evaluating the matching combinations of each autonomous vehicle with the fixed parking area and the temporary parking area based on each response value and the current location of each autonomous vehicle to be scheduled, and determining the target parking area corresponding to each autonomous vehicle, the process further includes: The response values ​​of each temporary parking area are determined to be greater than the response values ​​of the fixed parking area.

7. The method as described in claim 6, characterized in that, Based on each response value and the current location of each autonomous vehicle to be scheduled, the matching combination of each autonomous vehicle with the fixed parking area and the temporary parking area is evaluated to determine the target parking area corresponding to each autonomous vehicle, including: For each autonomous vehicle, the following operations are performed: For any one of the autonomous vehicles, the driving cost of the autonomous vehicle from the current location to the fixed parking area and each of the temporary parking areas is determined, as well as the parking cost of each autonomous vehicle parking in each of the temporary parking areas, wherein the parking cost of the fixed parking area is a preset value. Based on the response values, driving costs, and parking costs, a preset vehicle scheduling optimization model is used to optimize and evaluate the matching combinations of each autonomous vehicle with the fixed parking area and the temporary parking area, thereby obtaining the target parking area corresponding to each autonomous vehicle.

8. The method as described in claim 7, characterized in that, The determination of the driving cost of any autonomous vehicle traveling from its current location to the fixed parking area and each of the temporary parking areas, and the parking cost of each autonomous vehicle parking in the fixed parking area and each of the temporary parking areas, respectively, includes: Based on the current location and unit distance driving cost of any autonomous vehicle, determine the driving cost of any autonomous vehicle from the current location to the fixed parking area, and the driving cost from the current location to each of the temporary parking areas. Based on the unit-time parking cost of each temporary parking area, the parking cost of each autonomous vehicle parked in each temporary parking area is obtained.

9. An electronic device, characterized in that, include: Memory is used to store computer programs or instructions; A processor for executing a computer program or instructions in the memory such that the method described in any one of claims 1-8 is performed.

10. A computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor, the processor is able to perform the method as described in any one of claims 1-8.

Images