Intelligent parking guidance method for non-motorized vehicles

Abstract

The application discloses an intelligent parking guidance method for non-motor vehicles, and belongs to the field of intelligent transportation, and comprises the following steps: interacting with a parking destination of a non-motor vehicle; obtaining a destination parking lot set based on the parking destination, and selecting an optimal parking lot; interacting with a camera monitoring group of the optimal parking lot, and obtaining a parking distribution map of the optimal parking lot according to the camera monitoring group; selecting a recommended parking space set in the optimal parking lot according to the parking distribution map and the non-motor vehicle; planning a parking path according to the parking space set and the non-motor vehicle; and guiding the non-motor vehicle to park based on the parking path. The application solves the technical problems of low parking efficiency and high parking difficulty of non-motor vehicles in the prior art, and achieves the technical effects of improving the parking experience of non-motor vehicles and improving the parking efficiency of non-motor vehicles.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent transportation, in particular to an intelligent parking guidance method for non-motor vehicles. BACKGROUND

[0002] Non-motor vehicles play an important role in urban transportation as an environmentally friendly and convenient mode of transportation. However, existing non-motor vehicle navigation devices cannot obtain real-time parking lot vacancy information, providing accurate parking lot navigation services and vacancy recommendations for non-motor vehicle drivers, increasing the difficulty of non-motor vehicle drivers, and affecting their parking experience. SUMMARY

[0003] The present application provides an intelligent parking guidance method for non-motor vehicles to solve the technical problems of low parking efficiency and high parking difficulty in the prior art.

[0004] In view of the above problems, the present application provides an intelligent parking guidance method for non-motor vehicles.

[0005] The first aspect of the present application provides an intelligent parking guidance method for non-motor vehicles, the method comprising: interacting with the parking destination of the non-motor vehicle; obtaining a destination parking lot set based on the parking destination, and selecting an optimal parking lot; interacting with the camera monitoring group of the optimal parking lot, and obtaining a parking distribution map of the optimal parking lot according to the camera monitoring group; selecting a recommended parking space set in the optimal parking lot according to the parking distribution map and the non-motor vehicle; planning a parking path according to the parking space set and the non-motor vehicle; and guiding the non-motor vehicle to park based on the parking path.

[0006] Another aspect of the present application provides an intelligent parking guidance system for non-motor vehicles, the system comprising: a destination interaction module for interacting with the parking destination of the non-motor vehicle; an optimal parking lot module for obtaining a destination parking lot set based on the parking destination, and selecting an optimal parking lot; a parking distribution map module for interacting with the camera monitoring group of the optimal parking lot, and obtaining a parking distribution map of the optimal parking lot according to the camera monitoring group; a recommended parking space module for selecting a recommended parking space set in the optimal parking lot according to the parking distribution map and the non-motor vehicle; a parking path planning module for planning a parking path according to the parking space set and the non-motor vehicle; and a guidance parking module for guiding the non-motor vehicle to park based on the parking path.

[0007] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0008] The technical scheme comprises the following steps: collecting non-motor vehicle parking demand information and a destination; obtaining monitoring information of a parking lot of the destination, including real-time video streams and empty parking space detection results of each monitoring camera; analyzing the monitoring video streams to detect the empty parking space distribution of each monitoring area and construct a real-time empty parking space distribution map of the parking lot; recommending a set of empty parking spaces as a recommended parking space set according to the empty parking space distribution map and the non-motor vehicle information; planning a navigation path from the current position of the non-motor vehicle to the final recommended parking space as a parking path of the non-motor vehicle based on the recommended parking space set; and sending the recommended empty parking space and the planned parking path information to the navigation display interface of the non-motor vehicle to guide the non-motor vehicle driver to drive into the recommended empty parking space in real time and complete the parking into the garage. The technical scheme solves the technical problems of low parking efficiency and high parking difficulty of the non-motor vehicle in the prior art, and achieves the technical effects of improving the parking experience of the non-motor vehicle and improving the parking efficiency of the non-motor vehicle.

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

[0010] Figure 1 A possible flowchart of an intelligent parking guidance method for non-motor vehicles is provided for the embodiments of the present application.

[0011] Figure 2 A possible flowchart of obtaining a destination parking lot set in the intelligent parking guidance method for non-motor vehicles is provided for the embodiments of the present application.

[0012] Figure 3 A possible flowchart of selecting a recommended parking space set in the intelligent parking guidance method for non-motor vehicles is provided for the embodiments of the present application.

[0013] Figure 4 A possible structure diagram of an intelligent parking guidance system for non-motor vehicles is provided for the embodiments of the present application.

[0014] Explanation of reference signs: destination interaction module 11, optimal parking lot module 12, parking distribution map module 13, recommended parking space module 14, parking path planning module 15, and guidance parking module 16. DETAILED DESCRIPTION

[0015] The general idea of the technical scheme provided by the present application is as follows:

[0016] The embodiment of the present application provides a smart parking guidance method for non-motor vehicles. First, parking destination information of the non-motor vehicle is collected, monitoring information of a parking lot of the destination is acquired, including real-time video streams of each monitoring camera and empty parking space detection results; then computer vision technology is used to analyze the video streams, empty parking spaces in each monitoring area are detected, and a real-time empty parking space distribution map of the parking lot is constructed. On the basis of acquiring the real-time empty parking space information, a set of empty parking spaces is recommended as a recommended parking space set by combining demand information of the non-motor vehicle and using a computer intelligent algorithm; then, a navigation path from a current position of the non-motor vehicle to the empty parking space is planned as a parking path based on the recommended parking space set; finally, the recommended empty parking space and the planned parking path information are sent to a navigation display interface of the non-motor vehicle, real-time guidance is given to a driver of the non-motor vehicle to drive into the recommended empty parking space, and parking into the parking lot is completed.

[0017] After introducing the basic principle of the present application, various non-limiting embodiments of the present application will be specifically introduced in combination with the drawings of the specification. Embodiment

[0018] As shown in the figure, the embodiment of the present application provides a smart parking guidance method for non-motor vehicles, which comprises the following steps: Figure 1

[0019] Step S1000: interacting with a parking destination of the non-motor vehicle;

[0020] Specifically, the user inputs an address of the parking destination through voice or gesture instructions through a man-machine interface arranged inside the non-motor vehicle. After receiving the parking destination address input by the user, the address information is stored in a storage medium inside the non-motor vehicle as a basis for subsequent navigation and optimal parking lot selection.

[0021] Step S2000: acquiring a destination parking lot set based on the parking destination and selecting an optimal parking lot;

[0022] ​Specifically, first, a parking lot database is established, which stores detailed information of each parking lot, such as the name of the parking lot, the address of the parking lot, the number of parking spaces, the number of currently available parking spaces, parking fees, etc. Among them, the establishment of the parking lot database can be to obtain the data of each parking lot by field investigation, or to call the open interface of the existing parking lot information service platform to obtain the data. Then, the number of recommended parking lots is set, which represents the number of recommended parking lots calculated, such as setting the number to 3-5 according to actual needs. Next, the parking lot database is sorted according to the distance between the address of each parking lot and the address of the obtained parking destination, and a certain number of parking lots closest to the parking destination are selected, which constitute the destination parking lot set. Finally, the parking lot with the largest area, the most available parking spaces, the closest to the parking destination, and the best comprehensive conditions is selected from the obtained destination parking lot set, and is recommended to the non-motor vehicle user as the optimal parking lot.

[0023] Step S3000: interact the camera monitoring group of the optimal parking lot, and obtain the parking distribution map of the optimal parking lot according to the camera monitoring group;

[0024] Specifically, the parking lot is provided with a monitoring camera to monitor the use of parking spaces in the parking lot in real time. A communication connection is established with the monitoring system of the optimal parking lot, and real-time video images captured by each monitoring camera of the optimal parking lot are obtained through the communication connection. Then, the obtained real-time video images are processed, for example, image recognition technology is used to detect parking lines and vehicles in the video images, and the use state of each parking space in the optimal parking lot is calculated according to the geometric distribution characteristics of the parking lines and the detected vehicle positions, and a real-time parking distribution map of the parking lot is generated accordingly. In addition, if the monitoring system of the optimal parking lot provides a real-time parking space use state data interface, the interface can be directly called to obtain the use state of each parking space, and a parking distribution map is generated accordingly.

[0025] Step S4000: according to the parking distribution map and the non-motor vehicle in the optimal parking lot, a set of recommended parking spaces is selected;

[0026] Specifically, according to the acquired optimal parking lot parking distribution map, combined with the information of the non-motor vehicle, a certain number of recommended empty parking spaces are selected, which constitute a recommended parking space set. First, the position information of all idle parking spaces in the current optimal parking lot is acquired by scanning and analyzing the parking distribution map. Then, according to the information of the non-motor vehicle, such as the body size parameter, the empty parking spaces available for the current non-motor vehicle parking are filtered out. The filtering rule refers to whether the body size matches the size of the parking space to filter out the parking spaces that are not suitable for the current non-motor vehicle. Next, according to the recommendation strategy, a certain number of parking spaces are selected from the filtered empty parking spaces as recommended empty parking spaces. Among them, the recommendation strategy comprehensively considers multiple factors such as the distance between the empty parking space and the entrance of the optimal parking lot, the visibility of the empty parking space to the monitoring camera, and the last time the empty parking space was used, for example, the parking space with the shortest idle time, the shortest distance to the entrance, and the visible range of the monitoring camera is selected as the preferred recommended parking space. Then, the position information of the selected recommended empty parking spaces is extracted as the recommended parking space set.

[0027] Step S5000: planning a parking path according to the parking space set and the non-motor vehicle;

[0028] Specifically, the real-time position information of the non-motor vehicle is obtained, which comes from the GPS module or other positioning device set on the non-motor vehicle. Then, the layout information of the optimal parking lot is acquired, such as the connection relationship of each lane, the width of the lane, the position of the entrance and exit, etc. The information comes from the electronic map provided by the parking lot or is obtained by on-site investigation and is stored in the database. Next, the recommended parking spaces in the recommended parking space set are sorted, and a certain number of recommended parking spaces closest to the current position of the non-motor vehicle are selected, such as the first 2-3 closest parking spaces. These closest parking spaces will be used as the end point of the non-motor vehicle parking path planning. Then, a path planning algorithm such as A* algorithm is used to plan a path or multiple paths that can reach the closest recommended parking space according to the current position of the non-motor vehicle and the position of the closest recommended parking space on the road layout structure of the optimal parking lot. Finally, the path information of entering the parking lot and reaching the recommended parking space is integrated into a complete parking path, and the path information is sent to the non-motor vehicle to guide the non-motor vehicle to enter the parking lot and park in the recommended parking space.

[0029] Step S6000: guiding the non-motor vehicle to park based on the parking path.

[0030] Specifically, the planned parking path information is sent to the non-motor vehicle, and the non-motor vehicle displays the path information on a navigation display screen arranged in the non-motor vehicle to guide the driver to drive the non-motor vehicle into the parking lot. The parking path information is displayed in the form of graphical arrows or line segments, and is accompanied by voice navigation prompts, and the real-time position of the non-motor vehicle on the path is displayed at the same time, so as to facilitate the driver to understand and follow the path for driving, and to achieve the technical effects of improving the parking experience of the non-motor vehicle and improving the parking efficiency of the non-motor vehicle.

[0031] Further, as shown in Figure 2 The embodiment of the present application further includes:

[0032] Step S2100: establishing a parking lot database, the parking lot database including parking lot name, address, number of parking spaces, current number of remaining parking spaces, and cost;

[0033] Step S2200: setting a number of recommended parking lots;

[0034] Step S2300: traversing the addresses of the parking lots in the parking lot database, sorting the parking lots according to the addresses and the parking destination, and obtaining a parking space recommendation ranking table;

[0035] Step S2400: obtaining a destination parking lot set according to the number of recommended parking lots and the parking space recommendation ranking table.

[0036] Specifically, first-hand data is obtained by means of field research or the like, or relevant data is obtained by calling an open interface of a third-party parking lot information service platform to establish a parking lot database for storing detailed information of each parking lot. The database fields include parking lot name, parking lot address, number of parking spaces, current number of available parking spaces, parking cost, and the like. Then, a user sets a number of recommended parking lots according to actual needs, indicating the number of final recommended parking lots selected from the parking lot database. Next, the addresses of the parking lots in the parking lot database are traversed, a path planning algorithm is used to calculate the distances between the addresses of the parking lots and the address of the parking destination, and the parking lots are sorted according to the distances to select the parking lots closest to the address of the parking destination to form a parking space recommendation ranking table. Finally, the number of recommended parking lots set is selected from the obtained parking space recommendation ranking table to form a destination parking lot set.

[0037] Further, the embodiment of the present application further includes:

[0038] Step S2500: constructing a parking lot recommendation model;

[0039] Step S2600: traversing the parking space recommendation table to obtain first recommended parking space information;

[0040] Step S2700: obtaining first network recommendation information based on big data according to the first recommended parking space information;

[0041] Step S2800: inputting the first recommended parking space information and the first network recommendation information into the parking lot recommendation model to obtain a first parking lot recommendation index;

[0042] Step S2900: obtaining parking lot recommendation indexes of all parking spaces in the recommended parking space table to obtain a parking lot recommendation index set;

[0043] Step S21000: sorting the parking lot recommendation indexes to obtain an optimal parking lot.

[0044] Specifically, first, a parking lot information dataset containing basic information of each parking lot and user evaluation information of a corresponding network platform is obtained through big data, providing sample data for subsequent model training. Then, the model input features are determined, such as distance from the parking destination, parking fee, current number of available parking spaces, parking lot location score, order score, and scale score. Next, according to the parking lot information dataset and the input feature matrix, a model training sample matrix set is generated, each sample matrix corresponding to a parking lot and containing the input feature data of the parking lot. Then, the recommendation index of each parking lot is pre-assigned through artificial labeling to form the expected output of the sample matrix set for model training. Finally, a linear regression model is used to train the model with the feature data of the sample as input and the pre-assigned recommendation index as expected output. The trained model is the parking lot recommendation model.

[0045] Next, the information of each parking lot in the obtained preliminary recommended parking space table is input into the parking lot recommendation model in turn to obtain the corresponding parking lot recommendation index to form a parking lot recommendation index set. Finally, the parking lot recommendation index set is sorted, and the parking lot with the highest recommendation index is selected as the optimal parking lot to be recommended to the user.

[0046] Further, the embodiments of the application also include:

[0047] Step S2510: obtaining a parking lot dataset through a crawler technology, the parking lot dataset containing a corresponding network evaluation information set, wherein the parking lot data and the network evaluation information correspond one-to-one;

[0048] Step S2520: constructing a model input matrix, the model input matrix being a 2*3 matrix;

[0049] The first row of the 2*3 matrix is the distance, fee, and remaining number of parking spaces of the parking lot in turn; and the second row of the 2*3 matrix is the parking lot location score, parking lot order score, and parking lot scale score in turn.

[0050] Step S2530: According to the parking lot dataset, the network evaluation information set and the model input matrix, a sample input matrix set is obtained;

[0051] Step S2540: The sample input matrix set is matched with a recommendation index to obtain a recommendation index set;

[0052] Step S2550: A linear regression model is trained according to the sample input matrix set and the recommendation index set to obtain a parking recommendation model.

[0053] Specifically, first, the target website for obtaining data is determined, and the data source for obtaining parking lot information is located, such as some large parking lot information collection websites. Then, the website page layout is analyzed to find the structure of the parking lot information display. Subsequently, a targeted crawler code is written using a crawler library. The crawler code can log in to the target website, traverse the parking lot information display page, obtain the website address of each parking lot, enter the detail page of each parking lot, parse and extract basic information such as name, address, number of parking spaces, etc., enter the page of the corresponding parking lot on each evaluation platform, parse and extract evaluation information such as location score, order score, etc. Then, run the crawler to collect parking lot basic information and evaluation information, and save the collected parking lot basic information and corresponding evaluation information according to the parking lot ID.

[0054] Then, a 2x3 matrix is designed. The first row of the matrix respectively represents the distance (km) between the parking lot and the parking destination, the parking fee (yuan) and the current available parking space number, and the second row respectively represents the location score, the order score and the scale score of the parking lot on each evaluation platform, which are the main input features of the parking lot recommendation model. Subsequently, according to the obtained parking lot dataset and network evaluation information set, a sample matrix is generated for each parking lot, which contains the data of the 6 input features of the corresponding parking lot, i.e. the defined 2*3 model input matrix. All these sample matrices constitute the sample input matrix set. Then, the sample input matrix set is matched with the recommendation index of the corresponding parking lot one by one, and the recommendation index is added to the recommendation index set. The recommendation index can be evaluated by artificial evaluation or assigned by industry experts, and each sample matrix corresponds to a recommendation index. Finally, 80% of the sample matrix set and the recommendation index set are randomly selected as training data and 20% as test data; first, a linear regression model is initialized using the sklearn library, the 2*3 sample matrix in the training data set is set as the independent variable, and the corresponding recommendation index is set as the dependent variable, and the fit() function of the model is called for training. The test data set is used to check the model error, and the MSE is calculated as an evaluation index. According to the evaluation result, the model parameters are adjusted to reduce the error, and the finally trained linear regression model is used to obtain the recommendation index of any parking lot.

[0055] Further, the embodiment of the present application further comprises:

[0056] Step S2541: interacting with the parking demand of the non-motor vehicle;

[0057] Step S2542: setting position weight, order weight, and scale weight according to the parking demand;

[0058] Step S2543: traversing the generated sample input matrix set, obtaining a first parking lot position score, a first parking lot order score, and a first parking lot scale score;

[0059] Step S2544: obtaining a first recommendation index according to the first parking lot position score, the first parking lot order score, the first parking lot scale score, and the position weight, the order weight, and the scale weight;

[0060] Step S2545: adding the first recommendation index to the recommendation index set.

[0061] Specifically, on the navigation display screen in the non-motor vehicle, the following question is displayed for the user to select: Your parking demand mainly focuses on: A) short distance and short time; B) large parking lot scale and many cars; C) order and safety and comfort; the user only needs to select one of the three options A / B / C to obtain the user's parking demand. Then, according to the parking demand of the non-motor vehicle user, the position weight, the order weight, and the scale weight are set, for example, if the user's parking demand mainly focuses on the position factor, i.e., the distance to the destination is short, the position weight is set to 0.6, the order weight is set to 0.4, and the scale weight is set to 0.3.

[0062] Next, a sample matrix corresponding to a parking lot is taken out by traversing the generated sample input matrix set. Then, the position score, the order score, and the scale score of the parking lot on the network evaluation platform are obtained. According to the three scores of the obtained parking lot and the weights of the three factors, the comprehensive recommendation index of the first parking lot is calculated according to the formula: (position score * position weight) + (order score * order weight) + (scale score * scale weight), to obtain the comprehensive recommendation index of the parking lot. Then, the calculated recommendation index of the parking lot is added to the recommendation index set.

[0063] Further, as shown in Figure 3 the embodiment of the present application further comprises:

[0064] Step S4100: retrieving the parking database according to the non-motor vehicle type and the optimal parking lot to obtain a same-type vehicle parking set;

[0065] Step S4200: According to the parking frequency of different parking spaces in the same type of vehicle parking set, the parking spaces are sorted according to the parking frequency.

[0066] Step S4300: The sorting result is filtered according to the parking distribution map to obtain a recommended parking space set, and the recommended parking space set is sorted according to the parking frequency.

[0067] Specifically, vehicles are divided into different types, such as bicycles, electric bicycles, electric motorcycles, etc., and then a combined index is established according to the vehicle type and the parking lot ID. Non-motor vehicle types are obtained, and the ID of the optimal parking lot is obtained. According to the non-motor vehicle type and the optimal parking lot ID, the parking database is combined and indexed to retrieve all parking records that meet the conditions, forming a same type of vehicle parking set.

[0068] Then, according to the same type of vehicle parking set, the parking frequency of each parking space in the historical parking record is counted to determine which parking space has more historical parking of the same type of vehicle. Then, according to the parking frequency, each parking space is sorted from high to low. According to the real-time parking distribution map of the obtained optimal parking lot, the sorting result is filtered. Among them, the filtering rule is that if a parking space is currently occupied by a vehicle, it is excluded, and if the parking space is currently empty, it is retained. The final obtained parking space set is the recommended parking space set, and the sorting of each parking space is from high to low according to the historical parking frequency.

[0069] Further, the embodiments of the application also include:

[0070] Step S5100: Extracting the parking space according to the sorting of the recommended parking space set to obtain a first recommended parking space;

[0071] Step S5200: Establishing a real-time position of the non-motor vehicle and a first recommended path of the first recommended parking space;

[0072] Step S5300: Sending the first recommended path to the display screen of the non-motor vehicle to guide the non-motor vehicle to park;

[0073] Step S5400: Real-time interaction of the monitoring image of the first recommended parking space, and obtaining a second recommended parking space when the first recommended parking space has a vehicle.

[0074] Step S5500: Sending the first recommended path to the display screen of the non-motor vehicle to guide the non-motor vehicle to park.

[0075] Specifically, according to the ranking result of the obtained recommended parking space set, the top-ranked effective parking space is extracted, which is the first recommended parking space. Then, the current positioning information is obtained through the GPS receiver arranged on the non-motor vehicle, including the latitude and longitude coordinates, and the specific position of the first recommended parking space in the parking lot is located according to the extracted first recommended parking space. The geometric information such as the position and shape of the objects such as the lanes, paths and parking space lines in the parking lot is collected, and the information is imported into the geographic information system to construct the digital two-way road network of the parking lot. Taking the real-time position of the non-motor vehicle and the position of the first recommended parking space as the starting point and the ending point, the Dijkstra algorithm is used to search the nodes on the two-way road network of the parking lot, and the path with the shortest distance or the least time consumption is selected as the first recommended path. Then, the calculated first recommended path is sent to the navigation display screen in the non-motor vehicle, and the non-motor vehicle driver is guided to enter the parking lot and drive to the first recommended parking space according to the path in the form of text, picture and voice.

[0076] Moreover, the monitoring camera ID corresponding to the first recommended parking space is obtained, and a connection is established with the corresponding monitoring camera to obtain real-time video image data. By establishing a background model of the parking space without a vehicle in advance, and then comparing with the real-time video image, it is detected whether there is a vehicle in the video. When it is detected that there is a vehicle, it is judged that the first recommended parking space is already parked, and the second recommended parking space needs to be obtained. The second recommended parking space is the parking space ranked behind the first recommended parking space, and then a connection is established with the camera corresponding to the second recommended parking space to continuously analyze the video image and judge whether there is a vehicle parked. If it is detected that there is a vehicle, the third recommended parking space is obtained again until an idle parking space is obtained. Finally, the path to the second recommended parking space, i.e. the first recommended path, is sent to the display screen of the non-motor vehicle to guide the non-motor vehicle driver to drive to the second recommended parking space.

[0077] In summary, the intelligent parking guidance method for non-motor vehicles provided by the embodiments of the present application has the following technical effects:

[0078] The parking destination of the non-motor vehicle is interacted, and information support is provided for vehicle recommendation. The destination parking lot set is obtained based on the parking destination, and the optimal parking lot is selected to provide a basis for parking space recommendation and path planning. The camera monitoring group of the optimal parking lot is interacted, and the parking distribution map of the optimal parking lot is obtained according to the camera monitoring group. The target information is provided for path planning and final guidance according to the parking distribution map and the selection of the recommended parking space set in the optimal parking lot by the non-motor vehicle. The specific driving route is provided for the final guidance according to the parking space set and the non-motor vehicle planning parking path. The non-motor vehicle is guided to park based on the parking path, and the non-motor vehicle driver is real-time guided to drive into the recommended idle parking space to complete the parking and achieve the technical effects of improving the parking experience of the non-motor vehicle and improving the parking efficiency of the non-motor vehicle. Embodiments

[0079] Based on the same inventive concept as the intelligent parking guidance method for non-motor vehicles in the foregoing embodiments, the embodiments of the present application provide an intelligent parking guidance system for non-motor vehicles, as shown in the accompanying drawings, the system comprises: Figure 4

[0080] a destination interaction module 11 for interacting with a parking destination of the non-motor vehicle;

[0081] an optimal parking lot module 12 for obtaining a destination parking lot set based on the parking destination and selecting an optimal parking lot;

[0082] a parking distribution map module 13 for interacting with a camera monitoring group of the optimal parking lot and obtaining a parking distribution map of the optimal parking lot based on the camera monitoring group;

[0083] a recommended parking space module 14 for selecting a recommended parking space set based on the parking distribution map and the non-motor vehicle in the optimal parking lot;

[0084] a parking path planning module 15 for planning a parking path based on the parking space set and the non-motor vehicle;

[0085] a guidance parking module 16 for guiding the non-motor vehicle to park based on the parking path.

[0086] Further, the optimal parking lot module 12 comprises the following execution steps:

[0087] establishing a parking lot database, the parking lot database comprising a parking lot name, an address, a number of parking spaces, a current number of remaining parking spaces, and a fee;

[0088] setting a number of parking lot recommendations;

[0089] traversing the addresses of the parking lots in the parking lot database, sorting the parking lots based on the addresses and the parking destination, and obtaining a recommended parking space ranking table;

[0090] obtaining the destination parking lot set based on the number of parking lot recommendations and the recommended parking space ranking table.

[0091] Further, the optimal parking lot module 12 further comprises the following execution steps:

[0092] constructing a parking lot recommendation model;

[0093] traversing the recommended parking space table to obtain first recommended parking space information;

[0094] obtaining first network recommendation information based on the first recommended parking space information according to big data;

[0095] ​inputting the first recommended parking space information and the first network recommendation information into the parking lot recommendation model to obtain a first parking lot recommendation index;

[0096] obtaining parking lot recommendation indexes of all parking spaces in the parking space recommendation table to obtain a parking lot recommendation index set;

[0097] sorting the parking lot recommendation indexes to obtain an optimal parking lot.

[0098] Further, the optimal parking lot module 12 further includes the following execution steps:

[0099] obtaining a parking lot data set through a crawler technology, the parking lot data set including a corresponding network evaluation information set, wherein the parking lot data and the network evaluation information correspond one by one;

[0100] constructing a model input matrix, the model input matrix being a 2*3 matrix;

[0101] wherein the first row of the 2*3 matrix is the distance, the cost, and the number of remaining parking spaces of the parking lot data in sequence; and the second row of the 2*3 matrix is the parking lot location score, the parking lot order score, and the parking lot scale score in sequence;

[0102] obtaining a sample input matrix set according to the parking lot data set, the network evaluation information set, and the model input matrix;

[0103] matching the sample input matrix set with a recommendation index to obtain a recommendation index set;

[0104] training a linear regression model according to the sample input matrix set and the recommendation index set to obtain a parking recommendation model.

[0105] Further, the optimal parking lot module 12 further includes the following execution steps:

[0106] interacting with the parking demand of the non-motor vehicle;

[0107] setting a location weight, an order weight, and a scale weight according to the parking demand;

[0108] traversing the sample input matrix set to obtain a first parking lot location score, a first parking lot order score, and a first parking lot scale score;

[0109] obtaining a first recommendation index according to the first parking lot location score, the first parking lot order score, the first parking lot scale score, and the location weight, the order weight, and the scale weight;

[0110] adding the first recommendation index to the recommendation index set.

[0111] Further, the recommended parking space module 14 comprises the following execution steps:

[0112] According to the non-motor vehicle type and the optimal parking lot, a parking database is searched to obtain a same-type vehicle parking set;

[0113] According to the same-type vehicle parking set, the parking frequency of different parking spaces is counted, and the parking spaces are sorted according to the parking frequency;

[0114] According to the parking distribution map, the sorting result is eliminated to obtain a recommended parking space set, which is sorted according to the parking frequency.

[0115] Further, the parking path planning module 15 comprises the following execution steps:

[0116] According to the sorting of the recommended parking space set, a parking space is extracted to obtain a first recommended parking space;

[0117] The real-time position of the non-motor vehicle and the first recommended path of the first recommended parking space are established;

[0118] The first recommended path is sent to the display screen of the non-motor vehicle to guide the non-motor vehicle to park;

[0119] The monitoring image of the first recommended parking space is interacted in real time, and when the first recommended parking space has a vehicle, a second recommended parking space is obtained;

[0120] The first recommended path is sent to the display screen of the non-motor vehicle to guide the non-motor vehicle to park.

[0121] Any step of the above method can be stored in a computer memory without limitation as computer instructions or programs, and can be called and recognized by a computer processor without limitation to realize any method in the embodiments of the present application, and no redundant limitation is made herein.

[0122] Further, the above-mentioned first or second may not only represent an order relationship, but also may represent a specific concept, and / or refer to the selection of multiple elements individually or collectively. Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for intelligent parking guidance for non-motorized vehicles, characterized in that, The method comprises: interacting with the parking destination of the non-motor vehicle; acquiring the destination parking lot set based on the parking destination; interacting with the camera monitoring group of the optimal parking lot, and acquiring the parking distribution map of the optimal parking lot according to the camera monitoring group; selecting a recommended parking space set in the optimal parking lot according to the parking distribution map and the non-motor vehicle; planning a parking path according to the parking space set and the non-motor vehicle; guiding the non-motor vehicle to park based on the parking path; the acquiring of the destination parking lot set based on the parking destination comprises: establishing a parking lot database, the parking lot database comprising parking lot name, address, number of parking spaces, current number of remaining parking spaces, and cost; setting a number of parking lot recommendations; traversing the address of the parking lot in the parking lot database, sorting the parking lot according to the address and the parking destination, and acquiring a parking space recommendation ranking table; acquiring the destination parking lot set according to the number of parking lot recommendations and the parking space recommendation ranking table; the selecting of the optimal parking lot comprises: constructing a parking lot recommendation model; traversing the parking space recommendation table to acquire first recommended parking space information; acquiring first network recommendation information based on big data according to the first recommended parking space information; inputting the first recommended parking space information and the first network recommendation information into the parking lot recommendation model to acquire a first parking lot recommendation index; acquiring the parking lot recommendation index of all parking spaces in the parking space recommendation table to acquire a parking lot recommendation index set; sorting the parking lot recommendation index to acquire the optimal parking lot; the construction of the parking lot recommendation model comprises: acquiring a parking lot data set through a crawler technology, the parking lot data set containing a corresponding network evaluation information set, wherein the parking lot data and the network evaluation information correspond one by one; constructing a model input matrix, the model input matrix being a 2*3 matrix; wherein the first row of the 2*3 matrix is the distance, cost, and number of remaining parking spaces of the parking lot data in turn; and the second row of the 2*3 matrix is the parking lot location score, parking lot order score, and parking lot scale score in turn; acquiring a sample input matrix set according to the parking lot data set, the network evaluation information set, and the model input matrix; matching a recommendation index to the sample input matrix set to acquire a recommendation index set; training a linear regression model according to the sample input matrix set and the recommendation index set to acquire a parking recommendation model.

2. The method of claim 1, wherein, the matching of the recommendation index to the sample input matrix set to acquire the recommendation index set comprises: interacting with the parking demand of the non-motor vehicle; setting a location weight, an order weight, and a scale weight according to the parking demand; traversing the sample input matrix set to acquire a first parking lot location score, a first parking lot order score, and a first parking lot scale score; acquiring a first recommendation index according to the first parking lot location score, the first parking lot order score, the first parking lot scale score, and the location weight, the order weight, and the scale weight; adding the first recommendation index to the recommendation index set.

3. The method of claim 1, wherein, The selecting a recommended parking space set in the optimal parking lot according to the parking distribution map and the non-motor vehicle comprises: According to the non-motor vehicle type and the optimal parking lot, a parking database is searched to obtain a same-type vehicle parking set; According to the same-type vehicle parking set, the parking frequencies of different parking spaces are counted, and the parking spaces are sorted according to the parking frequencies; According to the parking distribution map, the sorting result is eliminated to obtain a recommended parking space set, which is sorted according to the parking frequencies.

4. The method of claim 1, wherein, The planning a parking path according to the parking space set and the non-motor vehicle comprises: According to the sorting of the recommended parking space set, a first recommended parking space is extracted to obtain a first recommended parking space; A first recommended path of the non-motor vehicle is established and the first recommended path is sent to a display screen of the non-motor vehicle to guide the non-motor vehicle to park; Real-time interaction of the first recommended parking space image is performed, and when the first recommended parking space has a vehicle, a second recommended parking space is obtained; The first recommended path is sent to the display screen of the non-motor vehicle to guide the non-motor vehicle to park. The system for implementing the intelligent parking guidance method for the non-motor vehicle according to any one of claims 1-4 comprises:

5. An intelligent parking guidance system for non-motorized vehicles, characterized in that, A destination interaction module for interacting with the parking destination of the non-motor vehicle; An optimal parking lot module for obtaining a destination parking lot set based on the parking destination and selecting an optimal parking lot; A parking distribution map module for interacting with a camera monitoring group of the optimal parking lot and obtaining a parking distribution map of the optimal parking lot according to the camera monitoring group; A recommended parking space module for selecting a recommended parking space set in the optimal parking lot according to the parking distribution map and the non-motor vehicle; A parking path planning module for planning a parking path according to the parking space set and the non-motor vehicle; A guidance parking module for guiding the non-motor vehicle to park based on the parking path. ​

Images