Three-dimensional intelligent dispatching platform and method under multi-modal traffic integration

Abstract

The application provides a three-dimensional intelligent scheduling platform and method under multi-mode traffic integration, relates to the technical field of three-dimensional parking lots, and acquires a plurality of traffic tool quantities and a plurality of parking position distributions through a distributed monitoring module; a quantity optimization module configures a scheduling optimization quantity for each time of three-dimensional parking position scheduling optimization according to the plurality of traffic tool quantities; a position optimization module performs scheduling optimization on the plurality of parking position distributions according to the scheduling optimization quantity to obtain a plurality of optimal parking position distributions; and a three-dimensional intelligent scheduling module performs three-dimensional intelligent scheduling on the multi-mode traffic tools according to the plurality of optimal parking position distributions. The application solves the technical problem that the traditional three-dimensional parking lot scheduling method lacks comprehensive scheduling means for multi-mode traffic tools, leading to unreasonable allocation of parking resources, and achieves the technical effect of improving the scheduling efficiency and overall resource utilization rate of the three-dimensional parking lot.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of stereoscopic parking lot, and particularly relates to a stereoscopic intelligent scheduling platform and a scheduling method under multi-mode traffic integration. BACKGROUND

[0002] With the continuous development of urban traffic, stereoscopic parking lots have been widely used as an important way to solve the problem of urban parking. The stereoscopic parking lot effectively improves the utilization rate of parking space through multi-level and multi-floor design, and provides more parking space for the growing number of vehicles. However, most of the existing stereoscopic parking lot scheduling methods adopt a fixed scheduling mode, which is usually optimized based on the parking demand of a single type of traffic tool (such as a car). The scheduling mode is mainly adjusted according to the static number and distribution of parking spaces. In the face of mixed parking of multi-mode traffic tools (such as electric cars and cars), there is a lack of effective coordination and management capability, such as the inability to reasonably allocate electric car charging spaces and car parking spaces, resulting in poor scheduling accuracy, unreasonable resource allocation, and the inability to well adapt to diversified traffic demand. SUMMARY

[0003] The present application provides a stereoscopic intelligent scheduling platform and a scheduling method under multi-mode traffic integration, which solves the technical problem of the lack of comprehensive scheduling means for multi-mode traffic tools in the traditional stereoscopic parking lot scheduling method, resulting in unreasonable allocation of parking resources, and achieves the technical effect of improving the scheduling efficiency and overall resource utilization rate of the stereoscopic parking lot.

[0004] In view of the above problems, on the one hand, the present application provides a stereoscopic intelligent scheduling platform under multi-mode traffic integration, which comprises: a distribution monitoring module for monitoring and obtaining the number and position of multi-mode traffic tools in the stereoscopic parking lot, obtaining a plurality of traffic tool numbers and a plurality of parking position distributions, wherein the multi-mode traffic tools include electric cars and cars; a number optimization module for configuring the scheduling optimization number of each stereoscopic parking position scheduling optimization according to the plurality of traffic tool numbers; a position optimization module for scheduling optimization of the plurality of parking position distributions according to the scheduling optimization number, obtaining a plurality of optimal parking position distributions; and a stereoscopic intelligent scheduling module for stereoscopic intelligent scheduling of the multi-mode traffic tools according to the plurality of optimal parking position distributions.

[0005] In another aspect, the application also provides a three-dimensional intelligent scheduling method under multi-modal traffic integration, which comprises: in a three-dimensional parking lot, monitoring and obtaining the parking quantity and parking position of multi-modal vehicles, obtaining a plurality of vehicle quantities and a plurality of parking position distributions, wherein the multi-modal vehicles include electric cars and cars; according to the plurality of vehicle quantities, configuring a scheduling optimization quantity for each time of three-dimensional parking position scheduling optimization; according to the scheduling optimization quantity, scheduling optimization is performed on the plurality of parking position distributions to obtain a plurality of optimal parking position distributions; and according to the plurality of optimal parking position distributions, performing three-dimensional intelligent scheduling on the multi-modal vehicles.

[0006] The one or more technical solutions provided in the application have at least the following technical effects or advantages:

[0007] The distribution monitoring module is used to monitor and obtain the parking quantity and position distribution of multi-modal vehicles in the three-dimensional parking lot in real time, thereby providing accurate basic data for subsequent scheduling optimization. According to the data provided by the distribution monitoring module, the quantity optimization module analyzes the vehicle quantity, and reasonably determines the optimization scale according to the actual vehicle quantity, thereby providing a precise optimization quantity reference for the position optimization module and improving the rationality and efficiency of scheduling. The position optimization module optimizes a plurality of parking positions in the parking lot according to the result after quantity optimization, converts the optimization scale determined in the early stage into a specific position optimization result, obtains an optimal parking space distribution, ensures that the parking lot space is used most effectively, and avoids waste or unreasonable use of parking space. On the basis of position optimization, the three-dimensional intelligent scheduling module intelligently schedules the vehicles in the parking lot, adjusts the parking order and position of the vehicles, realizes effective scheduling of multi-modal vehicles in the three-dimensional parking lot, and improves the scheduling efficiency.

[0008] In summary, the application provides a data basis through the distribution monitoring module, determines the optimization scale through the quantity optimization module, obtains the optimal position distribution through the position optimization module, and performs scheduling operations through the three-dimensional intelligent scheduling module. This whole process realizes effective management of multi-modal vehicles in the three-dimensional parking lot, improves the utilization rate of parking lot resources, improves the scheduling efficiency, can adapt to the mixed parking of multi-modal vehicles including electric cars and cars, reduces the time for vehicles to find parking spaces, reasonably allocates resources in the parking lot, and thus improves the scheduling flexibility of the three-dimensional parking lot and the user parking experience as a whole.

[0009] The above description is only a summary of the technical solutions of the application. In order to more clearly understand the technical means of the application, the application can be implemented in accordance with the content of the specification, and in order to make the above and other purposes, features and advantages of the application more obvious and easy to understand, the following specific embodiments of the application are described. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 This is a schematic diagram of the structure of a three-dimensional intelligent dispatching platform under multi-modal transportation integration provided in the embodiments of this application.

[0011] Figure 2 A flowchart illustrating the three-dimensional intelligent scheduling method under multi-modal traffic fusion provided in this application embodiment.

[0012] Explanation of reference numerals in the attached diagram: Distribution monitoring module 10, Quantity optimization module 20, Location optimization module 30, Three-dimensional intelligent scheduling module 40. Detailed Implementation

[0013] This application provides a three-dimensional intelligent scheduling platform and scheduling method under multi-modal transportation integration, which solves the technical problem that traditional three-dimensional parking lot scheduling methods lack comprehensive scheduling means for multi-modal transportation, resulting in unreasonable allocation of parking resources. It achieves the technical effect of improving the scheduling efficiency and overall resource utilization of three-dimensional parking lots.

[0014] Example 1, as Figure 1 As shown in the embodiment of this application, a three-dimensional intelligent dispatching platform under multi-modal transportation integration is provided, the platform comprising:

[0015] The distribution monitoring module 10 is used to monitor and acquire the number and location of multi-mode vehicles in a multi-level parking garage, and to obtain the distribution of multiple vehicle numbers and multiple parking locations, wherein the multi-mode vehicles include trams and cars.

[0016] Specifically, multimodal transportation refers to different types of vehicles, including traditional cars and electric vehicles. The number of vehicles parked refers to the number of vehicles in the parking lot at a given time, and the parking location refers to the specific location or space where the vehicle is parked. The distribution monitoring module 10 collects information in real time through sensors or cameras installed in the parking lot. These devices transmit data to the central control system via a wireless network, updating the status of each parking space in the parking lot in real time (whether it is vacant, what type of vehicle is parked there, etc.). For example, a parking lot can be configured with ground sensors (such as pressure sensors) and cameras to monitor the arrival and departure of vehicles and display their distribution in the parking lot in real time. Based on this information, a real-time map of the parking lot can be generated, showing the number and specific location of each type of vehicle.

[0017] The distributed monitoring module 10 provides accurate real-time data support for subsequent scheduling and optimization steps, enabling the system to understand the parking status at every moment and ensuring the real-time nature of scheduling decisions.

[0018] The quantity optimization module 20 is configured to determine the number of scheduling optimization operations based on the number of vehicles.

[0019] Specifically, the number of scheduling optimization operations refers to the number of vehicles that need to be adjusted and optimized each time the scheduling is performed. The quantity optimization module 20 analyzes the number of each type of vehicle in the parking lot based on the data provided by the distribution monitoring module 10, and determines the number of scheduling optimization operations for each time the three-dimensional parking location scheduling optimization is performed, to ensure that the scheduling system can reasonably arrange the scheduling tasks according to the actual situation in the parking lot, thereby improving the scheduling efficiency and resource utilization. For example, if the number of trams increases significantly during a certain period, the quantity optimization module 20 may decide to increase the number of scheduling optimization operations dedicated to trams to ensure that there is enough space and resources to meet the parking demand.

[0020] The location optimization module 30 is configured to perform scheduling optimization on the plurality of parking location distributions based on the number of scheduling optimization operations to obtain a plurality of optimal parking location distributions.

[0021] Specifically, the optimal parking location distribution refers to the parking location that is most suitable for each type of vehicle through scheduling optimization. Based on the quantity optimization, the location optimization module 30 adjusts the distribution of parking spaces through an algorithm. For example, if it is found that the trams are parked densely on a certain floor, and the parking spaces for other vehicles on that floor are idle, more trams need to be scheduled to the idle floor, and the locations of other vehicles need to be optimized. The location optimization module 30 improves the use efficiency of parking space, avoids excessive congestion or idleness in the parking area, and thus provides a more scientific and reasonable parking solution for the parking lot.

[0022] The three-dimensional intelligent scheduling module 40 is configured to perform three-dimensional intelligent scheduling on the plurality of multi-mode vehicles based on the plurality of optimal parking location distributions.

[0023] Specifically, the three-dimensional intelligent scheduling module 40 performs three-dimensional intelligent scheduling on the multi-mode vehicles based on the optimal parking location distributions obtained by the location optimization module 30, and generates corresponding control instructions by communicating with the automatic equipment of the parking lot, such as elevators, conveyors, etc., to realize the automatic parking and taking out of vehicles.

[0024] Further, the distribution monitoring module 10 of the embodiment is further configured to perform the following steps:

[0025] Step P11: In the three-dimensional parking lot, the number of multi-mode vehicles is monitored and obtained to obtain a plurality of vehicle quantities.

[0026] Step P12: The parking location of each vehicle is monitored and obtained, and a plurality of parking location distributions are constructed based on the category of each vehicle.

[0027] Specifically, the distribution monitoring module 10 monitors and records the number of various vehicles in the parking lot through sensors, cameras and other devices. For example, pressure sensors can detect whether a parking space is occupied, and image recognition technology can identify each parked vehicle. Based on these data, the module can obtain real-time parking numbers and classify statistics of the number of different types of vehicles, such as trams and cars.

[0028] The distribution monitoring module 10 not only monitors the number of each vehicle, but also records the specific location of these vehicles in the parking lot. Through sensors installed on parking spaces (such as geomagnetic sensors) or through image recognition by cameras, real-time parking space information for each vehicle can be obtained, and these information can be processed in combination with the type of each vehicle to construct the parking location distribution of the parking lot. These parking location distribution information includes the specific parking location of each vehicle on each parking layer, parking area, providing accurate spatial distribution data for subsequent parking location optimization and intelligent scheduling, ensuring that the parking location of each vehicle is reasonably allocated. By combining the category of vehicles with the parking location, different types of vehicles can be optimized and scheduled according to demand, and reasonable scheduling decisions can be made.

[0029] Further, the quantity optimization module 20 of the embodiment of the present application is also used to perform the following steps:

[0030] Step P21: According to the number of the plurality of vehicles, the total number of vehicles is calculated and obtained.

[0031] Step P22: According to the total number of vehicles, the scheduling optimization quantity for each time of three-dimensional parking location scheduling optimization is configured.

[0032] Specifically, the quantity optimization module 20 aggregates the number of different types of vehicles according to the data provided by the distribution monitoring module 10, and calculates the total number of all vehicles in the parking lot. According to the total number of vehicles, considering the vehicle density, parking demand and distribution of vehicle categories in the parking lot, the number of vehicles that need to be optimized each time is determined, i.e. the scheduling optimization quantity. If the total number of vehicles is large, the scheduling optimization quantity can be increased to ensure that there is enough space and resources to meet the demand; if the total number of vehicles is small, the scheduling optimization quantity can be reduced to avoid waste of resources. The quantity optimization module 20 can flexibly configure the number of vehicles that need to be optimized each time according to the parking demand and the number of vehicles, so that the scheduling optimization is more in line with the actual demand, improves the rational utilization rate of parking spaces, and ensures the overall operation efficiency of the parking lot.

[0033] Further, step P22 includes:

[0034] Step P221: Obtain an average vehicle quantity of the stereoscopic parking lot.

[0035] Step P222: Obtain a preset scheduling optimization quantity.

[0036] Step P223: Calculate a ratio of the total vehicle quantity to the average vehicle quantity, multiply the preset scheduling optimization quantity, and take an integer to obtain a scheduling optimization quantity for each scheduling optimization of the stereoscopic parking position.

[0037] Specifically, the average vehicle quantity is a value reflecting the average level of the vehicle parking quantity of the stereoscopic parking lot in a certain period, which can be obtained by counting the vehicle quantities of the parking lot at different times and then taking an average. The quantity optimization module 20 extracts the vehicle quantity records of the parking lot in a historical period (such as every day, every week, or every month) from the database of the parking lot management system, takes an average of these historical data to calculate the average vehicle quantity. For example, the database stores the vehicle parking quantities of the past week as (50, 60, 45, 55, 65, 40, 50), and then the average vehicle quantity is obtained by adding these historical data and dividing by the number of days 7, which is about 52 vehicles.

[0038] The preset scheduling optimization quantity is a quantity value set in advance, which is a reference value suitable for scheduling optimization according to the design of the parking lot, operation experience, or theoretical calculation. The preset scheduling optimization quantity is read from the configuration file or parameter setting of the system. Then, the ratio of the total vehicle quantity to the average vehicle quantity is calculated, the ratio is multiplied by the preset scheduling optimization quantity, and finally the result is rounded to obtain the scheduling optimization quantity for each scheduling optimization of the stereoscopic parking position.

[0039] Through such a calculation method, a more reasonable scheduling optimization quantity can be obtained according to the current actual vehicle quantity of the parking lot, the normal situation, and the pre-set optimization scale, so that the scheduling optimization operation can better adapt to the actual operation demand of the parking lot.

[0040] Further, the position optimization module 30 of the embodiment is further used to perform the following steps:

[0041] Step P31: Fuse the plurality of parking position distributions to obtain a total parking position distribution.

[0042] Step P32: Randomly select the scheduling optimization quantity of vehicles for parking position random adjustment in the total parking position distribution according to the scheduling optimization quantity to obtain a first adjusted total parking position distribution.

[0043] Step P33: Analyze the first parking danger information of the first adjusted total parking position distribution.

[0044] Step P34: According to the first parking danger information, calculate the first scheduling fitness of the first adjusted total parking position distribution by combining with the promotion of parking uniformity.

[0045] Step P35: Continue to schedule optimization on the total parking position distribution according to the scheduling optimization number until convergence, output the total parking position distribution with the maximum scheduling fitness, and obtain the optimal total parking position distribution, wherein the optimal total parking position distribution includes a plurality of optimal parking position distributions of the multi-mode vehicles.

[0046] Specifically, the total parking position distribution is a comprehensive position distribution obtained by fusing a plurality of parking position distributions, which covers the parking position information of all vehicles in the parking lot. The position optimization module 30 collects a plurality of parking position distribution data constructed previously. These data can exist in the form of matrices, lists or database tables, and each element represents the parking position information of a vehicle. Then these data are fused. For example, if the parking position distribution is represented in the form of matrices, the module will perform addition operation or other logical merging operation on these matrices, and integrate all the position information into a new matrix, which represents the total parking position distribution. By fusing a plurality of parking position distributions, comprehensive parking position information in the parking lot can be obtained, providing complete basic data for subsequent scheduling optimization.

[0047] The position optimization module 30 obtains the scheduling optimization number, and randomly selects the vehicles of the scheduling optimization number in the total parking position distribution by using a random number generation algorithm. For example, if the total parking position distribution is a list, wherein each element represents the position information of a vehicle, the module randomly selects a specified number of indexes in the index range of the list by using a random number generation function, and the vehicles corresponding to these indexes are selected. For the selected vehicles, the module randomly adjusts the parking positions of these vehicles, and reconstructs a new parking position distribution, which is the first adjusted total parking position distribution. By randomly selecting and adjusting the parking positions of part of the vehicles, an initial adjustment scheme is provided for subsequent analysis and optimization, which helps to explore the possibility of different parking position combinations.

[0048] The first adjustment total parking position distribution is analyzed to obtain first parking danger information. The first parking danger information is information related to parking danger obtained by analyzing the first adjustment total parking position distribution. The parking danger analysis process includes judgment of multiple danger sources. For example, the module determines the distance relationship between the parking position of each vehicle and the dangerous area (such as the area around the fire-fighting facilities, emergency passage, etc.) according to the layout map of the parking lot. If the distance is too close, it indicates a higher danger. At the same time, whether the distance between vehicles meets the safety standard is also considered. If the electric vehicles are too close, it may cause the battery temperature to be too high, and if the vehicles are too close, it may increase the risk of collision. The module quantitatively or qualitatively evaluates these factors, and finally obtains the first parking danger information, which can be the probability of a dangerous event, such as the probability of battery overheating and the probability of collision. Accurate assessment of parking danger information helps to avoid safety hazards in subsequent scheduling optimization and improves the safety of the parking lot.

[0049] The parking distribution quality of the first adjustment total parking position distribution is evaluated in combination with the safety of parking (i.e., the first parking danger information) and the uniformity (e.g., whether the parking space distribution is reasonable), and a comprehensive fitness value, i.e., the first scheduling fitness, is calculated. The first scheduling fitness reflects the pros and cons of the current parking distribution. The higher the fitness, the more reasonable the distribution. The parking uniformity refers to whether the distribution of the parking positions of the vehicles in the parking lot is balanced, avoiding excessive concentration or excessive dispersion, which can be represented by the reciprocal of the parking density. In the calculation process, weights can be set for the first parking danger information and the parking uniformity, and the first scheduling fitness can be obtained by weighted average algorithm. By calculating the scheduling fitness, the pros and cons of the adjusted parking position distribution can be quantified, providing a basis for subsequent optimization decisions.

[0050] The total parking position distribution is further optimized multiple times according to the scheduling optimization quantity. Each operation obtains an adjusted total parking position distribution, and the scheduling fitness is calculated. As the number of operations increases, the values of the scheduling fitness are compared continuously. When the value of the scheduling fitness no longer increases significantly or reaches a certain convergence standard (e.g., the change in the scheduling fitness after continuous adjustment is less than a certain threshold), the total parking position distribution with the maximum scheduling fitness is output, which is the optimal total parking position distribution.

[0051] Through the continuous optimization process, an optimal parking distribution that is safe, uniform, and has the highest space utilization rate can be found, improving the space utilization rate, safety, and management efficiency of the entire parking lot.

[0052] Further, step P33 includes:

[0053] Step P331: Collect a sample total parking position distribution set from historical parking data of the same stereoscopic parking lot, and label the probability of each sample total parking position distribution appearing dangerous events to obtain a sample parking danger information set.

[0054] Step P332: Use the sample total parking position distribution set and the sample parking danger information set as training data to train a parking danger identification model.

[0055] Step P333: Input the first adjusted total parking position distribution into the parking danger identification model to obtain first parking danger information.

[0056] Specifically, in analyzing the first parking danger information, a parking danger identification model can be constructed using historical data and machine learning algorithms to achieve intelligent identification. This parking danger identification model learns from historical parking distribution and corresponding danger information, and can predict potential dangerous situations in the parking lot according to new parking position distribution data.

[0057] Historical parking data refers to the parking records of vehicles in a parking lot within a certain period of time. These data include the types of vehicles, parking time, parking location, etc. The sample total parking position distribution set is a collection of a series of total parking position distribution samples collected from the historical parking data of the same stereoscopic parking lot. Each sample represents the overall parking position situation of the parking lot in a certain period of time in the past. The sample parking danger information set corresponds to the sample total parking position distribution set, and is obtained by labeling the probability of each sample total parking position distribution appearing dangerous events. These information reflect the danger level under different parking position distribution. The probability of dangerous events refers to the possibility of unsafe events (such as traffic accidents, emergency passage being blocked, etc.) occurring under a certain parking position distribution. This probability is calculated through actual events in historical data.

[0058] A sample set of parking position distribution is constructed from historical parking data. For each sample distribution, whether a dangerous event has occurred is labeled according to historical records, and the probability of dangerous event occurrence is calculated for each distribution. Through these data, a sample parking danger information set containing different parking distributions and corresponding danger information is established, providing high-quality training data for subsequent training of danger identification model.

[0059] Each sample in the sample total parking position distribution set is taken as input data, and the information in the corresponding sample parking danger information set is taken as output label. A suitable machine learning model is selected, such as a neural network model (e.g., a multilayer perceptron) or a decision tree model, etc. The prepared data is used to train the model. During the training process, the model will adjust its parameters according to the input sample total parking position distribution and the corresponding output label, so as to minimize the error between the predicted result and the true label. For example, a simple three-layer neural network model is selected, the number of nodes in the input layer is determined according to the feature dimension of the sample total parking position distribution, the number of nodes in the hidden layer can be determined according to experience or experiment, and the number of nodes in the output layer is 1 (representing the probability of dangerous event). The sample data is input into the model for multiple iteration training, and the connection weight between neurons is adjusted through the back propagation algorithm until the model converges. Through the training of the parking danger identification model, the model can accurately predict the probability of dangerous events according to the input parking position distribution, and provide an effective risk assessment tool for the safety management of parking lots.

[0060] The first adjusted total parking position distribution is taken as input data and input into the trained parking danger identification model. The model analyzes and calculates the input parking position distribution according to the learned rules and parameters before, and finally outputs the first parking danger information, which is the evaluation result of the probability of dangerous events in the first adjusted total parking position distribution. Using the trained model to identify the danger information of the first adjusted total parking position distribution can quickly and accurately evaluate the safety of the current parking layout, and provide an important basis for subsequent scheduling optimization.

[0061] Further, step P34 includes:

[0062] Step P341: randomly selecting the first adjusted total parking position distribution according to the preset range anchor frame, calculating the parking density, outputting the maximum value, and obtaining the first maximum parking density.

[0063] Step P342: calculating the first scheduling fitness of the first adjusted total parking position distribution according to the first parking danger information and the first maximum parking density, as follows: wherein, F D is the scheduling fitness, w x is the danger weight, w m is the density weight, X is the parking danger information, and M is the maximum parking density.

[0064] Specifically, the preset range anchor frame is a pre-set range defining frame used to determine a specific selected area in the first adjusted total parking position distribution to facilitate the calculation of the parking density. The first maximum parking density is the highest density value calculated after the first adjusted total parking position distribution is randomly selected according to the preset range anchor frame, representing the maximum vehicle aggregation degree in the first adjusted total parking position distribution. The danger weight (w x ) and the density weight (w m ) are weight coefficients respectively given to the parking danger information and the maximum parking density in the calculation of the scheduling fitness formula, used to measure the relative importance of the two in the overall scheduling fitness.

[0065] The position optimization module 30 randomly selects in the first adjusted total parking position distribution according to the preset range anchor frame. This preset range anchor frame can be a rectangular area or a range defined by other shapes. For example, if the preset range anchor frame is a framework that divides the parking lot into a plurality of small rectangular areas, the module randomly selects some areas in these small rectangular areas for subsequent operations. For each selected area, the parking density is calculated. The parking density can be obtained by dividing the number of vehicles in the area by the effective parking area of the area. For example, if there are 5 vehicles in a small rectangular area, and the effective parking area of the area is 20 square meters, the parking density is 5 / 20 = 0.25 vehicles / square meter.

[0066] The obtained first parking danger information (X) and first maximum parking density (M), and the pre-set danger weight w x and the density weight w m are substituted into the formula to calculate the first scheduling fitness of the first adjusted total parking position distribution. This calculation formula comprehensively considers two important factors, the parking danger and the parking density, to quantitatively evaluate the pros and cons of the first adjusted total parking position distribution in scheduling, so as to make subsequent optimization decisions. For the new parking position distribution after each optimization, the corresponding scheduling fitness is calculated in a similar way. The distribution with a higher scheduling fitness indicates that the parking scheme of the parking lot is more reasonable and safer.

[0067] In summary, the three-dimensional intelligent scheduling platform under the multi-modal traffic fusion provided by the embodiments of the present application has the following technical effects:

[0068] This application embodiment utilizes multi-mode fusion scheduling and dynamic optimization to achieve a more rational allocation of parking spaces (including special resources such as trolley charging stations) within the parking lot, reducing resource idleness and waste, and improving the overall resource utilization rate of the multi-level parking garage. The phased optimization process and quantity-based dynamic scheduling optimization can quickly and accurately assign suitable parking positions to vehicles, reducing vehicle waiting time and driving time within the parking lot, thereby improving the overall scheduling efficiency of the multi-level parking garage. Overall, this application embodiment can simultaneously manage different types of transportation, such as trolleys and cars, adapting to diverse travel needs and providing a more convenient parking experience for users of different types of transportation.

[0069] Example 2, as Figure 2 As shown in the embodiments of this application, a three-dimensional intelligent scheduling method under multi-modal traffic fusion is provided, the method comprising:

[0070] Step S1: In the multi-level parking garage, monitor and obtain the number and location of multi-mode vehicles, and obtain the distribution of multiple vehicle numbers and multiple parking locations, wherein the multi-mode vehicles include trams and cars.

[0071] Step S2: Based on the number of vehicles, configure the number of scheduling optimizations to be performed each time for the three-dimensional parking location scheduling optimization.

[0072] Step S3: Based on the number of scheduling optimizations, optimize the multiple parking location distributions to obtain multiple optimal parking location distributions.

[0073] Step S4: Based on the distribution of the multiple optimal parking locations, perform three-dimensional intelligent scheduling of the multi-mode vehicles.

[0074] Furthermore, step S1 in the embodiments of this application includes:

[0075] Within a multi-level parking garage, the number of parking vehicles of various modes of transportation is monitored to obtain the total number of vehicles; the parking location of each vehicle is monitored and obtained, and combined with the category of each vehicle, the distribution of multiple parking locations is constructed.

[0076] Furthermore, step S2 in this embodiment includes:

[0077] Based on the number of the various vehicles, the total number of vehicles is calculated; based on the total number of vehicles, the number of scheduling optimizations for each time the multi-level parking location scheduling optimization is configured.

[0078] Furthermore, based on the total number of vehicles, the number of scheduling optimizations for each instance of multi-level parking location scheduling optimization is configured, including:

[0079] obtaining a preset scheduling optimization quantity; calculating a ratio of the total vehicle quantity and the average vehicle quantity, multiplying the preset scheduling optimization quantity, and taking an integer to obtain a scheduling optimization quantity of each time of scheduling optimization of the stereoscopic parking position.

[0080] Further, the step S3 of the embodiment of the application comprises:

[0081] fusing the plurality of parking position distributions to obtain a total parking position distribution; randomly selecting the scheduling optimization quantity of vehicles in the total parking position distribution to perform parking position random adjustment according to the scheduling optimization quantity, to obtain a first adjusted total parking position distribution; analyzing first parking danger information of the first adjusted total parking position distribution; calculating a first scheduling fitness of the first adjusted total parking position distribution according to the first parking danger information and in combination with promotion of parking uniformity; continuing to perform scheduling optimization on the total parking position distribution according to the scheduling optimization quantity until convergence, outputting a total parking position distribution with the largest scheduling fitness to obtain an optimal total parking position distribution, wherein the optimal total parking position distribution comprises a plurality of optimal parking position distributions of the multi-mode vehicles.

[0082] Further, analyzing the first parking danger information of the first adjusted total parking position distribution comprises:

[0083] According to historical parking data of the same stereoscopic parking lot, a sample total parking position distribution set is collected, and a probability of occurrence of a dangerous event for each sample total parking position distribution is labeled to obtain a sample parking danger information set; the sample total parking position distribution set and the sample parking danger information set are used as training data to train a parking danger identification model; the first adjusted total parking position distribution is input into the parking danger identification model to identify and output first parking danger information.

[0084] Further, calculating the first scheduling fitness of the first adjusted total parking position distribution according to the first parking danger information and in combination with promotion of parking uniformity comprises:

[0085] The first adjusted total parking position distribution is randomly selected according to a preset range anchor frame, a parking density is calculated, a maximum value is output, and a first maximum parking density is obtained; the first scheduling fitness of the first adjusted total parking position distribution is calculated according to the first parking danger information and the first maximum parking density, as follows: wherein F D is the scheduling fitness, w x is a danger weight, w m is a density weight, X is the parking danger information, and M is the maximum parking density.

[0086] The foregoing detailed description of the multi-modal traffic integration under the three-dimensional intelligent scheduling platform enables those skilled in the art to clearly understand the multi-modal traffic integration under the three-dimensional intelligent scheduling method in the embodiment. For the method disclosed in Embodiment Two, since it corresponds to the platform disclosed in Embodiment One, it has corresponding execution steps and beneficial effects. For the related parts, refer to the platform part for explanation.

[0087] The foregoing description of the disclosed embodiments enables a person skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A three-dimensional intelligent scheduling platform under the integration of multi-modal transportation, characterized in that, The platform comprises: a distribution monitoring module configured to monitor and obtain a number and a distribution of parking positions of multi-mode vehicles in a multi-level parking lot, wherein the multi-mode vehicles comprise trams and cars; a number optimization module configured to configure a scheduling optimization number for each time of multi-level parking position scheduling optimization according to the number of vehicles; a position optimization module configured to perform scheduling optimization on the distribution of parking positions according to the scheduling optimization number to obtain a plurality of optimal parking position distributions; a multi-level intelligent scheduling module configured to perform multi-level intelligent scheduling on the multi-mode vehicles according to the plurality of optimal parking position distributions; the position optimization module is further configured to perform the following steps: fuse the plurality of parking position distributions to obtain a total parking position distribution; select a number of vehicles in the total parking position distribution at random according to the scheduling optimization number to obtain a first adjusted total parking position distribution; analyze first parking danger information of the first adjusted total parking position distribution; calculate a first scheduling fitness of the first adjusted total parking position distribution according to the first parking danger information and in combination with parking uniformity improvement; continue to perform scheduling optimization on the total parking position distribution according to the scheduling optimization number until convergence, and output a total parking position distribution with the largest scheduling fitness to obtain an optimal total parking position distribution, wherein the optimal total parking position distribution comprises a plurality of optimal parking position distributions of the multi-mode vehicles; the step of analyzing the first parking danger information of the first adjusted total parking position distribution comprises: collect a sample total parking position distribution set according to historical parking data of the same multi-level parking lot, and label a probability of occurrence of a dangerous event for each sample total parking position distribution to obtain a sample parking danger information set; use the sample total parking position distribution set and the sample parking danger information set as training data to train a parking danger identification model; input the first adjusted total parking position distribution into the parking danger identification model to identify and output first parking danger information; the step of calculating the first scheduling fitness of the first adjusted total parking position distribution according to the first parking danger information and in combination with parking uniformity improvement comprises: select the first adjusted total parking position distribution at random according to a preset range anchor frame, calculate a parking density, output a maximum value, and obtain a first maximum parking density; According to the first parking danger information and the first maximum parking density, a first scheduling fitness of the first adjusted total parking position distribution is calculated as follows: ; wherein, is the scheduling fitness, is the danger weight, is the density weight, X is the parking danger information, and M is the maximum parking density. 2.The three-dimensional intelligent scheduling platform under multi-modal traffic integration of claim 1, wherein, the distribution monitoring module is further configured to perform the following steps: monitor and obtain a number of multi-mode vehicles in a multi-level parking lot to obtain a plurality of vehicle numbers; monitor and obtain a parking position of each vehicle, and construct a plurality of parking position distributions in combination with a category of each vehicle. 3.The three-dimensional intelligent scheduling platform under multi-modal traffic integration of claim 1, wherein, the number optimization module is further configured to perform the following steps: calculate a total number of vehicles according to the number of vehicles; configure a scheduling optimization number for each time of multi-level parking position scheduling optimization according to the total number of vehicles.

4. The three-dimensional intelligent scheduling platform under the integration of multi-modal transportation according to claim 3, characterized in that, According to the total vehicle quantity, a scheduling optimization quantity for each time of stereo parking position scheduling optimization is configured, and the execution step comprises: Obtaining an average vehicle quantity of the stereo parking lot; Obtaining a preset scheduling optimization quantity; Calculating a ratio of the total vehicle quantity to the average vehicle quantity, multiplying the preset scheduling optimization quantity, and taking an integer to obtain the scheduling optimization quantity for each time of stereo parking position scheduling optimization.

5. A three-dimensional intelligent scheduling method under multi-modal traffic integration, characterized in that, The method is executed by the stereo intelligent scheduling platform under the multi-mode traffic integration according to any one of claims 1-4, comprising: In the stereo parking lot, the parking quantity and the parking position of the multi-mode vehicles are monitored and obtained to obtain a plurality of vehicle quantities and a plurality of parking position distributions, wherein the multi-mode vehicles include electric cars and cars; According to the plurality of vehicle quantities, a scheduling optimization quantity for each time of stereo parking position scheduling optimization is configured; According to the scheduling optimization quantity, the plurality of parking position distributions are scheduled and optimized to obtain a plurality of optimal parking position distributions; According to the plurality of optimal parking position distributions, the multi-mode vehicles are intelligently scheduled in stereo.

Images