Intelligent parking reservation control and adjustment method and system

Abstract

The invention provides an intelligent parking reservation control adjustment method and system, and the method comprises the steps: obtaining user information with a parking space reservation demand, the user information comprises a user license plate number, a user vehicle type and a user current position, and recommending a parking space for a user based on a scoring model; feedback of a user to the recommended parking space is obtained, and the weight in the scoring model is updated according to the feedback of the user; and applying collaborative filtering enhancement to each user to update the weight again every preset time. Through the method and the corresponding system, the personalized experience of the user can be improved, and the overall operation efficiency and recommendation accuracy of the system can be improved through continuous learning and optimization.

Description

Technical Field

[0001] The present invention provides an intelligent parking reservation control and adjustment method and system, belonging to the technical field of ... Background Art

[0002] With the acceleration of urbanization and the rapid growth of vehicle ownership, urban parking problems are becoming increasingly serious, leading to traffic congestion and environmental pollution. Traditional parking methods often rely on car owners to find vacant parking spaces by themselves, which is not only time-consuming and labor-intensive, but also aggravates the burden of urban traffic. To this end, intelligent parking systems have emerged with the aim of optimizing the allocation of parking resources and reducing the idle time of vehicles when looking for parking spaces through efficient technical means. Most existing intelligent parking systems adopt a central control method, which centrally processes all parking information and requests and then dispatches them uniformly. These systems usually rely on fixed rules to operate, such as distance priority, price priority, etc., but often ignore the personalized needs and preferences of users; with the rise of personalized services and the development of artificial intelligence technology, it has become possible to provide users with customized parking solutions. However, to achieve this goal, it is necessary to process and analyze a large amount of user data and update the system in real time to respond to changes in user preferences. In addition, the addition of new users also brings about the cold start problem, that is, how to effectively provide services without sufficient user information. Therefore, there is an urgent need to develop an intelligent parking reservation control method and system that can effectively integrate and utilize user preferences, dynamically learn user behaviors in real time, and use collaborative filtering technology to solve the cold start problem. Summary of the invention

[0003] The present invention provides an intelligent parking reservation control and adjustment method and system to solve the above-mentioned problems:

[0004] The present invention provides an intelligent parking reservation control and adjustment method, the method comprising:

[0005] Obtaining user information that needs to reserve a parking space, the user information includes: the user's license plate number, the user's vehicle model, and the user's current location, and recommending a parking space for the user based on a scoring model;

[0006] Obtaining user feedback on the recommended parking space, and updating the weights in the scoring model according to the user feedback;

[0007] At preset intervals, collaborative filtering enhancement is applied to each user to update the weights again.

[0008] Further, obtaining user information of a user who needs to reserve a parking space, the user information including: the user's license plate number and the user's current location, and selecting a parking space for the user based on the scoring model, including:

[0009] Obtaining user information that requires a parking space reservation, the user information including: the user's license plate number, the size of the user's vehicle model, and the user's current location;

[0010] Obtain all available parking spaces whose distance from the user's current location does not exceed the user's preset distance threshold, obtain the weather during the user's reserved parking time period through the API, and recommend a parking space to the user based on the score calculated by the scoring model. Specifically, the scoring model is:

[0011] s(p,u,t,w)=w a D(u,p)+w b C(p)+w c W(w,p)+w d T(t,p)+w e S(p)

[0012] Among them, s(p,u,t,w) represents the score of the parking space, w a 、w b 、w c 、w d 、w e and w f represents the weight of each sub-score, D(u, p) represents the distance adjustment factor, C(p) represents the parking difficulty coefficient, W(w, p) represents the weather adjustment factor, F(p) represents the coverage facility addition coefficient, T(t, p) represents the time sensitivity coefficient, and S(p) represents the safety factor;

[0013]

[0014] Where d(u, p) represents the driving distance from the user's current location to parking space p, and δ is the distance sensitivity coefficient;

[0015] C(p)=e -θ×c(p)

[0016] Among them, c(p) represents the parking difficulty index, which can be represented by the average time spent on historical parking in the parking space, and θ represents the difficulty sensitivity coefficient.

[0017] W(w,p)=max(0,1-ξ×f(w,p))

[0018] Among them, f(w, p) represents the impact of weather on the availability of the parking space, and ξ represents the weather sensitivity coefficient;

[0019] S(p)=1+μ×s(p)

[0020] Among them, s(p) represents the safety index of the parking space, and μ represents the safety weighted coefficient;

[0021]

[0022] Among them, λ represents the time sensitivity coefficient, T is the period, a day is 24 hours, t is the current time, which needs to be converted into a decimal form of hours, φ is the phase offset, which is adjusted according to the specific time when the peak demand begins, and C is the vertical offset.

[0023] Furthermore, obtaining user feedback on the recommended parking space and updating the weights in the scoring model according to the user feedback include:

[0024] Initialize each user’s preference vector P u ;

[0025] For each parking request, the user's feedback score on the recommended parking space is obtained, and the user preference weight is updated based on the dynamic weight learning model. Specifically, the dynamic weight learning model is:

[0026]

[0027] in, is the weight of user u on factor i at time step t, η is the learning rate, r t is the user's feedback rating for the current recommendation, is the average feedback score, f i (s t ) is the score of the current recommended parking space on factor i, is the average score of factor i.

[0028] Furthermore, at preset intervals, collaborative filtering enhancement is applied to each user to update the weight again, including:

[0029] Get the k most similar users to the updated weighted user;

[0030] The updated weight of the user is calculated based on the collaborative filtering model. Specifically, the collaborative filtering model is:

[0031]

[0032] Among them, w u，i is the weight of user u on factor i, η is the learning rate, The weight after combining collaborative information, α is the balance parameter between individual preference and group preference, N k (u) is the k most similar users to user u, and sim(u,v) is the similarity between users u and v.

[0033] The present invention proposes an intelligent parking reservation control and adjustment system, the system comprising:

[0034] The parking space recommendation module is used to obtain user information that needs to reserve a parking space, including the user's license plate number, the user's vehicle model, and the user's current location, and recommend a parking space to the user based on the scoring model;

[0035] An updating weight module is used to obtain user feedback on the recommended parking space and update the weights in the scoring model according to the user feedback;

[0036] The enhanced updating weight module is used to apply collaborative filtering enhancement to each user and update the weight again at preset intervals.

[0037] Furthermore, the parking space recommendation module includes:

[0038] The user information acquisition module is used to acquire the user information of the user who needs to reserve a parking space, and the user information includes: the user's license plate number, the size of the user's vehicle model and the user's current location;

[0039] The scoring module is used to obtain all available parking spaces whose distance from the user's current location does not exceed the user's preset distance threshold, obtain the weather during the user's reserved parking time period through the API, and recommend parking spaces to the user based on the score calculated by the scoring model. Specifically, the scoring model is:

[0040] s(p,u,t,w)=w a D(u,p)+w b C(p)+w c W(w,p)+w d T(t,p)+w e S(p)

[0041] Among them, s(p,u,t,w) represents the score of the parking space, w a 、w b 、w c 、w d 、w e and w f represents the weight of each sub-score, D(u, p) represents the distance adjustment factor, C(p) represents the parking difficulty coefficient, W(w, p) represents the weather adjustment factor, F(p) represents the coverage facility addition coefficient, T(t, p) represents the time sensitivity coefficient, and S(p) represents the safety factor;

[0042]

[0043] Where d(u, p) represents the driving distance from the user's current location to parking space p, and δ is the distance sensitivity coefficient;

[0044] C(p)=e -θ×c(p)

[0045] Among them, c(p) represents the parking difficulty index, which can be represented by the average time spent on historical parking in the parking space, and θ represents the difficulty sensitivity coefficient.

[0046] W(w,p)=max(0,1-ξ×f(w,p))

[0047] Among them, f(w, p) represents the impact of weather on the availability of the parking space, and ξ represents the weather sensitivity coefficient;

[0048] S(p)=1+μ×s(p)

[0049] Among them, s(p) represents the safety index of the parking space, and μ represents the safety weighted coefficient;

[0050]

[0051] Among them, λ represents the time sensitivity coefficient, T is the period, a day is 24 hours, t is the current time, which needs to be converted into a decimal form of hours, φ is the phase offset, which is adjusted according to the specific time when the peak demand begins, and C is the vertical offset.

[0052] Furthermore, the weight updating module includes:

[0053] Initialize preference vector module, used to initialize the preference vector P of each user u ;

[0054] The calculation and update weight module is used to obtain the user's feedback score on the recommended parking space for each parking request, and update the user preference weight based on the dynamic weight learning model. Specifically, the dynamic weight learning model is:

[0055]

[0056] in, is the weight of user u on factor i at time step t, η is the learning rate, r t is the user's feedback rating for the current recommendation, is the average feedback score, f i (s t ) is the score of the current recommended parking space on factor i, is the average score of factor i

[0057] Furthermore, the enhanced update weight module includes:

[0058] A similar user acquisition module is used to acquire the k most similar users to the updated weighted user;

[0059] The filter weight calculation module is used to calculate the updated weight of the user based on the collaborative filtering model. Specifically, the collaborative filtering model is:

[0060]

[0061] Among them, w u，i is the weight of user u on factor i, η is the learning rate, The weight after combining collaborative information, α is the balance parameter between individual preference and group preference, N k (u) is the k most similar users to user u, and sim(u,v) is the similarity between users u and v.

[0062] The beneficial effects of the present invention are as follows: personalized service is improved, by recommending parking spaces based on the user's personal information and preferences, the user experience is more personalized and the satisfaction is higher; the recommendation accuracy is improved, the dynamic update of the scoring model and the regular collaborative filtering enhancement ensure that the system is constantly learning and optimizing, and the accuracy and adaptability of the recommendations are improved; user satisfaction is enhanced, and the recommendation strategy is adjusted in response to user feedback in a timely manner, thereby enhancing the user's trust in and satisfaction with the system; resource allocation is optimized, and more accurate parking recommendations reduce the time users spend looking for parking spaces, help alleviate urban traffic pressure, and improve the efficiency of parking resource utilization; new user problems are solved, and collaborative filtering technology helps solve the cold start problem of new users or users with less data, and can provide them with reasonable parking recommendations more quickly; this technical solution can not only enhance the user's personalized experience, but also improve the overall operating efficiency and recommendation accuracy of the system through continuous learning and optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a schematic diagram of an intelligent parking reservation control and adjustment method described in the present invention. DETAILED DESCRIPTION

[0064] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0065] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. The embodiments described are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0067] One embodiment of the present invention provides a smart parking reservation control and adjustment method, the method comprising:

[0068] Obtaining user information that needs to reserve a parking space, the user information includes: the user's license plate number, the user's vehicle model, and the user's current location, and recommending a parking space for the user based on a scoring model;

[0069] Obtaining user feedback on the recommended parking space, and updating the weights in the scoring model according to the user feedback;

[0070] At preset intervals, collaborative filtering enhancement is applied to each user to update the weights again.

[0071] The working principle and effect of the above technical solution are as follows: to obtain user information, the system first captures the license plate number, vehicle model and current location entered by the user. This step is the beginning of collecting necessary information to ensure the accuracy of subsequent recommended parking spaces. The user's vehicle model can be used to determine the size of the required parking space, and the current location is used to optimize the distance calculation; based on the information provided by the user, the system uses the established scoring model to calculate the score of each parking space. The scoring model integrates the user's personalized preferences (such as distance, price, safety and other factors) and the specific conditions of the parking space (such as location, cost, space size, etc.). The system will recommend the best parking space to the user based on the calculated score; collect user feedback, and the user will provide feedback on the recommendation results, which can be acceptance or rejection of the recommendation, or evaluation of the parking experience after parking. This user feedback is a very important data source for adjusting the scoring model; dynamic weight update, based on the user's actual feedback, the system automatically adjusts the weights in the scoring model. This step uses an online learning algorithm to ensure that the model can reflect the user's latest preferences and needs in real time; the system uses collaborative filtering technology to make a comprehensive analysis and optimization of the preferences of all users at preset time intervals (such as weekly or monthly). This step improves the quality of recommendations for new users or users with less data by considering the preferences of similar users, and also recalibrates the preference models of all users. Personalized service improvement: by integrating users' personal information and preferences to recommend parking spaces, the user experience is more personalized and the satisfaction is higher; the accuracy of recommendations is improved: dynamically updating the scoring model and regular collaborative filtering enhancements ensure that the system is constantly learning and optimizing, improving the accuracy and adaptability of recommendations; user satisfaction is enhanced: timely response to user feedback to adjust the recommendation strategy, and user trust and satisfaction with the system are strengthened; resource allocation is optimized: more accurate parking recommendations reduce the time users spend looking for parking spaces, help alleviate urban traffic pressure, and improve the efficiency of parking resource utilization; solving new user problems: collaborative filtering technology helps solve the cold start problem of new users or users with less data, and can provide them with reasonable parking recommendations more quickly; this technical solution can not only enhance users' personalized experience, but also improve the overall system operation efficiency and recommendation accuracy through continuous learning and optimization.

[0072] An embodiment of the present invention is a smart parking reservation control and adjustment method, which obtains user information that requires a parking space reservation, wherein the user information includes: the user's license plate number and the user's current location, and selects a parking space for the user based on a scoring model, including:

[0073] Obtaining user information that requires a parking space reservation, the user information including: the user's license plate number, the size of the user's vehicle model, and the user's current location;

[0074] Obtain all available parking spaces whose distance from the user's current location does not exceed the user's preset distance threshold, obtain the weather during the user's reserved parking time period through the API, and recommend a parking space to the user based on the score calculated by the scoring model. Specifically, the scoring model is:

[0075] S(p,u,t,w)=w a D(u,p)+w b C(p)+w c W(w,p)+w d T(t,p)+w e S(p)

[0076] Among them, S(p,u,t,w) represents the score of the parking space, w a 、w b 、w c 、w d 、w e and w f represents the weight of each sub-score, D(u, p) represents the distance adjustment factor, C(p) represents the parking difficulty coefficient, W(w, p) represents the weather adjustment factor, F(p) represents the coverage facility addition coefficient, T(t, p) represents the time sensitivity coefficient, and S(p) represents the safety factor;

[0077]

[0078] Where d(u, p) represents the driving distance from the user's current location to parking space p, and δ is the distance sensitivity coefficient;

[0079] C(p)=e -θ×c(p)

[0080] Among them, c(p) represents the parking difficulty index, which can be represented by the average time spent on historical parking in the parking space, and θ represents the difficulty sensitivity coefficient.

[0081] W(w,p)=max(0,1-ξ×f(w,p))

[0082] Among them, f(w, p) represents the impact of weather on the availability of the parking space, and ξ represents the weather sensitivity coefficient;

[0083] S(p)=1+μ×s(p)

[0084] Among them, s(p) represents the safety index of the parking space, and μ represents the safety weighted coefficient;

[0085]

[0086] Among them, λ represents the time sensitivity coefficient, T is the period, a day is 24 hours, t is the current time, which needs to be converted into a decimal form of hours, φ is the phase offset, which is adjusted according to the specific time when the peak demand begins, and C is the vertical offset.

[0087] In one embodiment of the present invention, obtaining user feedback on the recommended parking space and updating the weights in the scoring model according to the user feedback include:

[0088] Initialize each user’s preference vector P u ;

[0089] For each parking request, the user's feedback score on the recommended parking space is obtained, and the user preference weight is updated based on the dynamic weight learning model. Specifically, the dynamic weight learning model is:

[0090]

[0091] in, is the weight of user u on factor i at time step t, η is the learning rate, r t is the user's feedback rating for the current recommendation, is the average feedback score, f i (s t ) is the score of the current recommended parking space on factor i, is the average score of factor i.

[0092] In one embodiment of the present invention, the weight is updated again by applying collaborative filtering enhancement to each user at a preset time interval, including:

[0093] Get the k most similar users to the updated weighted user;

[0094] The updated weight of the user is calculated based on the collaborative filtering model. Specifically, the collaborative filtering model is:

[0095]

[0096] Among them, w u，i is the weight of user u on factor i, η is the learning rate, The weight after combining collaborative information, α is the balance parameter between individual preference and group preference, N k(u) is the k most similar users to user u, and sim(u,v) is the similarity between users u and v.

[0097] One embodiment of the present invention provides an intelligent parking reservation control and adjustment system, the system comprising:

[0098] The parking space recommendation module is used to obtain user information that needs to reserve a parking space, including the user's license plate number, the user's vehicle model, and the user's current location, and recommend a parking space to the user based on the scoring model;

[0099] An updating weight module is used to obtain user feedback on the recommended parking space and update the weights in the scoring model according to the user feedback;

[0100] The enhanced updating weight module is used to apply collaborative filtering enhancement to each user and update the weight again at preset intervals.

[0101] In one embodiment of the present invention, a smart parking reservation control and adjustment system is provided, wherein the parking space recommendation module comprises:

[0102] The user information acquisition module is used to acquire the user information of the user who needs to reserve a parking space, and the user information includes: the user's license plate number, the size of the user's vehicle model and the user's current location;

[0103] The scoring module is used to obtain all available parking spaces whose distance from the user's current location does not exceed the user's preset distance threshold, obtain the weather during the user's reserved parking time period through the API, and recommend parking spaces to the user based on the score calculated by the scoring model. Specifically, the scoring model is:

[0104] s(p,u,t,w)=w a D(u,p)+w b C(p)+w c W(w,p)+w d T(t,p)+w e S(p)

[0105] Among them, s(p,u,t,w) represents the score of the parking space, w a 、w b 、w c 、w d 、w e and w f represents the weight of each sub-score, D(u, p) represents the distance adjustment factor, C(p) represents the parking difficulty coefficient, W(w, p) represents the weather adjustment factor, F(p) represents the coverage facility addition coefficient, T(t, p) represents the time sensitivity coefficient, and S(p) represents the safety factor;

[0106]

[0107] Where d(u, p) represents the driving distance from the user's current location to parking space p, and δ is the distance sensitivity coefficient;

[0108] C(p)=e -θ×c(p)

[0109] Among them, c(p) represents the parking difficulty index, which can be represented by the average time spent on historical parking in the parking space, and θ represents the difficulty sensitivity coefficient.

[0110] W(w,p)=max(0,1-ξ×f(w,p))

[0111] Among them, f(w, p) represents the impact of weather on the availability of the parking space, and ξ represents the weather sensitivity coefficient;

[0112] S(p)=1+μ×s(p)

[0113] Among them, s(p) represents the safety index of the parking space, and μ represents the safety weighted coefficient;

[0114]

[0115] Among them, λ represents the time sensitivity coefficient, T is the period, a day is 24 hours, t is the current time, which needs to be converted into a decimal form of hours, φ is the phase offset, which is adjusted according to the specific time when the peak demand begins, and C is the vertical offset.

[0116] In one embodiment of the present invention, a smart parking reservation control and adjustment system, the weight update module includes:

[0117] Initialize preference vector module, used to initialize the preference vector P of each user u ;

[0118] The calculation and update weight module is used to obtain the user's feedback score on the recommended parking space for each parking request, and update the user preference weight based on the dynamic weight learning model. Specifically, the dynamic weight learning model is:

[0119]

[0120] in, is the weight of user u on factor i at time step t, η is the learning rate, r t is the user's feedback rating for the current recommendation, is the average feedback score, f i (s t ) is the score of the current recommended parking space on factor i, is the average score of factor i

[0121] In one embodiment of the present invention, a smart parking reservation control and adjustment system, the enhanced update weight module includes:

[0122] A similar user acquisition module is used to acquire the k most similar users to the updated weighted user;

[0123] The filter weight calculation module is used to calculate the updated weight of the user based on the collaborative filtering model. Specifically, the collaborative filtering model is:

[0124]

[0125] Among them, w u，i is the weight of user u on factor i, η is the learning rate, The weight after combining collaborative information, α is the balance parameter between individual preference and group preference, N k (u) is the k most similar users to user u, and sim(u,v) is the similarity between users u and v.

[0126] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An intelligent parking reservation control and adjustment method, characterized in that: The method comprises: Obtaining user information that needs to reserve a parking space, the user information includes: the user's license plate number, the user's vehicle model, and the user's current location, and recommending a parking space for the user based on a scoring model; Obtaining user feedback on the recommended parking space, and updating the weights in the scoring model according to the user feedback; At preset intervals, collaborative filtering enhancement is applied to each user to update the weights again.

2. According to claim 1, a smart parking reservation control and adjustment method is characterized in that: Obtaining user information that needs to reserve a parking space, the user information includes: the user's license plate number and the user's current location, and selecting a parking space for the user based on the scoring model, including: Obtaining user information that requires a parking space reservation, the user information including: the user's license plate number, the size of the user's vehicle model, and the user's current location; Obtain all available parking spaces whose distance from the user's current location does not exceed the user's preset distance threshold, obtain the weather during the user's reserved parking time period through the API, and recommend a parking space to the user based on the score calculated by the scoring model. Specifically, the scoring model is: S(p,u,t,w)=w a D(u,p)+w b C(p)+w c W(w,p)+w d T(t,p)+w e S(p) Among them, S(p,u,t,w) represents the score of the parking space, w a 、w b 、w c 、w d 、w e and w f represents the weight of each sub-score, D(u, p) represents the distance adjustment factor, C(p) represents the parking difficulty coefficient, W(w, p) represents the weather adjustment factor, F(p) represents the coverage facility addition coefficient, T(t, p) represents the time sensitivity coefficient, and S(p) represents the safety factor; Where d(u, p) represents the driving distance from the user's current location to parking space p, and δ is the distance sensitivity coefficient; C(p)=e -θ×c(p) Among them, c(p) represents the parking difficulty index, which can be represented by the average time spent on historical parking in the parking space, and θ represents the difficulty sensitivity coefficient. W(w,p)=max(0,1-ξ×f(w,p)) Among them, f(w, p) represents the impact of weather on the availability of the parking space, and ξ represents the weather sensitivity coefficient; S(p)=1+μ×s(p) Among them, s(p) represents the safety index of the parking space, and μ represents the safety weighted coefficient; Among them, λ represents the time sensitivity coefficient, T is the period, a day is 24 hours, t is the current time, which needs to be converted into a decimal form of hours, φ is the phase offset, which is adjusted according to the specific time when the peak demand begins, and C is the vertical offset.

3. According to claim 1, a smart parking reservation control and adjustment method is characterized in that: Obtaining user feedback on the recommended parking space, and updating the weights in the scoring model according to the user feedback, including: Initialize each user’s preference vector P u ; For each parking request, the user's feedback score on the recommended parking space is obtained, and the user preference weight is updated based on the dynamic weight learning model. Specifically, the dynamic weight learning model is: in, is the weight of user u on factor i at time step t, η is the learning rate, r t is the user's feedback rating for the current recommendation, is the average feedback score, f i (s t ) is the score of the current recommended parking space on factor i, f i is the average score of factor i.

4. According to claim 3, a smart parking reservation control and adjustment method is characterized in that: At preset intervals, collaborative filtering enhancement is applied to each user to update the weight again, including: Get the k most similar users to the updated weighted user; The updated weight of the user is calculated based on the collaborative filtering model. Specifically, the collaborative filtering model is: Among them, w u，i is the weight of user u on factor i, η is the learning rate, The weight after combining collaborative information, α is the balance parameter between individual preference and group preference, N k (u) are the k most similar users to user u, and sim(u, v) is the similarity between users u and v.

5. An intelligent parking reservation control and adjustment system, characterized in that: The system comprises: The parking space recommendation module is used to obtain user information that needs to reserve a parking space, including the user's license plate number, the user's vehicle model, and the user's current location, and recommend a parking space to the user based on the scoring model; An updating weight module is used to obtain user feedback on the recommended parking space and update the weights in the scoring model according to the user feedback; The enhanced updating weight module is used to apply collaborative filtering enhancement to each user and update the weight again at preset intervals.

6. The intelligent parking reservation control and adjustment system according to claim 5, characterized in that: The recommended parking space module includes: The user information acquisition module is used to acquire the user information of the user who needs to reserve a parking space, and the user information includes: the user's license plate number, the size of the user's vehicle model and the user's current location; The scoring module is used to obtain all available parking spaces whose distance from the user's current location does not exceed the user's preset distance threshold, obtain the weather during the user's reserved parking time period through the API, and recommend parking spaces to the user based on the score calculated by the scoring model. Specifically, the scoring model is: S(p,u,t,w)=w a D(u,p)+w b C(p)+w c W(w,p)+w d T(t,p)+w e S(p) Among them, S(p,u,t,w) represents the score of the parking space, w a 、w b 、w c 、w d 、w e and w f represents the weight of each sub-score, D(u, p) represents the distance adjustment factor, C(p) represents the parking difficulty coefficient, W(w, p) represents the weather adjustment factor, F(p) represents the coverage facility addition coefficient, T(t, p) represents the time sensitivity coefficient, and S(p) represents the safety factor; Where d(u, p) represents the driving distance from the user's current location to parking space p, and δ is the distance sensitivity coefficient; C(p)=e -θ×c(p) Among them, c(p) represents the parking difficulty index, which can be represented by the average time spent on historical parking in the parking space, and θ represents the difficulty sensitivity coefficient. W(w,p)=max(0,1-ξ×f(w,p)) Among them, f(w, p) represents the impact of weather on the availability of the parking space, and ξ represents the weather sensitivity coefficient; S(p)=1+μ×s(p) Among them, s(p) represents the safety index of the parking space, and μ represents the safety weighted coefficient; Among them, λ represents the time sensitivity coefficient, T is the period, a day is 24 hours, t is the current time, which needs to be converted into a decimal form of hours, φ is the phase offset, which is adjusted according to the specific time when the peak demand begins, and C is the vertical offset.

7. The intelligent parking reservation control and adjustment system according to claim 5, characterized in that: The weight updating module comprises: Initialize preference vector module, used to initialize the preference vector P of each user u ; The calculation and update weight module is used to obtain the user's feedback score on the recommended parking space for each parking request, and update the user preference weight based on the dynamic weight learning model. Specifically, the dynamic weight learning model is: in, is the weight of user u on factor i at time step t, η is the learning rate, r t is the user's feedback rating for the current recommendation, is the average feedback score, f i (s t ) is the score of the current recommended parking space on factor i, is the average score of factor i.

8. The intelligent parking reservation control and adjustment system according to claim 5, characterized in that: The enhanced update weight module includes: A similar user acquisition module is used to acquire the k most similar users to the updated weighted user; The filter weight calculation module is used to calculate the updated weight of the user based on the collaborative filtering model. Specifically, the collaborative filtering model is: Among them, w u，i is the weight of user u on factor i, η is the learning rate, The weight after combining collaborative information, α is the balance parameter between individual preference and group preference, N k (u) are the k most similar users to user u, and sim(u, v) is the similarity between users u and v.

Images