A supermarket truck special parking space sharing method, device, equipment and medium

Abstract

The application relates to a supermarket truck special parking space sharing method, device, equipment and medium, belonging to the field of intelligent transportation and information technology, wherein the supermarket truck special parking space sharing method comprises the following steps: acquiring a truck driving track, constructing a truck arrival time prediction model, determining a truck arrival time based on the truck arrival time prediction model and the truck driving track; a dynamic parking space time period distribution model considering time fragmentation degree and time importance is constructed, and based on the dynamic parking space time period distribution model and the truck arrival time, parking spaces are distributed to social vehicles according to parking requests of the social vehicles, so that the parking demands of supermarket distribution trucks and the social vehicles are considered, and the utilization rate of parking space resources is improved.

Description

Technical Field

[0001] This invention relates to the fields of intelligent transportation and information technology, and in particular to a method, apparatus, equipment and medium for sharing dedicated parking spaces for supermarket trucks. Background Technology

[0002] Logistics and distribution are the foundation for chain supermarkets to achieve large-scale operations. However, delivery trucks often face parking difficulties, which leads to delayed delivery, increased delivery costs, and even traffic congestion. In order to alleviate the parking difficulties of supermarket logistics and delivery trucks, some cities have planned dedicated parking spaces for delivery trucks around supermarkets. However, the exclusive parking spaces for logistics and delivery trucks will exacerbate the problem of scarce urban parking resources.

[0003] Due to the unique characteristics of supermarket delivery in terms of parking time and frequency, the sharing of parking spaces for supermarket delivery trucks differs significantly from the sharing of private or fixed parking spaces in residential areas and office buildings in terms of sharing time periods. Traditional methods of sharing private parking spaces are difficult to apply to the sharing scenario of supermarket delivery truck parking spaces.

[0004] Considering parking space sharing to improve parking space utilization while addressing the parking difficulties of supermarket delivery trucks is of practical significance and research value. Summary of the Invention

[0005] In view of this, it is necessary to provide a method, device, equipment and medium for sharing dedicated parking spaces for supermarket delivery trucks, in order to solve the technical problem of using dedicated parking spaces for supermarket delivery trucks for parking space sharing.

[0006] To address the above problems, this invention provides a method for sharing dedicated parking spaces for supermarket trucks, comprising:

[0007] Obtain the truck's driving trajectory, construct a truck arrival time prediction model, and determine the truck's arrival time based on the truck arrival time prediction model and the truck's driving trajectory;

[0008] A dynamic parking space time-slot allocation model is constructed that considers the degree of time fragmentation and time importance. Based on the parking requests of social vehicles, parking spaces are allocated to social vehicles according to the dynamic parking space time-slot allocation model and the arrival time of the truck.

[0009] In one possible implementation, the truck arrival time prediction model includes a GRU neural network; determining the truck arrival time based on the truck arrival time prediction model and the truck's travel trajectory includes:

[0010] Obtain the truck's driving trajectory, and extract the truck's historical travel time data and date attributes based on the truck's driving trajectory;

[0011] The historical travel time data and date attributes are input into the GRU neural network to determine the truck's arrival time.

[0012] In one possible implementation, inputting the historical trip time data and date attributes into the GRU neural network to determine the truck arrival time includes:

[0013] A dataset is obtained based on the historical travel time data and date attributes;

[0014] The dataset was normalized using the Min-Max standardization formula to obtain the input vector;

[0015] The dataset is divided into training samples and test samples;

[0016] Initialize the hyperparameters and weights of the GRU neural network;

[0017] The input vector is input into the GRU neural network, and the parameters are tuned using a grid search algorithm to train the GRU neural network.

[0018] The mean squared error is used as the fitness function, and a genetic algorithm is used to iteratively optimize the weights of the attention layer until the fitness function reaches the optimal value. The calculation is then terminated to obtain the optimal weights of the attention layer.

[0019] The state value weight distribution and the state vector of the hidden layer are obtained by calculation.

[0020] After weighted summation of the state value weight distribution and the state vector, inverse normalization is performed to obtain the predicted value.

[0021] The test sample is input into a fully trained GRU neural network to obtain a single-step prediction value, wherein the single-step prediction value is the arrival time of the truck.

[0022] In one possible implementation, the input vector is calculated as follows:

[0023] ,

[0024] in, The input vector after normalization. For actual input data, The maximum value of the actual input data. This is the minimum value of the actual input data.

[0025] In one possible implementation, the allocation of parking spaces to social vehicles based on the dynamic parking time allocation model and the truck arrival time includes:

[0026] Set up shared parking space locks, obtain real-time parking space information based on the shared parking space locks, and determine available parking spaces for social vehicles based on the real-time parking space information;

[0027] The duration for which social vehicles can use shared parking spaces is determined based on the arrival time of the truck.

[0028] Based on the parking requests and parking space reservation process of social vehicles, the parking time period of social vehicles is obtained based on the duration of time that social vehicles can use shared parking spaces.

[0029] The system makes decisions on parking requests from private vehicles based on the parking time period, real-time parking space information, and dynamic parking space allocation model, and allocates parking spaces to private vehicles based on the available parking spaces available to them.

[0030] In one possible implementation, the decision-making process for parking requests from private vehicles based on the parking time period, real-time parking space information, and a dynamic parking space time allocation model includes:

[0031] Step 1: Construct a parking status matrix and a parking request status matrix for parking spaces. Based on the parking status matrix and the parking request status matrix, determine the comprehensive value of the parking space status, wherein the comprehensive value of the parking space status includes 0, 1, and 2.

[0032] Step 2: Traverse each parking space and obtain the comprehensive parking space status value. Judge the comprehensive parking space status value. When the comprehensive parking space status value is equal to 1, jump to step 3. When the comprehensive parking space status value is greater than 1, determine the remaining parking time of the parking space based on the parking time period, obtain the arrival time of the social vehicle, and compare the remaining parking time of the parking space with the arrival time of the social vehicle. When the remaining parking time of the parking space is less than the arrival time of the social vehicle, jump to step 3. When the remaining parking time of the parking space is greater than the arrival time of the social vehicle, add a parking space mark to the parking space and set a first mark value. Based on the first mark value, reject the parking request of the social vehicle.

[0033] Step 3: Add parking space markers to the parking spaces and set second marker values. Calculate the time fragmentation degree and time importance of the parking spaces. Based on the time fragmentation degree and time importance, determine the comprehensive time value of the parking spaces, and then proceed to Step 4.

[0034] Step 4: Traverse each parking space, search for the parking space marked with the second mark value, and compare the parking time comprehensive value of the parking space. When the parking time comprehensive value of the parking space is the smallest, accept the parking request of the social vehicle and use the parking space as the parking space of the social vehicle.

[0035] In one possible implementation, the degree of time fragmentation is calculated as follows:

[0036] ,

[0037] in, For the degree of time fragmentation, For the total fragmented time, The parking duration with the highest frequency in the parking space. For parking periods of private vehicles;

[0038] The formula for calculating the time importance is:

[0039] ,

[0040] in, For time importance, This refers to the number of truck delivery tasks performed during fragmented time periods. This represents the total number of truck delivery tasks completed in a single day.

[0041] On the other hand, the present invention also provides a supermarket truck-specific parking space sharing device, comprising:

[0042] The truck arrival time determination module is used to acquire the truck's driving trajectory, construct a truck arrival time prediction model, and determine the truck arrival time based on the truck arrival time prediction model and the truck's driving trajectory.

[0043] The shared parking space allocation module is used to construct a dynamic parking space time-slot allocation model that considers the degree of time fragmentation and time importance. Based on the parking requests of social vehicles, the module allocates parking spaces to social vehicles according to the dynamic parking space time-slot allocation model and the arrival time of the truck.

[0044] On the other hand, the present invention also provides an electronic device, including: a processor and a memory;

[0045] The memory stores a computer-readable program that can be executed by the processor;

[0046] When the processor executes the computer-readable program, it implements the steps in the supermarket truck-specific parking space sharing method described above.

[0047] On the other hand, the present invention also provides a computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the steps in the supermarket truck-specific parking space sharing method described above.

[0048] The beneficial effects of this invention are as follows: It utilizes a truck arrival time prediction model to achieve more accurate truck arrival time prediction; it constructs a dynamic parking space time-slot allocation model that considers the degree of time fragmentation and the importance of time; by considering the degree of time fragmentation, it maximizes the resources of shared parking spaces; by considering the importance of time, it reduces the impact of shared parking spaces on truck usage; based on the parking requests of social vehicles, it allocates parking spaces to social vehicles according to the dynamic parking space time-slot allocation model and truck arrival times; it intelligently analyzes parking demand, optimizes the allocation and management of parking spaces, effectively reduces truck waiting time, improves truck delivery efficiency, and takes into account the parking needs of supermarket delivery trucks and social vehicles, making the shared time slots for dedicated parking spaces for supermarket delivery trucks scientific and reasonable, improving parking space utilization while ensuring that truck parking needs are met. Attached Figure Description

[0049] Figure 1 A flowchart illustrating an embodiment of the supermarket truck-specific parking space sharing method provided by the present invention;

[0050] Figure 2 A schematic diagram illustrating the fragmented time of the supermarket truck-specific parking space sharing method provided by the present invention;

[0051] Figure 3 A schematic diagram of an embodiment of the supermarket truck-specific parking space sharing device provided by the present invention;

[0052] Figure 4 A schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation

[0053] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0054] This invention discloses a method, apparatus, equipment, and medium for sharing dedicated parking spaces for supermarket trucks, which can be used in a computer. The method, apparatus, or computer-readable storage medium involved in this invention can be integrated with the aforementioned equipment or can be relatively independent.

[0055] One specific embodiment of the present invention discloses a method for sharing dedicated parking spaces for supermarket trucks, which can be executed by a computer, specifically by one or more processors of the computer. Figure 1 This is a flowchart of the supermarket truck-specific parking space sharing method provided in an embodiment of the present invention. Please refer to [link / reference]. Figure 1 Methods for sharing dedicated parking spaces for supermarket delivery trucks include:

[0056] S101. Obtain the truck's driving trajectory, construct a truck arrival time prediction model, and determine the truck's arrival time based on the truck arrival time prediction model and the truck's driving trajectory.

[0057] S102. Construct a dynamic parking space time-slot allocation model that considers the degree of time fragmentation and time importance. Based on the parking requests of social vehicles, allocate parking spaces to social vehicles according to the dynamic parking space time-slot allocation model and the arrival time of trucks.

[0058] The truck arrival time prediction model is an EA-GRU neural network. The EA-GRU neural network predicts the arrival time of trucks, laying the foundation for the setting of subsequent parking space sharing time periods. It takes into account the degree of time fragmentation to avoid waste of parking space resources, and takes into account the importance of time to reduce the impact on trucks' use of parking spaces, so as to maximize the utilization of truck parking space time and space resources.

[0059] Compared with existing technologies, the supermarket truck-dedicated parking space sharing method provided in this embodiment obtains the truck's driving trajectory, constructs a truck arrival time prediction model, determines the truck's arrival time based on the truck arrival time prediction model and the truck's driving trajectory, and uses a GRU neural network to predict the truck's arrival time to obtain a more accurate truck arrival time. It also constructs a dynamic parking space time-slot allocation model that considers the degree of time fragmentation and time importance, and allocates parking spaces to social vehicles based on the dynamic parking space time-slot allocation model and the truck's arrival time. By intelligently analyzing parking demand, it optimizes the allocation and management of parking spaces, effectively reduces truck parking waiting time, improves truck delivery efficiency, and simultaneously takes into account the parking needs of supermarket delivery trucks and social vehicles, thereby improving the utilization rate of parking space resources.

[0060] In some embodiments, in step S101, the truck's driving trajectory is obtained, and historical travel time data and date attributes of the truck are extracted based on the truck's driving trajectory. The truck's positioning information is obtained through the Global Positioning System (GPS), and the optimal route for the truck to its destination is obtained in conjunction with navigation software. Historical travel time data and date attributes of the truck are extracted from the truck's driving trajectory on each branch road. Real-time and historical traffic flow information of the surrounding area is obtained, and a truck arrival time prediction model is constructed. The truck arrival time prediction model includes a GRU neural network. The truck arrival time is determined based on the truck arrival time prediction model and the truck's driving trajectory. The historical travel time data and date attributes are input into the GRU neural network to determine the truck arrival time.

[0061] First, the data is preprocessed. A dataset is obtained based on historical travel time data and date attributes. To eliminate the adverse effects caused by outlier data, the data is limited to... The dataset is normalized using the Min-Max standardization formula to obtain the input vector, which is calculated as follows:

[0062] ,

[0063] in, The input vector after normalization. For actual input data, The maximum value of the actual input data. The minimum value of the actual input data;

[0064] The historical travel time data features the travel times of the three days prior to the prediction date and the current day itself, which are most correlated with the prediction date. This includes 12 dimensions of travel time: t-1, t, and t+1 of the three days prior to the prediction date, and t-2, t-1, and t of the current day. The date attribute factors mainly include 4 dimensions of data: the date attributes of the three days prior to the prediction date and the prediction date itself. The date attributes are divided into two types: weekdays and non-weekdays. The date attribute value is 1 for weekdays and 0 for non-weekdays. Therefore, the truck arrival time prediction model has a total of 16 input variables.

[0065] Based on preprocessing, a GRU neural network is used for data processing, using input vectors. Predicted output vector ,make At this time, the GRU neural network is a single-step prediction model, meaning that the input window predicts only one future value each time. This prediction method has higher accuracy than multi-step prediction. Since the GRU neural network performs multi-step prediction, and the prediction error accumulates with the number of steps, resulting in increasingly inaccurate results, this is not considered. Therefore, a single-step prediction GRU neural network is used to predict the truck arrival time. The specific calculation process of the single-step prediction GRU neural network is as follows:

[0066] The dataset is divided into training and testing samples. After normalizing the input data, the input value at time t is obtained, and its calculation formula is:

[0067] ,

[0068] The hyperparameters and weights of the GRU neural network are initialized. The input vector is fed into the GRU neural network, and the hyperparameters are tuned using a grid search algorithm to train the GRU neural network. The mean squared error is used as the fitness function, and a genetic algorithm is used to iteratively optimize the weights of the attention layer until the fitness function reaches its optimal value. The calculation is terminated when the optimal weights of the attention layer are obtained. The state value weight distribution and the state vector of the hidden layer are calculated. After weighted summation of the state value weight distribution and the state vector, inverse normalization is performed to obtain the predicted value. The test sample is input into the fully trained GRU neural network to obtain the single-step predicted value, where the single-step predicted value is the arrival time of the truck.

[0069] In some embodiments, in step S102, since supermarket delivery trucks occupy parking spaces for a relatively short period, parking space usage time periods are shared to make full use of parking space resources. However, the shared time periods need to be designed to ensure the parking needs of delivery trucks. A dynamic parking space time period allocation model considering the degree of time fragmentation and time importance is constructed. Based on the dynamic parking space time period allocation model and the arrival time of the trucks, parking spaces are allocated to social vehicles. First, shared parking space locks are set up. Using smart parking space locks, parking spaces are reserved and opened, and parked vehicles are identified to prevent parking spaces from being occupied. Real-time parking space information is obtained based on the shared parking space locks, and available parking spaces for social vehicles are determined based on the real-time parking space information. The smart parking space locks realize the conversion between information and parking space locks and control through a host computer. The system features improved D-type smart parking locks installed on roadside parking spaces outside the supermarket. These locks are controlled by a server. When a vehicle is detected approaching a designated parking space, the lock's panoramic camera captures images of the surrounding environment, processes them, identifies the license plate number, and checks access permissions. If the vehicle has parking rights, an unlocking command is sent to the lock. Upon receiving the command, the lock arm lowers, allowing the vehicle to enter; otherwise, it remains locked and entry is prohibited. After a vehicle enters the space, the server records the time it entered and records entry information. When a vehicle leaves the space, an internal motor raises the lock arm, locking the space. The server records the departure time and confirms and records exit information.

[0070] Secondly, the duration for which private vehicles can use shared parking spaces is determined based on the arrival time of the truck. The time of submission of the parking request is obtained after the private vehicle submits the request. The duration for which private vehicles can use shared parking spaces when submitting a request will be determined based on the truck delivery schedule. The process is as follows:

[0071] When a truck has no delivery assignments, determine the number of feasible routes the truck can take from the warehouse to the delivery point. For each path The arrival time of the trucks was calculated using a truck arrival time prediction model. (Considering the distribution of large supermarket warehouses and stores in the city, most warehouses take more than 1.5 hours to reach delivery points; therefore, the delivery time for delivery trucks to reach delivery points is set to be greater than 1.5 hours.) The average arrival time of trucks for all feasible routes is taken as a reference for the duration of shared parking space usage by private vehicles. The formula for calculating the average arrival time of trucks is:

[0072] ,

[0073] in, This represents the average arrival time of the trucks. The number of feasible routes for trucks from the warehouse to the distribution point. This refers to the truck's arrival time;

[0074] When a truck is on a delivery mission, obtain the time period required for the truck to arrive in real time. Furthermore, parking spaces are reserved in advance based on the real-time arrival time and historical truck parking and unloading times. When a private vehicle submits a request, the duration for which it can use the shared parking space is equal to the truck's arrival time. The formula for calculating the duration for which a private vehicle can use the shared parking space, based on the truck's delivery schedule, is as follows:

[0075] ,

[0076] Based on the parking requests and parking space reservation process of private vehicles, the parking time period of private vehicles is obtained based on the available shared parking space duration. After a private vehicle issues a parking request according to the parking space reservation process, the available shared parking space duration is fed back to the private vehicle to determine the start time of parking. and the time of parking end The parking time period for private vehicles is determined based on the start and end times of parking, and the calculation formula is as follows:

[0077] ,

[0078] in, For parking periods of private vehicles;

[0079] The system makes decisions on parking requests from social vehicles based on parking time periods, real-time parking space information, and dynamic parking space allocation models. It also allocates parking spaces to social vehicles based on available parking spaces, aiming to maximize the utilization of time and space resources for truck parking spaces through parking space sharing. The utilization of time and space resources in parking lots includes two concepts: the degree of fragmentation of time generated by parking space allocation and the importance of fragmented time.

[0080] Due to the randomness of parking times, vehicle parking may divide a period of free time in a shared parking space into one or two short, discontinuous time segments. Time segments shorter than 40 minutes are defined as fragmented time. For a diagram illustrating fragmented time, please refer to [link / reference needed]. Figure 2 ,like Figure 2 As shown, when the reserved vehicle is parked, the parking start time of the reserved vehicle is obtained. and the time of parking end Get the starting time of parking when a vehicle parks in that parking space. and the time of parking end By reserving the start time of parking for the vehicle The time when the vehicle in that parking space ended. Calculations are performed to obtain the fragmented time. , Too much fragmented time can lead to a large number of idle periods going unused, resulting in a waste of parking space resources.

[0081] The length of idle time periods can be characterized by the degree of time fragmentation. The degree of time fragmentation can be obtained by collecting and analyzing parking data to obtain the distribution of vehicle parking time. The formula for calculating the degree of time fragmentation is:

[0082] ,

[0083] in, For the degree of time fragmentation, For the total fragmented time, The parking duration with the highest frequency in the parking space. The parking time period for private vehicles, as defined, is... Smaller parking spaces are more efficiently utilized.

[0084] Parking space resources are more valuable during truck delivery periods than when trucks are not on delivery. When allocating parking spaces, in addition to minimizing the fragmentation of time slots, these time slots should be allocated as far away from truck delivery periods as possible, meaning their temporal importance should be minimized to reduce their impact on truck parking space usage. Arrival data for truck delivery times is collected for each time slot within a 24-hour period. The temporal importance of fragmented time slots generated during the parking time of other vehicles is as follows:

[0085] ,

[0086] in, For time importance, This refers to the number of truck delivery tasks performed during fragmented time periods. This represents the total number of truck delivery tasks completed in a single day.

[0087] The parking lot sharing time is set with a 15-minute time interval as the smallest granularity of time measurement. The 24 hours are divided into 96 time intervals, and the time corresponding to time interval t is: to The unit is min; set the initial state matrix for shared parking spaces. This refers to the parking status matrix of parking spaces. When parking space a is vacant at time t, Conversely, it is 1, representing the parking request vector. That is, the request parking state matrix and the time importance vector. and parking space markings The process of making decisions on parking requests from social vehicles based on the dynamic parking space time allocation model is as follows:

[0088] Step 1: Construct the parking status matrix and the request parking status matrix of the parking space. Based on the parking status matrix and the request parking status matrix, determine the comprehensive value of the parking space status. The formula for calculating the comprehensive value of the parking space status is:

[0089] ,

[0090] in, This is a comprehensive value representing the parking space status. This is the parking status matrix value; a value of 0 indicates a parking space is vacant, and a value of 1 indicates a parking space is being used. The value requested is the parking status matrix value. A value of 0 indicates no parking request, and a value of 1 indicates a parking request. The overall parking space status value includes 0, 1, and 2. When the parking space is available and no other vehicles are requesting parking, it indicates that the parking space is vacant and no other vehicles are requesting parking. When the parking space is occupied and there is no request to park there, or when the parking space is vacant but there is a request to park there, it indicates that the parking space is occupied and there is no request to park there. When the parking space is occupied and a vehicle requests parking, it indicates that the parking space is in use and a vehicle has requested parking.

[0091] Step 2: Iterate through each parking space and obtain the overall parking space status value. The system assesses the overall parking space status. If the overall parking space status value is 1, it proceeds to step 3. If the overall parking space status value is greater than 1, the remaining parking time of the parking space is determined based on the parking time periods of other vehicles. To obtain the arrival time of other vehicles, calculate the remaining parking time for the parking space. Compared with the arrival time of other vehicles, the remaining parking time of the parking space If the arrival time is less than that of other vehicles, proceed to step 3, where the remaining parking time is... If the arrival time of other vehicles exceeds the arrival time of other vehicles, add a parking space marker to the parking space. Then, set the first flag value, that is... When the first flag value is equal to 0, the parking request of a social vehicle is rejected. The remaining parking time of the parking space is calculated as follows:

[0092] ,

[0093] in, This represents the remaining parking time for the parking space. This refers to the actual time during the parking space allocation process. The time when the vehicle leaves the parking space;

[0094] Step 3: For parking spaces that meet the conditions in Step 2, add a parking space marker and set a second marker value. Calculate the degree of time fragmentation of parking spaces. and the importance of time The comprehensive value of parking space time is determined based on the degree of time fragmentation and time importance. Then, proceed to step 4, where the formula for calculating the comprehensive parking time value is:

[0095] ,

[0096] in, This represents the overall parking time value.

[0097] Step 4: Iterate through each parking space and search for the parking space marker among all parking spaces. For the second tag value The parking spaces, and the combined parking time value for all parking spaces marked with a value of 1. Comparison, when the parking space time is comprehensive value When the minimum parking time is reached, the system accepts parking requests from other vehicles and selects the parking space with the lowest overall parking time value as the designated parking space, while modifying the parking space markings of other vehicles. The parking space marker has a value of 0 or 1. When it is 0, it means that the parking space is identified as a shared parking space. When it is 1, the parking space is not shared.

[0098] The process for private vehicles to apply for shared parking spaces is as follows: When a private vehicle issues a parking request, the maximum parking time for the private vehicle is determined, and the start and end time periods of parking for the private vehicle are obtained. The usage time of the private vehicle is determined, and a dynamic parking space time allocation model is used to decide whether to accept the specific parking time request issued by the car owner, and a parking space is allocated. After successfully reserving a parking space, when the vehicle is within five meters of the parking space, the parking space lock identifies and verifies the vehicle and determines whether to lock or unlock.

[0099] To regulate parking space usage, reservation rules and a reward / penalty mechanism were established. Big data analytics were used to analyze and mine data on parking lot usage near the supermarket, the behavior of other vehicle users, and parking fees. This determined the supermarket's delivery truck demand and the parking needs of other vehicles. A procedure for reserving parking spaces was designed, along with a penalty mechanism for vehicles failing to leave within the reserved time slot. Specifically, a credit score system was established to regulate the behavior of other vehicles. When a vehicle violates the rules or parks improperly, the user is penalized by lowering their credit score. When the credit score drops to a certain level, the vehicle will be unable to reserve a parking space.

[0100] The penalty mechanism for vehicles not leaving within the reserved time slot is as follows: When a vehicle parks beyond the time limit, its deposit will be deducted, and a text message reminder will be sent. The smart parking lock's visual recognition system will determine whether the vehicle has left. If the vehicle leaves, the overtime behavior will be recorded. If the vehicle does not leave, the fee will be increased to three times the usual price until the vehicle leaves, and the overtime behavior will be recorded. If the same vehicle is detected to have more than 5 overtime violations, it will be blacklisted, and the smart parking lock will not be open to that vehicle.

[0101] To better implement the supermarket truck parking space sharing method in this embodiment of the invention, based on the supermarket truck parking space sharing method, correspondingly, as follows: Figure 3 As shown, this embodiment of the invention also provides a supermarket truck-specific parking space sharing device. The supermarket truck-specific parking space sharing device 300 includes:

[0102] The truck arrival time determination module 301 is used to acquire the truck's driving trajectory, construct a truck arrival time prediction model, and determine the truck's arrival time based on the truck arrival time prediction model and the truck's driving trajectory.

[0103] The shared parking space allocation module 302 is used to construct a dynamic parking space time-slot allocation model that considers the degree of time fragmentation and time importance. Based on the parking requests of social vehicles, the module allocates parking spaces to social vehicles according to the dynamic parking space time-slot allocation model and the arrival time of the truck.

[0104] like Figure 4 As shown, based on the method for sharing dedicated parking spaces for supermarket trucks, the present invention also provides an electronic device 400, which can be a mobile terminal, desktop computer, laptop, handheld computer, server, or other computing device. The electronic device 400 includes a processor 401, a memory 402, and a display 403. Figure 4 Only some components of the electronic device 400 are shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead.

[0105] In some embodiments, memory 402 may be an internal storage unit of the electronic device 400, such as a hard disk or memory of the electronic device 400. In other embodiments, memory 402 may be an external storage device of the electronic device 400, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 400. Furthermore, memory 402 may include both internal and external storage units of the electronic device 400. Memory 402 is used to store application software and various types of data installed on the electronic device 400, such as program code installed on the electronic device 400. Memory 402 may also be used to temporarily store data that has been output or will be output. In one embodiment, memory 402 stores a supermarket truck parking space sharing program, which can be executed by processor 401 to implement the supermarket truck parking space sharing method of various embodiments of the present invention.

[0106] In some embodiments, processor 401 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 402 or process data, such as a method for sharing dedicated parking spaces for supermarket trucks.

[0107] In some embodiments, display 403 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 403 is used to display identification information for the supermarket truck-specific parking space sharing program and to display a visual user interface. Components 401-403 of electronic device 400 communicate with each other via a system bus.

[0108] In some embodiments, when the processor 401 executes the supermarket truck parking space sharing program in the memory 402, it implements each step of the supermarket truck parking space sharing method as described in the above embodiments. Since the supermarket truck parking space sharing method has been described in detail above, it will not be repeated here.

[0109] In summary, the supermarket truck-dedicated parking space sharing method, device, equipment, and medium provided by this invention acquires the truck's driving trajectory, constructs a truck arrival time prediction model, and determines the truck's arrival time based on the truck arrival time prediction model and the truck's driving trajectory; it also constructs a dynamic parking space time-slot allocation model that considers the degree of time fragmentation and time importance, and allocates parking spaces to social vehicles based on the parking requests of social vehicles, taking into account the parking needs of supermarket delivery trucks and social vehicles, thereby improving the utilization rate of parking space resources.

[0110] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0111] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for sharing dedicated parking spaces for supermarket trucks, characterized in that, include: Obtain the truck's driving trajectory, construct a truck arrival time prediction model, and determine the truck's arrival time based on the truck arrival time prediction model and the truck's driving trajectory; A dynamic parking space allocation model considering the degree of time fragmentation and time importance is constructed. Based on the parking requests of social vehicles, parking spaces are allocated to the social vehicles according to the dynamic parking space allocation model and the arrival time of the truck. This includes: making decisions on the parking requests of social vehicles based on parking time periods, real-time parking space information, and the dynamic parking space allocation model, specifically: Step 1: Construct a parking status matrix and a parking request status matrix for parking spaces. Based on the parking status matrix and the parking request status matrix, determine the comprehensive value of the parking space status, wherein the comprehensive value of the parking space status includes 0, 1, and 2. Step 2: Traverse each parking space and obtain the comprehensive parking space status value. Judge the comprehensive parking space status value. When the comprehensive parking space status value is equal to 1, jump to step 3. When the comprehensive parking space status value is greater than 1, determine the remaining parking time of the parking space based on the parking time period, obtain the arrival time of the social vehicle, and compare the remaining parking time of the parking space with the arrival time of the social vehicle. When the remaining parking time of the parking space is less than the arrival time of the social vehicle, jump to step 3. When the remaining parking time of the parking space is greater than the arrival time of the social vehicle, add a parking space mark to the parking space and set a first mark value. Based on the first mark value, reject the parking request of the social vehicle. Step 3: After adding a parking space marker and setting a second marker value, calculate the time fragmentation degree and time importance of the parking space. Based on the time fragmentation degree and time importance, determine the comprehensive time value of the parking space, and then proceed to Step 4. The formula for calculating the time fragmentation degree is: ， in, For the degree of time fragmentation, For the total fragmented time, The parking duration with the highest frequency in the parking space. For parking periods of private vehicles; The formula for calculating the time importance is: ， in, For time importance, This refers to the number of truck delivery tasks performed during fragmented time periods. The total number of truck delivery tasks in a day; Step 4: Traverse each parking space, search for the parking space marked with the second mark value, and compare the parking time comprehensive value of the parking space. When the parking time comprehensive value of the parking space is the smallest, accept the parking request of the social vehicle and use the parking space as the parking space of the social vehicle.

2. The method for sharing dedicated parking spaces for supermarket trucks according to claim 1, characterized in that, The truck arrival time prediction model includes a GRU neural network; the determination of the truck arrival time based on the truck arrival time prediction model and the truck's driving trajectory includes: Obtain the truck's driving trajectory, and extract the truck's historical travel time data and date attributes based on the truck's driving trajectory; The historical travel time data and date attributes are input into the GRU neural network to determine the truck's arrival time.

3. The method for sharing dedicated parking spaces for supermarket trucks according to claim 2, characterized in that, The step of inputting the historical trip time data and date attributes into the GRU neural network to determine the truck arrival time includes: A dataset is obtained based on the historical travel time data and date attributes; The dataset was normalized using the Min-Max standardization formula to obtain the input vector; The dataset is divided into training samples and test samples; Initialize the hyperparameters and weights of the GRU neural network; The input vector is input into the GRU neural network, and the parameters are tuned using a grid search algorithm to train the GRU neural network. The mean squared error is used as the fitness function, and a genetic algorithm is used to iteratively optimize the weights of the attention layer until the fitness function reaches the optimal value. The calculation is then terminated to obtain the optimal weights of the attention layer. The state value weight distribution and the state vector of the hidden layer are obtained by calculation. After weighted summation of the state value weight distribution and the state vector, inverse normalization is performed to obtain the predicted value. The test sample is input into a fully trained GRU neural network to obtain a single-step prediction value, wherein the single-step prediction value is the arrival time of the truck.

4. The method for sharing dedicated parking spaces for supermarket trucks according to claim 3, characterized in that, The formula for calculating the input vector is: ， in, The input vector after normalization. For actual input data, The maximum value of the actual input data. This is the minimum value of the actual input data.

5. The method for sharing dedicated parking spaces for supermarket trucks according to claim 3, characterized in that, The method of allocating parking spaces to social vehicles based on the dynamic parking space time allocation model and the truck arrival time also includes: Set up shared parking space locks, obtain real-time parking space information based on the shared parking space locks, and determine available parking spaces for social vehicles based on the real-time parking space information; The duration for which social vehicles can use shared parking spaces is determined based on the arrival time of the truck. Based on the parking requests and parking space reservation process of social vehicles, the parking time period of social vehicles is obtained based on the duration of time that social vehicles can use shared parking spaces. The system makes decisions on parking requests from private vehicles based on the parking time period, real-time parking space information, and dynamic parking space allocation model, and allocates parking spaces to private vehicles based on the available parking spaces available to them.

6. A shared parking space device for supermarket trucks, characterized in that, include: The truck arrival time determination module is used to acquire the truck's driving trajectory, construct a truck arrival time prediction model, and determine the truck arrival time based on the truck arrival time prediction model and the truck's driving trajectory. The shared parking space allocation module is used to construct a dynamic parking space time-slot allocation model that considers the degree of time fragmentation and time importance. Based on the parking requests of social vehicles, and using the dynamic parking space time-slot allocation model and the arrival time of the truck, it allocates parking spaces to the social vehicles. This includes: making decisions on the parking requests of social vehicles based on parking time slots, real-time parking space information, and the dynamic parking space time-slot allocation model, specifically: Step 1: Construct a parking status matrix and a parking request status matrix for parking spaces. Based on the parking status matrix and the parking request status matrix, determine the comprehensive value of the parking space status, wherein the comprehensive value of the parking space status includes 0, 1, and 2. Step 2: Traverse each parking space and obtain the comprehensive parking space status value. Judge the comprehensive parking space status value. When the comprehensive parking space status value is equal to 1, jump to step 3. When the comprehensive parking space status value is greater than 1, determine the remaining parking time of the parking space based on the parking time period, obtain the arrival time of the social vehicle, and compare the remaining parking time of the parking space with the arrival time of the social vehicle. When the remaining parking time of the parking space is less than the arrival time of the social vehicle, jump to step 3. When the remaining parking time of the parking space is greater than the arrival time of the social vehicle, add a parking space mark to the parking space and set a first mark value. Based on the first mark value, reject the parking request of the social vehicle. Step 3: After adding a parking space marker and setting a second marker value, calculate the time fragmentation degree and time importance of the parking space. Based on the time fragmentation degree and time importance, determine the comprehensive time value of the parking space, and then proceed to Step 4. The formula for calculating the time fragmentation degree is: ， in, For the degree of time fragmentation, For the total fragmented time, The parking duration with the highest frequency in the parking space. For parking periods of private vehicles; The formula for calculating the time importance is: ， in, For time importance, This refers to the number of truck delivery tasks performed during fragmented time periods. The total number of truck delivery tasks in a day; Step 4: Traverse each parking space, search for the parking space marked with the second mark value, and compare the parking time comprehensive value of the parking space. When the parking time comprehensive value of the parking space is the smallest, accept the parking request of the social vehicle and use the parking space as the parking space of the social vehicle.

7. An electronic device, characterized in that, Including memory and processor; The memory stores a computer-readable program that can be executed by the processor; When the processor executes the computer-readable program, it implements the steps of the supermarket truck-specific parking space sharing method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, which can be executed by one or more processors to implement the steps in the supermarket truck-specific parking space sharing method as described in any one of claims 1-5.

Images