A Service Area Remaining Parking Space Prediction and Parking Guidance Method Based on Time Series Analysis

Abstract

This invention relates to the field of parking space prediction and parking guidance technology, specifically to a service area remaining parking space prediction and parking guidance method based on time-series analysis. The method includes: obtaining a comprehensive stability coefficient based on the vehicle type and any parking space type; obtaining a highway traffic flow impact coefficient based on the number of vehicles on the highway heading to the service area that match the parking space type and the parking space's vacancy time, and the corresponding number of available parking spaces; obtaining a comparison benchmark based on the historical time period, traffic density, and dwell time corresponding to the current moment; obtaining a deviation degree of traffic flow dwelling pattern based on the comparison benchmark and a preset off-peak time benchmark; obtaining the probability that a vehicle will occupy an available parking space of a certain type based on the comprehensive stability coefficient, highway traffic flow impact coefficient, and traffic flow dwelling pattern deviation degree; and guiding the vehicle based on the comparison result between the probability and a preset threshold. Using this invention, parking space prediction and parking guidance can be performed with relatively high accuracy.

Description

Technical Field

[0001] This invention relates to the field of parking space prediction and parking guidance technology, specifically to a method for predicting and guiding remaining parking spaces in service areas based on time-series analysis. Background Technology

[0002] In service area parking management, existing technologies typically rely on time-series data to predict parking space availability and provide guidance. This new method continuously collects parking space status and traffic flow data, combining historical patterns to predict parking space supply and demand trends in different areas, and then recommends optimal parking areas to drivers. This aims to help drivers quickly find parking spaces, avoid congestion caused by disorderly searching for spaces, promote the balanced use of parking resources, and effectively improve overall parking efficiency.

[0003] However, the aforementioned traditional forecasting methods are mainly designed for ordinary parking lots with a single type of vehicle and uniform parking space specifications. Service areas, on the other hand, face the core challenge of complex vehicle types (small cars, new energy vehicles, and large trucks coexisting) and diverse parking space types (standard parking spaces, charging parking spaces, and large parking spaces). The parking needs of different vehicle types vary significantly, and traditional methods, lacking consideration of the fine-grained matching logic between "vehicle type and parking space," often lead to a mismatch between the predicted available parking spaces and the actual parking demand of vehicles. Summary of the Invention

[0004] To address the technical problem of inaccurate parking space prediction and parking guidance in service areas, this invention aims to provide a method for predicting remaining parking spaces and guiding parking in service areas based on time-series analysis. The specific technical solution adopted is as follows:

[0005] Get the vehicle model of any vehicle currently waiting to stop at the service area entrance;

[0006] The overall stability coefficient is obtained based on the vehicle model and any parking space type.

[0007] The highway traffic flow impact coefficient is obtained based on the number of vehicles on the highway that meet the parking space type and parking space vacancy time and the corresponding number of available parking spaces.

[0008] The comparison benchmark is obtained based on the historical time period, traffic density, and dwell time corresponding to the current moment. The deviation of the traffic dwell pattern is obtained based on the comparison benchmark and the preset off-peak time benchmark.

[0009] The probability of a vehicle waiting to stop occupying an empty parking space of the specified parking space type is obtained based on the comprehensive stability coefficient, the highway traffic flow influence coefficient, and the deviation of the traffic flow dwelling pattern. The vehicle waiting to stop is then guided based on the comparison result between the probability and a preset threshold.

[0010] Furthermore, the process of obtaining the comprehensive stability coefficient includes:

[0011] Calculate the size matching coefficient based on the vehicle model and the parking space type. During a preset historical period, when a vehicle of the vehicle model is parked in a parking space of the parking space type, calculate the function utilization rate based on the duration of the vehicle's use of the parking space function and the total duration of its stay.

[0012] Based on the size matching coefficient and the functional utilization rate, obtain the appropriate rationality coefficient;

[0013] Within the preset historical time period, the basic dwell stability is obtained based on the dwell time of the vehicle of the vehicle type in the parking space of the parking space type.

[0014] The appropriate rationality coefficient is used to correct the basic dwell stability to obtain the dwell regularity coefficient;

[0015] Within the preset historical time period, the necessity coefficient of mixed parking is obtained based on the mixed parking information of vehicles of the vehicle type in parking spaces of the parking space type.

[0016] The comprehensive stability coefficient is obtained based on the mixed-stop necessity coefficient and the dwell regularity coefficient.

[0017] Furthermore, the process of obtaining the dwell regularity coefficient includes:

[0018] The relative stability is obtained based on the aforementioned basic dwell stability.

[0019] The dwell regularity coefficient is quantified by the product of the relative stability and the fit rationality coefficient.

[0020] Furthermore, the process of obtaining the mixed-stop necessity coefficient includes:

[0021] The mixed-stop information includes: the number of random mixed-stop records and the total number of mixed-stop records;

[0022] The necessity coefficient for random parking is obtained based on the proportion of the number of random parking records to the total number of random parking records.

[0023] Furthermore, the process of obtaining the comprehensive stability coefficient includes:

[0024] The mixed-stop necessity coefficient and the dwell regularity coefficient are used to characterize the actual disturbance stability;

[0025] The dwell regularity coefficient is used to characterize ideal stability;

[0026] The comprehensive stability coefficient is obtained based on the arithmetic mean of the actual disturbance stability and the ideal stability.

[0027] Furthermore, the process of obtaining the highway traffic flow impact coefficient includes:

[0028] The intensity of parking space competition is determined by the ratio of the number of vehicles on the highway that meet the parking space type and the parking space vacancy time to the corresponding number of available parking spaces.

[0029] The highway traffic flow impact coefficient is obtained by limiting the intensity of competition for parking spaces within a reasonable probability range.

[0030] Furthermore, the process of obtaining the deviation of the traffic flow dwell pattern includes:

[0031] The difference between the benchmark to be compared and the off-peak time benchmark is calculated, and the deviation of the traffic flow dwell pattern is used to measure the difference.

[0032] Furthermore, the process of obtaining the benchmark to be compared includes:

[0033] Obtain vehicle information for the historical time period corresponding to the current moment, use the traffic density and the dwell time as interval features, and record the number of vehicles that meet the interval features as interval record data;

[0034] The interval probability is obtained based on the interval recorded data and the total interval recorded data. The comparison benchmark is the set of interval probabilities, wherein the total interval recorded data is the sum of the interval recorded data.

[0035] Furthermore, the method further includes: determining the off-peak period before obtaining the off-peak period benchmark, wherein the process of determining the off-peak period includes:

[0036] Divide the historical days corresponding to the current day into at least two time periods according to a preset time length;

[0037] The traffic density, dwell time, and turnover rate are used as clustering features to cluster the time period;

[0038] The class corresponding to the one with the highest stability is selected as the off-peak period.

[0039] Furthermore, the process of obtaining the probability includes:

[0040] The initial idle potential is determined based on the aforementioned comprehensive stability coefficient;

[0041] The highway traffic flow influence coefficient and the deviation of the traffic flow dwell pattern are used to correct the initial idle potential, and the corrected results are normalized to a non-negative unit interval to obtain the probability.

[0042] The present invention has the following beneficial effects:

[0043] First, obtain the vehicle model of any vehicle currently waiting to stop at the service area entrance. This vehicle model is used for subsequent analysis of the overall stability coefficient.

[0044] Secondly, a comprehensive stability coefficient is obtained based on the vehicle model and any parking space type. This comprehensive stability coefficient measures the compatibility between the vehicle model and the parking space type. The higher the comprehensive stability coefficient, the greater the compatibility.

[0045] Next, the highway traffic flow impact coefficient is obtained based on the number of vehicles on the highway heading to the service area that match the parking space type and the parking space vacancy time, and the corresponding number of available parking spaces. The highway traffic flow impact coefficient represents the degree of competition of highway traffic for available parking spaces of the specified parking space type. The larger the highway traffic flow impact coefficient, the less likely the vehicles waiting to stop are to occupy available parking spaces of the specified parking space type.

[0046] Then, a comparison benchmark is obtained based on the historical time period, traffic density, and dwell time corresponding to the current moment. The deviation of the traffic flow dwell pattern is obtained based on the comparison benchmark and the preset off-peak time benchmark. The deviation of the traffic flow dwell pattern reflects the degree of consumption of the parking space vacancy potential: the more serious the deviation, the faster the potential is consumed, and the more difficult it is for parking spaces to become vacant.

[0047] Finally, based on the comprehensive stability coefficient, the highway traffic flow impact coefficient, and the deviation of the traffic flow dwelling pattern, the probability that a vehicle waiting to stop will occupy an empty parking space of the specified parking type is obtained. The vehicle is then guided based on a comparison of this probability with a preset threshold. When the probability is higher than the preset threshold, the vehicle is guided to park in the service area, and a prompt can be displayed on the service area entrance guidance screen: "Please proceed to area XX," etc. When the probability is lower than the preset threshold, the vehicle is guided to wait or leave, and a prompt can be displayed on the service area entrance guidance screen: "Parking spaces are currently limited; please wait patiently," etc.

[0048] This invention can analyze the matching rules between vehicle type and parking space, clarify the adaptation relationship and dwell characteristics, and then incorporate the time dimension to build a prediction framework that is more in line with the real scene of the service area, so that the prediction results can accurately serve parking space allocation and vehicle guidance. Attached Figure Description

[0049] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart of the service area remaining parking space prediction and parking guidance method based on time series analysis provided in the first embodiment of the present invention;

[0051] Figure 2 A flowchart illustrating the process of obtaining the comprehensive stability coefficient provided in the second embodiment of the present invention;

[0052] Figure 3 This is a flowchart illustrating the process of obtaining the highway traffic flow impact coefficient according to the third embodiment of the present invention.

[0053] Figure 4 A flowchart illustrating the process of obtaining the benchmark to be compared, as provided in the fourth embodiment of the present invention;

[0054] Figure 5 A flowchart illustrating the process of determining off-peak hours provided in the fifth embodiment of the present invention;

[0055] Figure 6 This is a flowchart of the probability acquisition process provided in the sixth embodiment of the present invention. Detailed Implementation

[0056] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the service area remaining parking space prediction and parking guidance method based on time-series analysis proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0058] The specific scheme of the service area remaining parking space prediction and parking guidance method based on time series analysis provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0059] Please see Figure 1 The diagram illustrates a flowchart of a service area remaining parking space prediction and parking guidance method based on time-series analysis provided in the first embodiment of the present invention. The method includes:

[0060] S101. Obtain the vehicle model of any vehicle currently waiting to stop at the service area entrance.

[0061] Images of vehicles waiting to stop are captured by cameras installed at the entrance of the service area. The vehicle models identified from the images are existing technologies and will not be described in detail here.

[0062] S102. Obtain the comprehensive stability coefficient based on the vehicle model and any parking space type.

[0063] The process of obtaining the comprehensive stability coefficient will be described in detail in the second embodiment, and will not be repeated here.

[0064] S103. Obtain the highway traffic flow impact coefficient based on the number of vehicles on the highway that meet the parking space type and parking space vacancy time and the corresponding number of vacant parking spaces.

[0065] The process of obtaining the highway traffic flow impact coefficient will be described in detail in the third embodiment, and will not be repeated here.

[0066] S104. Obtain the comparison benchmark based on the historical time period, traffic density, and dwell time corresponding to the current moment, and obtain the deviation of the traffic dwell pattern based on the comparison benchmark and the preset off-peak time benchmark.

[0067] The process of obtaining the benchmark to be compared will be described in detail in the fourth embodiment, and will not be repeated here.

[0068] The method further includes: determining the off-peak period before obtaining the off-peak period benchmark. The process of obtaining the off-peak period will be described in detail in the fifth embodiment, and will not be repeated here.

[0069] The process of obtaining the off-peak time period benchmark is the same as the process of obtaining the benchmark to be compared, and the historical time period in the benchmark to be compared is equivalent to the off-peak time period in the process of obtaining the off-peak time period benchmark.

[0070] Furthermore, the process of obtaining the deviation of the traffic flow dwell pattern includes:

[0071] The difference between the benchmark to be compared and the off-peak time benchmark is calculated, and the deviation of the traffic flow dwell pattern is used to measure the difference.

[0072] The deviation of the traffic flow dwell pattern can be expressed by the formula:

[0073] ;

[0074] Among them, the Indicates time At that time, the vehicle model k and the parking space y correspond to the comparison benchmark, the The off-peak time period reference corresponding to the vehicle type k and the parking space y, the This represents the relative entropy function, used to calculate the difference between the two. Indicates time At that time, the deviation of the traffic flow dwell pattern corresponding to the vehicle type k and the parking space y is measured.

[0075] S105. Based on the comprehensive stability coefficient, the highway traffic flow influence coefficient, and the traffic flow dwelling pattern deviation, obtain the probability that the vehicle waiting to stop occupies an empty parking space of the parking space type, and guide the vehicle waiting to stop based on the comparison result of the probability and a preset threshold.

[0076] The process of obtaining the probability will be described in detail in the sixth embodiment, and will not be repeated here.

[0077] When the probability is higher than the preset threshold, the vehicle waiting to park is guided to enter the service area for parking, and a prompt can be made directly through the service area entrance guidance screen: Please go to area XX, etc. When the probability is lower than the preset threshold, the vehicle waiting to park is guided to wait or leave, and a prompt can be made directly through the service area entrance guidance screen: Parking spaces are limited, please wait patiently, etc.

[0078] The preset threshold can be set independently, preferably 0.8.

[0079] Figure 2 The flowchart illustrates the process of obtaining the comprehensive stability coefficient according to the second embodiment of the present invention. The process of obtaining the comprehensive stability coefficient includes:

[0080] S201. Calculate the size matching coefficient based on the vehicle model and the parking space type. During a preset historical period, when a vehicle of the vehicle model is parked in a parking space of the parking space type, calculate the function utilization rate based on the duration of the vehicle's use of the parking space function and the total duration of its stay.

[0081] For any vehicle type k and any parking space type y, the core of stability lies in achieving a long-term, orderly, and predictable occupancy state. In service area scenarios, there are various parking space types, and each type typically corresponds to a suitable vehicle type. If incorrect occupancy occurs (e.g., a car occupies a truck space, or a gasoline car occupies a charging space), even if the vehicle can park, it will result in a waste of the parking space's functionality (e.g., a charging space is occupied by a gasoline car, preventing electric vehicles that need charging from using it).

[0082] Based on the above analysis, the compatibility between parking spaces and vehicle models needs to consider the following two dimensions, with equal weight:

[0083] 1. Size matching: Whether the vehicle size is suitable for the parking space. For example, parking a truck in a small car space is a size mismatch.

[0084] 2. Functional utilization rate: Does the parking space function have actual utility for the vehicle type? For example, if a gasoline vehicle occupies a charging space, it is a waste of function.

[0085] Only when both size matching and effective functional utilization are satisfied can it be considered a true fit, thereby supporting the realization of overall stability.

[0086] The size matching degree can be assessed by calculating the relative difference between the vehicle size and the parking space size. It should be noted that the size here refers to the area, and the specific formula is as follows:

[0087] ;

[0088] Among them, the This indicates the degree of size matching between the vehicle of model k and the parking space y. This refers to the vehicle area corresponding to the vehicle model k. This refers to the area of ​​the parking space y. This refers to the absolute value function, the... The closer the value is to 1, the higher the size matching degree, and the better the parking space is matched with the vehicle model. norm represents linear normalization.

[0089] The area of ​​the vehicle corresponding to model k and the area of ​​the parking space y are both known data.

[0090] The function utilization rate can be defined as the ratio of the effective duration of a vehicle using a specific function of a parking space to its total dwell time. The specific formula is as follows:

[0091] ;

[0092] Among them, the The utilization rate of parking space y for vehicle model k is described in the above description. This indicates the duration of time a vehicle of model k stays in parking space y using a specific function. The total dwell time of vehicle model k in parking space y represents the total time a vehicle spends in parking space y. It quantifies the degree to which parking space functions are effectively utilized, with norm representing linear normalization.

[0093] S202. Obtain the appropriate rationality coefficient based on the size matching coefficient and the functional utilization rate.

[0094] The appropriate rationality coefficient can be expressed by the formula:

[0095] ;

[0096] Among them, the The fitting rationality coefficient represents the relationship between vehicle type k and parking space y. The utilization rate of parking space y for vehicle model k is described in the above description. This indicates the degree of size matching between the vehicle of model k and the parking space y.

[0097] S203. Within the preset historical time period, obtain the basic dwell stability based on the dwell time of the vehicle of the vehicle type in the parking space of the parking space type.

[0098] If the compatibility between vehicle type and parking space is low, even small fluctuations in dwell time may only indicate temporary occupancy and cannot represent a long-term stable state. Only under the premise of high compatibility can small fluctuations in dwell time signify the formation of a predictable stable pattern. Therefore, the basic dwell time stability index is introduced, which is measured by the ratio of the standard deviation to the mean of the dwell time of vehicle type k in parking space y (i.e., the coefficient of variation). The smaller this value, the higher the stability.

[0099] The basic stability of the residence can be expressed by the formula: , wherein The standard deviation of the time that a vehicle of model k stays in parking space y is represented. The mean time that vehicles of model k stay in parking space y is represented by norm, which indicates normalization.

[0100] S204. The appropriate rationality coefficient is used to correct the basic dwell stability to obtain the dwell regularity coefficient.

[0101] The process of obtaining the dwell regularity coefficient includes:

[0102] The relative stability is obtained based on the aforementioned basic dwell stability.

[0103] The dwell regularity coefficient is quantified by the product of the relative stability and the fit rationality coefficient.

[0104] The relative stability can be expressed by the formula: .

[0105] The meaning of relative stability is the degree of consistency of the dwell time fluctuating around its mean. When the value is close to 1, it indicates that the standard deviation is much smaller than the mean, the dwell time is very stable, the volatility is very low, and the predictability is very high. When the value is close to 0 or negative, it indicates that the standard deviation is close to or greater than the mean, the dwell time is extremely unstable, the fluctuation is huge, and it is almost unpredictable.

[0106] The dwell regularity coefficient can be expressed by the formula:

[0107] ;

[0108] Among them, the The coefficient representing the regularity of vehicle model k's stay in parking space y.

[0109] The dwell time regularity coefficient expresses the effective regularity under the premise of correct adaptation. True regularity must meet two conditions simultaneously: first, "stable dwell time" (relative stability), and second, "correct parking behavior" (appropriate adaptation). It is a comprehensive evaluation index that combines "behavioral quality" and "time regularity".

[0110] S205. Within the preset historical time period, obtain the mixed parking necessity coefficient based on the mixed parking information of vehicles of the vehicle type in the parking space of the parking space type.

[0111] Furthermore, the process of obtaining the mixed-stop necessity coefficient includes:

[0112] The mixed-stop information includes: the number of random mixed-stop records and the total number of mixed-stop records;

[0113] The necessity coefficient for random parking is obtained based on the proportion of the number of random parking records to the total number of random parking records.

[0114] The impact of mixed parking behavior on stability can be divided into two categories based on its necessity: rigid mixed parking and random mixed parking. Rigid mixed parking refers to the unavoidable choice when the target parking space is full (such as a car parking in a truck parking space when the car parking space is full). This type of behavior has a relatively small impact on overall stability. Random mixed parking, on the other hand, is the disorderly occupation that occurs even when the target parking space is available (such as a car parking in a truck parking space when the car parking space is empty). This will directly disrupt the stable order of the system.

[0115] To quantify the degree of disorder in mixed parking behavior, a mixed parking necessity coefficient can be introduced. This coefficient is measured by calculating the proportion of random mixed parking records of vehicle type k in parking space y to the total number of mixed parking records, i.e. , wherein This represents the number of random mixed parking records of vehicle type k in parking space y. This represents the total number of mixed parking records for vehicles of model k in parking space y. The higher the value, the less necessary the mixed parking behavior is, and the greater the negative impact on stability.

[0116] The necessity coefficient for mixed parking can be expressed by the formula: , wherein The mixed parking necessity coefficient represents the parking space y of vehicle model k. It is a contrarian indicator: it does not measure how good the mixed-stop behavior itself is, but rather how necessary the behavior is and how little it is disruptive to the system.

[0117] S206. Obtain the comprehensive stability coefficient based on the mixed-stop necessity coefficient and the dwell regularity coefficient.

[0118] Furthermore, obtaining the comprehensive stability coefficient based on the mixed-stop necessity coefficient and the dwell regularity coefficient includes:

[0119] The mixed-stop necessity coefficient and the dwell regularity coefficient are used to characterize the actual disturbance stability;

[0120] The dwell regularity coefficient is used to characterize ideal stability;

[0121] The comprehensive stability coefficient is obtained based on the arithmetic mean of the actual disturbance stability and the ideal stability.

[0122] Random mixed parking significantly weakens the stability characterized by the parking regularity coefficient, while the impact of essential mixed parking is relatively small. To quantify this interference effect, an inverse factor of the mixed parking necessity coefficient is introduced. The dwell regularity coefficient is corrected to construct a comprehensive stability coefficient, which can be expressed by the formula:

[0123] ;

[0124] Among them, the The coefficient representing the regularity of vehicle model k's stay in parking space y is... The mixed parking necessity coefficient of vehicle model k in parking space y represents the requirement for mixed parking. The overall stability coefficient of vehicle model k at parking space y represents the overall stability coefficient of the vehicle. This is the "unnecessary factor for mixed parking," which directly quantifies the degree of negative impact of random mixed parking. This represents the ideal stability under conditions of no stoppage interference. The overall stability coefficient represents the actual stability after the disturbance of mixed stopping. It is the arithmetic mean of the ideal stability and the actual stability, which realizes the balance between theoretical expectation and reality.

[0125] The comprehensive stability coefficient, as a quantitative comprehensive indicator, is used to predict future trends from the current state. It indicates that the future state of a parking space is not randomly generated, but rather a continuation of the current comprehensive stability state. Therefore, based on the real-time calculated comprehensive stability coefficient, the future occupancy probability of a parking space can be efficiently and accurately predicted, providing the prediction system with a unified and reliable core basis and avoiding the predicament of multiple indicators causing confusion.

[0126] Figure 3 This is a flowchart illustrating the process of obtaining the highway traffic flow impact coefficient according to the third embodiment of the present invention. The process of obtaining the highway traffic flow impact coefficient includes:

[0127] S301. The parking space competition intensity is obtained based on the ratio of the number of vehicles on the current highway that meet the parking space type and parking space vacancy time to the corresponding number of available parking spaces.

[0128] Vehicles on the highway that are about to arrive at a service area and match a parking space y (e.g., a car for a car space) will directly compete for an available parking space. Only when the estimated arrival time of a vehicle is less than or equal to the next available time of the parking space, and the vehicle type matches, is it considered part of the traffic flow competition. The intensity of this competition is determined by the ratio of the number of vehicles competing for parking spaces to the number of available spaces.

[0129] The ratio can be expressed by the formula: , wherein Indicates time At that time, the number of vehicles on the current highway heading towards the service area that meet the parking space type k and the parking space vacancy time (the next vacancy time of parking space y), the Indicates time At that time, the number of available empty parking spaces y for car model k.

[0130] in, G represents the total number of vacant parking spaces y. The comprehensive stability coefficient represents the number of parking spaces y for vehicles of model k. The simple total number of vacant parking spaces often includes a large number of spaces that are ineffective for model k due to functional or dimensional mismatches. Therefore, a comprehensive stability coefficient is introduced as an "effective utilization rate" to correct for these inefficiencies. The calculation result characterizes the number of suitable parking spaces from the perspective of model k that can be used by it and meet the conditions for long-term stable parking.

[0131] S302. The road traffic flow impact coefficient is obtained by limiting the intensity of competition for parking spaces within a reasonable probability range.

[0132] The highway traffic flow impact coefficient can be expressed by the formula:

[0133] ;

[0134] Among them, the highway traffic flow impact coefficient Used to characterize time At that time, the degree of impact of highway traffic flow on the competition for parking spaces y suitable for vehicle type k. This will affect the upper limit constraint to 100% (i.e., at most all of them will be occupied).

[0135] After analysis, the stated Generally, it is a non-negative number and cannot be negative, but the stated Indicates time When the number of available vacant parking spaces y for vehicle type k is 0, the following situation applies: When the denominator is 0, the Unable to calculate, can be set as described It is a specific value, such as 0.

[0136] Figure 4 The flowchart below shows the process of obtaining the benchmark to be compared, as provided in the fourth embodiment of the present invention. The process of obtaining the benchmark to be compared includes:

[0137] S401. Obtain vehicle information for the historical time period corresponding to the current moment, use the traffic density and the dwell time as interval features, and record the number of vehicles that meet the interval features as interval record data.

[0138] The historical time period corresponding to the current moment can be the same time period of the same day in the previous year. It should be noted that historical days are divided into at least two historical time periods according to a preset time length.

[0139] The vehicle information includes the traffic density and the dwell time.

[0140] Within the historical time period, the traffic density and the dwell time are used as interval features to record the number of vehicles that meet the interval features.

[0141] First, fixed intervals are divided based on preset thresholds: traffic density is divided into low (0-10 vehicles / hour), medium (10-30 vehicles / hour), and high (over 30 vehicles / hour); dwell time is divided into short (0-10 minutes), medium (10-30 minutes), and long (over 30 minutes). It should be noted that the preset thresholds can be set independently and are not limited here. Simply put, traffic density and dwell time are divided into high, medium, and low intervals respectively. This forms a 3×3 grid, with each cell representing a specific traffic flow-dwelling behavior pattern. Therefore, there are 9 traffic flow-dwelling behavior patterns: high traffic density - long dwell time, high traffic density - medium dwell time, high traffic density - short dwell time, medium traffic density - long dwell time, medium traffic density - medium dwell time, medium traffic density - short dwell time, low traffic density - long dwell time, low traffic density - medium dwell time, and low traffic density - short dwell time. Subsequently, vehicle information for all vehicles within the historical time period was collected, and based on the specific values ​​of their traffic density and dwell time, they were categorized into the corresponding traffic flow-dwelling behavior patterns.

[0142] For example, if 5 out of all vehicle information belong to the traffic flow density - long dwell time traffic flow behavior pattern, then the interval record data for the traffic flow density - long dwell time interval is 5.

[0143] S402. The interval probability is obtained based on the interval recorded data and the total interval recorded data, wherein the comparison benchmark is the set of interval probabilities, and the total interval recorded data is the sum of the interval recorded data.

[0144] The interval probability is obtained by dividing the interval record data by the total number of records in the interval. Since each interval corresponds to one probability, there are multiple interval probabilities, and the comparison benchmark is the set of interval probabilities.

[0145] Figure 5 The flowchart below shows the process for determining off-peak hours according to the fifth embodiment of the present invention. The process for determining off-peak hours includes:

[0146] S501. Divide the historical days corresponding to the current day into at least two time periods according to the preset time length.

[0147] The preset time length can be 30 minutes, meaning that the entire historical day is divided into 48 30-minute segments. Of course, the preset time length can be set arbitrarily and is not limited here.

[0148] S502. Cluster the time period using the traffic density, dwell time, and turnover rate as clustering features.

[0149] The clustering method used is an existing technique, such as the K-Means algorithm, which will not be elaborated upon here. The number of clusters can be three.

[0150] S503. Select the class corresponding to the one with the highest stability as the off-peak period.

[0151] Stability is assessed by calculating the average of the coefficients of variation (standard deviation ÷ mean) of the clustering features in each of the three clustered groups. The higher the average value, the lower the stability.

[0152] Specifically, for the three clusters obtained, firstly, the coefficient of variation (standard deviation ÷ mean) of the three features of traffic flow density, dwell time and turnover rate within each cluster is calculated. Then, the average of the coefficients of variation of the three features of each cluster is taken. Finally, the cluster with the smallest average value is selected from the three clusters and determined as the most stable time period category as the off-peak time period.

[0153] Figure 6 The flowchart below shows the probability acquisition process provided in the sixth embodiment of the present invention. The probability acquisition process includes:

[0154] S601. Establish the initial idle potential based on the comprehensive stability coefficient.

[0155] The initial vacancy potential characterizes the inherent tendency of vehicle type k and parking space y to remain vacant, and its value consists of two parts: basic potential and stability contribution. Specifically, it is defined as: Initial vacancy potential = Wherein, the constant 1 represents the basic idle potential, ensuring that even the comprehensive stability coefficient Even at extremely low levels, there is still a possibility that parking spaces may remain vacant; the aforementioned comprehensive stability coefficient... This quantifies the positive contribution of the stability of the vehicle type-parking space combination to improving the vacancy potential.

[0156] S602. The highway traffic flow influence coefficient and the deviation of the traffic flow dwelling pattern are used to correct the initial idle potential, and the corrected results are normalized to a non-negative unit interval to obtain the probability.

[0157] The probability can be expressed by the formula:

[0158] ;

[0159] Among them, the Indicates time At that time, the deviation of the traffic flow dwell pattern corresponding to the vehicle type k and the parking space y is... The comprehensive stability coefficient of vehicle model k at parking space y represents the overall stability coefficient of the vehicle. The highway traffic flow impact coefficient is used to characterize the impact of traffic flow over time. At that time, the degree of influence of highway traffic flow on the competition for parking spaces y suitable for vehicle type k, the aforementioned It can be any function used to restrict the value within the brackets to the range [0,1], such as the Sigmoid function.

[0160] Initial buffer ( ) is the foundation of the parking space's resistance to pressure; two types of pressure ( The pressure is greater and the remaining buffer is less. The remaining buffer is then converted into the probability of [0,1].

[0161] The present invention has the following beneficial effects:

[0162] First, obtain the vehicle model of any vehicle currently waiting to stop at the service area entrance. This vehicle model is used for subsequent analysis of the overall stability coefficient.

[0163] Secondly, a comprehensive stability coefficient is obtained based on the vehicle model and any parking space type. This comprehensive stability coefficient measures the compatibility between the vehicle model and the parking space type. The higher the comprehensive stability coefficient, the greater the compatibility.

[0164] Next, the highway traffic flow impact coefficient is obtained based on the number of vehicles on the highway heading to the service area that match the parking space type and the parking space vacancy time, and the corresponding number of available parking spaces. The highway traffic flow impact coefficient represents the degree of competition of highway traffic for available parking spaces of the specified parking space type. The larger the highway traffic flow impact coefficient, the less likely the vehicles waiting to stop are to occupy available parking spaces of the specified parking space type.

[0165] Then, a comparison benchmark is obtained based on the historical time period, traffic density, and dwell time corresponding to the current moment. The deviation of the traffic flow dwell pattern is obtained based on the comparison benchmark and the preset off-peak time benchmark. The deviation of the traffic flow dwell pattern reflects the degree of consumption of the parking space vacancy potential: the more serious the deviation, the faster the potential is consumed, and the more difficult it is for parking spaces to become vacant.

[0166] Finally, based on the comprehensive stability coefficient, the highway traffic flow impact coefficient, and the deviation of the traffic flow dwelling pattern, the probability that a vehicle waiting to stop will occupy an empty parking space of the specified parking type is obtained. The vehicle is then guided based on a comparison of this probability with a preset threshold. When the probability is higher than the preset threshold, the vehicle is guided to park in the service area, and a prompt can be displayed on the service area entrance guidance screen: "Please proceed to area XX," etc. When the probability is lower than the preset threshold, the vehicle is guided to wait or leave, and a prompt can be displayed on the service area entrance guidance screen: "Parking spaces are currently limited; please wait patiently," etc.

[0167] This invention can analyze the matching rules between vehicle type and parking space, clarify the adaptation relationship and dwell characteristics, and then incorporate the time dimension to build a prediction framework that is more in line with the real scene of the service area, so that the prediction results can accurately serve parking space allocation and vehicle guidance.

[0168] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0169] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for predicting remaining parking spaces in service areas and providing parking guidance based on time-series analysis, characterized in that, The method includes: Get the vehicle model of any vehicle currently waiting to stop at the service area entrance; The overall stability coefficient is obtained based on the vehicle model and any parking space type. The highway traffic flow impact coefficient is obtained based on the number of vehicles on the highway that meet the parking space type and parking space vacancy time and the corresponding number of available parking spaces. The comparison benchmark is obtained based on the historical time period, traffic density, and dwell time corresponding to the current moment. The deviation of the traffic dwell pattern is obtained based on the comparison benchmark and the preset off-peak time benchmark. The probability of a vehicle waiting to stop occupying an empty parking space of the specified parking space type is obtained based on the comprehensive stability coefficient, the highway traffic flow influence coefficient, and the deviation of the traffic flow dwelling pattern. The vehicle waiting to stop is then guided based on the comparison result between the probability and a preset threshold. The process of obtaining the comprehensive stability coefficient includes: Calculate the size matching coefficient based on the vehicle model and the parking space type. During a preset historical period, when a vehicle of the vehicle model is parked in a parking space of the parking space type, calculate the function utilization rate based on the duration of the vehicle's use of the parking space function and the total duration of its stay. Based on the size matching coefficient and the functional utilization rate, obtain the appropriate rationality coefficient; Within the preset historical time period, the basic dwell stability is obtained based on the dwell time of the vehicle of the vehicle type in the parking space of the parking space type. The appropriate rationality coefficient is used to correct the basic dwell stability to obtain the dwell regularity coefficient; Within the preset historical time period, the necessity coefficient of mixed parking is obtained based on the mixed parking information of vehicles of the vehicle type in parking spaces of the parking space type. The comprehensive stability coefficient is obtained based on the mixed-stop necessity coefficient and the dwell regularity coefficient.

2. The service area remaining parking space prediction and parking guidance method based on time series analysis as described in claim 1, characterized in that, The process of obtaining the dwell regularity coefficient includes: The relative stability is obtained based on the aforementioned basic dwell stability. The dwell regularity coefficient is quantified by the product of the relative stability and the fit rationality coefficient.

3. The service area remaining parking space prediction and parking guidance method based on time series analysis as described in claim 1, characterized in that, The process of obtaining the mixed parking necessity coefficient includes: The mixed-stop information includes: the number of random mixed-stop records and the total number of mixed-stop records; The necessity coefficient for random parking is obtained based on the proportion of the number of random parking records to the total number of random parking records.

4. The service area remaining parking space prediction and parking guidance method based on time series analysis as described in claim 1, characterized in that, The comprehensive stability coefficient is obtained based on the mixed-stop necessity coefficient and the stop regularity coefficient, including: The mixed-stop necessity coefficient and the dwell regularity coefficient are used to characterize the actual disturbance stability; The dwell regularity coefficient is used to characterize ideal stability; The comprehensive stability coefficient is obtained based on the arithmetic mean of the actual disturbance stability and the ideal stability.

5. The service area remaining parking space prediction and parking guidance method based on time series analysis as described in claim 1, characterized in that, The process of obtaining the highway traffic flow impact coefficient includes: The intensity of parking space competition is determined by the ratio of the number of vehicles on the highway that meet the parking space type and the parking space vacancy time to the corresponding number of available parking spaces. The highway traffic flow impact coefficient is obtained by limiting the intensity of competition for parking spaces within a reasonable probability range.

6. The service area remaining parking space prediction and parking guidance method based on time series analysis as described in claim 1, characterized in that, The process of obtaining the deviation of the traffic flow dwell pattern includes: The difference between the benchmark to be compared and the off-peak time benchmark is calculated, and the deviation of the traffic flow dwell pattern is used to measure the difference.

7. The service area remaining parking space prediction and parking guidance method based on time series analysis as described in claim 1, characterized in that, The process of obtaining the benchmark to be compared includes: Obtain vehicle information for the historical time period corresponding to the current moment, use the traffic density and the dwell time as interval features, and record the number of vehicles that meet the interval features as interval record data; The interval probability is obtained based on the interval recorded data and the total interval recorded data. The comparison benchmark is the set of interval probabilities, wherein the total interval recorded data is the sum of the interval recorded data.

8. The service area remaining parking space prediction and parking guidance method based on time series analysis as described in claim 1, characterized in that, The method further includes: determining the off-peak period before obtaining the off-peak period benchmark, wherein the process of determining the off-peak period includes: Divide the historical days corresponding to the current day into at least two time periods according to a preset time length; The traffic density, dwell time, and turnover rate are used as clustering features to cluster the time period; The class corresponding to the one with the highest stability is selected as the off-peak period.

9. The service area remaining parking space prediction and parking guidance method based on time series analysis as described in claim 1, characterized in that, The process of obtaining the probability includes: The initial idle potential is determined based on the aforementioned comprehensive stability coefficient; The highway traffic flow influence coefficient and the deviation of the traffic flow dwell pattern are used to correct the initial idle potential, and the corrected results are normalized to a non-negative unit interval to obtain the probability.

Images