JPH05120599A - Parking prediction management method and device - Google Patents

Abstract

(57) [Summary] [Purpose] To make effective use of available parking space. [Configuration] The vehicle ID identification device 3 identifies whether the entry vehicle is a registered vehicle or a general vehicle, and when the entry vehicle is a registration vehicle, the entry is immediately permitted, and when the entry vehicle is a general vehicle, the computer 1 is required for the registration vehicle at that time. Predict the number of areas, secure those areas, and allow entry if there is room.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a parking prediction management method for predicting an inflow / outflow amount of a vehicle and performing entrance control, and particularly when a registered vehicle and a general vehicle are both parked, an empty space of the registered vehicle can be saved. The present invention relates to a parking prediction management method and device that can be effectively used.

[0002]

2. Description of the Related Art In a self-propelled parking lot, management of a parking lot having a parking area for registered vehicles such as a monthly pole is divided into a parking area for registered vehicles and an area for general vehicles. Even if there is an empty space, there are many things that do not use that area. In addition, there is a case where the rental space is rented again for effective use of the vacant space in the registered vehicle area, but when it is not actually used, it is still vacant.

[0003]

At present, the problem of insufficient parking lots in urban areas is becoming more serious. Further, when the parking lot is full, a line of vehicles waiting for parking causes further traffic congestion on the road. Therefore, improving the efficiency of using parking lots is a very important issue. However, in the above-mentioned conventional technology in the self-propelled parking lot, the parking area for registered vehicles such as the monthly pole is not effectively used even during the daytime when it is not used, and general vehicles such as temporary storage are used in the parking area. There was a drawback that you could not use.

An object of the present invention is to provide a parking prediction management method which can make the most effective use of the available parking area by making the parking area free allocation.

[0005]

[Means for Solving the Problems] The above-mentioned object is to measure vehicle inflow and outflow vehicles in a parking lot, and vehicle identification means for identifying whether the inflow vehicle and outflow vehicle are registered vehicles or general vehicles. For counting the number of inflowing and outflowing vehicles, the storing means for obtaining and storing the inflowing number, the outflowing number and the number of existing vehicles of the registered vehicle and the general vehicle for each time from the outputs of the inflowing and outgoing means and the vehicle identifying means. Prediction means for predicting the amount of change in the number of occupants indicating how much the number of occupants changes after a certain time from each time point is provided, and when the vehicle enters the entrance, the vehicle identification means detects the vehicle Whether the vehicle is a registered vehicle or a general vehicle is identified, and if the vehicle is a registered vehicle, the admission is allowed immediately. If the vehicle is a general vehicle, the presence of the registered vehicle stored in the amount of change in the number of existing vehicles obtained by the prediction means is determined. Car platform Then, the estimated number of registered vehicles in the vehicle is calculated, and a value obtained by adding the predetermined number of spares and the stored number of general vehicles in the above to the calculated estimated number of registered vehicles is When the number of vehicles allowed for parking is less than the permitted number, parking is permitted by allowing general vehicles to enter.

[0006]

If the estimated information is estimated by the predicting means by grasping the in-vehicle amount of the registered vehicle and the general vehicle, the inflow / outflow amount statistical information, the external information, the in-vehicle information and the inflow / outflow information of the day, the registered vehicle It is possible to effectively use the vacant space for general vehicles while securing the necessary space for. Furthermore, if the past data is analyzed and the estimation accuracy is improved, it is possible to use the space more effectively.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which the present invention is applied to a self-propelled parking lot will be described below with reference to the drawings. FIG. 2 shows an embodiment of a predictive parking management system to which the method of the present invention is applied. The computer 1 has an admission determination function and a control function for each connected parking lot device, and data communication with each device is performed. , Admission judgment and guidance for general vehicles, and control of equipment. The storage device 2 directly connected to the computer 1 stores statistical data such as registered vehicle information, registered vehicle parking position data, past vehicle presence patterns and inflow / outflow patterns for creating prediction information. As the parking lot device, a vehicle ID identification device 3 for identifying whether the entry vehicle is a registered vehicle or a general vehicle (as an example of the current device, a vehicle number identification device using image processing, or a non-contact type ID card (Equipment combined with loop coil, etc.), entrance gate device 4 for restricting entry vehicles, guidance guide plate 8 for reliably guiding registered vehicles to the parking position in the hall, presence of vehicle for discriminating parking in the passenger compartment The sensor 9, the exit gate 5 for restricting the vehicle to leave, and the parking status guide display plate 6 for individually guiding and displaying registered vehicles and general vehicles near the parking lot before entering, the general vehicle when the parking status is full On the other hand, a waiting time guide display plate 7 for guiding the waiting time until entry is installed, and these are centrally managed by the computer 1.

FIG. 1 shows the overall operation flow of this embodiment. The management operation will be described on the premise of the device arrangement shown in FIG. First, in process 100, the vehicle ID identification device 3
Detects the vehicle that is about to enter, recognizes the ID information of the vehicle, and transmits the recognized ID information to the computer 1. The computer 1 uses the ID information transmitted in the process 101 and the storage device 2
The registered vehicle information is compared with the registered vehicle information of 1. to identify whether the entry vehicle is a registered vehicle or a general vehicle. If the identified entry vehicle is a registered vehicle, the computer 1 performs the process 1
In 02, the current on-site parking status and gate status are checked from the outputs of the in-vehicle sensor 9 and the identification device 3 to find an enterable passage, which is displayed on the guide display board 6. If the entering vehicle is a general vehicle, it is determined by the prediction function of the computer 1 described later whether or not the vehicle is full.
In 3, the entrance waiting time and the like are displayed on the guide display board 6 using the prediction information. When the result of the determination is that the vehicle is empty, or when the result of the above prediction is that the vehicle is empty while waiting for the vehicle to be full, the entrance gate guide is displayed. Then, the process 10 is performed while the general vehicle is moving toward the gate.
At 4 and 105, the number of vehicles parked in the hall is confirmed from the output of the in-vehicle sensor 9. If the number of registered vehicles required is predicted and it is finally determined in step 106 that entry is possible, the entrance gate device 4 is opened in step 107. If the entry vehicle is a registered vehicle, the entrance gate device 4 is opened in step 107 immediately after the display in step 102.

When the vehicle enters and the entrance gate 4 is closed, the computer 1 guides the entering vehicle to the parking position at step 108 because the parking position is not fixed even in the registered vehicle. FIG. 4 shows an example inside the parking lot, and the parking position is divided into several blocks, and the vehicle ID identification device 3 is arranged at the entrance and exit of each block. At this time, when the entry vehicle is a registered vehicle, the parking position desired by the registrant and the occupancy status of the block (for example, close to the entrance / exit) stored in the storage device 2 are confirmed, and the vehicle is closest to the position. Guide to position. In addition, for general vehicles, avoid parking blocks for registered vehicles as much as possible, assign parking positions, and provide guidance. This guidance is provided by the guide plate 8. Subsequently, in process 109, the number of inflowing vehicles and the number of existing vehicles are counted up. This is because when it is confirmed from the output of the vehicle ID identification device 3 that the vehicle that entered a certain block in FIG. 4 did not exit from that block and parking is confirmed by the in-vehicle sensor 9, the number of vehicles in that block Is to count up.

In the case of leaving the vehicle, processing is performed according to the flow described as when leaving the vehicle in FIG. As shown in FIG. 3, first, in process 120, the vehicle ID identification device 3 installed in front of the exit gate device 5 reads the ID information of the leaving vehicle, and in process 121, it is determined whether or not the leaving vehicle is a registered vehicle. judge. After that, in process 122, the exit gate device 5 is opened, and when the device confirms the passage, the process 123 counts down the number of in-vehicle vehicles.

Hereinafter, the method of judging entrance of a general vehicle, which is a feature of the present invention, will be described in detail. When the number of required areas for registered vehicles at the current time obtained by the prediction described later is NEi and the number of ordinary vehicles in the vehicle is UQi, the sum of these TQi = NEi + UQi is obtained as the number of parking areas. If PQ ≧ TQi is satisfied for the allowable number of parked cars PQ in the parking lot, admission is permitted. Here, the number of areas NEi required for the registered vehicle is calculated as the sum of the predicted number of registered vehicles FEi and the spare vacant area RSi required to reliably park the registered vehicle. Two methods, a statistical processing method and a fuzzy processing method, will be described as the calculation method of the number NEi.

First, an example of a method for calculating the predicted number of registered vehicles FEi using the statistical processing will be described. Calculator 1 uses the daily statistical presence pattern of registered vehicles, which is stored in storage device 2 in advance as past statistical data for each variable factor (for example, month, day, day of the week, weather, etc.), to determine the conditions for that day. Select the pattern that was there. FIG. 5 shows an example of the daily statistical vehicle presence pattern PEQ (dotted line) of the registered vehicle and the actual vehicle presence data EQ (solid line) of the day. In addition, although one statistical in-vehicle presence pattern is shown in the figure, there are generally a plurality of statistical in-vehicle presence patterns that match the conditions of the day.
When expressed as EQ, j = 1, 2, ..., First, a process of taking out one of them is performed. Therefore, the computer uses the actual vehicle data up to the current time ti of the day, and the number of vehicle vehicles EQi and the rate of change of vehicle vehicle numbers per unit time at each time point ΔEQi = (EQi-EQi-1) / (ti-ti -1) is calculated. Further, the computer 1 similarly obtains values PjEQi and ΔPjEQi = (PjEQi-PjEQi-1) / (ti-ti-1) corresponding to each statistical pattern PjEQ at each time until the current time. The deviation Eri between the actual vehicle data and each selected statistical pattern is calculated by the following formula: Eri = (EQi−PEQi) ² + (ΔEQi−ΔPEQi) ² and the computer 1 is the current time ti of Eri. The cumulative value up to is obtained for each statistical pattern, and the one that has the smallest value, that is, the one with the highest internal matching rate of each statistical pattern is selected. When the statistical pattern is determined in this way, the predicted vehicle number FEi of registered vehicles is equal to the vehicle number EQi-1 at time ti-1 which is a unit time before the current time ti of the actual vehicle data EQ.
FEi = EQi-1 + PEQi by adding the change rate ΔPEQi of the number of vehicles at the time ti-1 and the time ti of the statistical vehicle presence pattern selected earlier to FEi = EQi-1 + PEQi. In addition, the inflow / outflow prediction information necessary for creating the parking waiting time guide information by the same method is stored in the storage device 2 in advance as past data for each variation factor. It is possible to select and create a pattern that meets the conditions of the day from the statistical inflow pattern and the statistical outflow pattern of. And the number of required areas NEi
Is calculated by adding the spare empty area RSi at each time to this FEi. Further, these pieces of prediction information can be used to perform prediction before the current time by expanding the value for time ti-1 to the value for ti and the value for time ti to the value for ti + 1. ..

Next, an example of calculation of prediction information using fuzzy theory will be described. 6 and 7 are vehicle IDs
An example of the actual inflow amount EI and the actual outflow amount EO of the registered vehicle up to the present time ti counted by the identification device 3 and the entrance gate device 4 is shown. The computer 1 calculates the actual inflow change rate ΔEIi and the actual outflow change rate ΔEOi at the current time ti from the current time ti and the inflow and outflow rates per unit time at the time ti-1 which is a unit time before the current time. ΔEIi = (EIi−EIi−1) / (ti−ti−1) ΔEOi = (EOi−EOi−1) / (ti−ti−1). 8 and 9 show examples of patterns of the inflow amount PI and the outflow amount PO of the day stored in the storage device 2 in advance as past statistical data. Calculator 1
Is the statistical inflow change rate ΔPIi and the statistical outflow change rate ΔPOi per unit time of the time ti predicted from the time ti−1 from the previously stored daily inflow and outflow patterns. [Delta] PIi = (PIi-PIi-1) / (ti-ti-1) [Delta] EOi = (POi-POi-1) / (ti-ti-1) Next, the computer 1 has determined as above. Predicted inflow rate evaluation set U using ΔEIi, ΔPIi, ΔEOi and ΔPOi as input and using a membership function that has been qualitatively evaluated as a fuzzy amount in advance.
(ΔFEIi) and predicted outflow quantity evaluation set U (ΔFEOi) are output.

10 and 11 show examples of membership functions, and FIG. 10 shows membership functions with respect to the actual inflow change rate ΔEi. Increase, constant,
Gives a qualitative assessment of reduction. A value of 0.0 to 1.0 is defined as the degree of qualitative evaluation on the vertical axis. The horizontal axis represents the actual rate of change of inflow ΔEIi, which is a signed value in this example, where the increase is positive (+) and the decrease is negative (-). As an example,
If ΔEIi = 7 (units / hour), the suitability for each evaluation is as follows. V + (ΔEIi) = 0.3 V0 (ΔEIi) = 0.63 V- (ΔEIi) = 0.0 However, V is a function indicating the fitness for each evaluation. This evaluation is qualitatively expressed as "the actual inflow change rate ΔEIi is 0.3 in the forecast time zone Ti = ti-ti-1 and 0.63 in the constant degree and 0.0 in the decreasing degree". The evaluation set is expressed as follows: U (ΔEIi) = (V + (ΔEIi), V0 (ΔEIi), V- (ΔEIi)) = (0.3, 0.63, 0.0) For ΔEOi, ΔPIi and ΔPOi Also obtains a qualitative evaluation set using each membership function. 11
Is an example of the membership function for the statistical inflow change rate ΔPIi.

Next, the estimated inflow amount ΔFEIi is inferred and output using these evaluation sets. This inference is based on FIG. 12 and FIG.
The inference rule is based on the a priori information shown in 3.
FIG. 12 is a rule definition table for inferring the predicted inflow amount ΔFEIi from the actual inflow change rate ΔEIi and the statistical inflow change rate ΔPIi. The rule definition table is as follows. (Rule 1) if ΔEIi> 0 and ΔPIi> 0 then ΔFEIi> 0 (Rule 2) if ΔEIi> 0 and ΔPIi = 0 then ΔFEIi> 0 (Rule 3) if ΔEIi> 0 and ΔPIi <0 then △ FEIi = 0 (Rule 4) if ΔEIi = 0 and △ PIi> 0 then △ FEIi> 0 (Rule 5) if ΔEIi = 0 and △ PIi = 0 then △ FEIi = 0 (Rule 6) if ΔEIi = 0 and △ PIi <0 then ΔFEIi <0 (rule 7) if ΔEIi <0 and ΔPIi> 0 then ΔFEIi = 0 (rule 8) if ΔEIi <0 and ΔPIi = 0 then ΔFEIi <0 (rule 9) if ΔEIi <0 and ΔPIi <0 then ΔFEIi <0 Further, FIG. 13 shows the actual outflow change rate ΔEOi and the statistical outflow change rate ΔP.
It is a rule definition table for inferring a predicted outflow from Oi. Based on these rules, the estimated inflow evaluation set U (△ FEI
i) is obtained as follows. As an example, △ EIi = 7
(Unit / hour) and ΔPIi = 5 (unit / hour), FIGS. 8 and 9
The evaluation set is as follows: U (△ EIi) = (0.3,0.63,0.0) U (△ PIi) = (0.22,0.6,0.0) Using this value, the predicted inflow is inferred as follows. It Based on FIG. 12, the rule 1 is V1 + (ΔEIi + ΔPIi) = min (V + (ΔEIi), V + (ΔPIi)) = min (0.3,0.22) = 0.22 where V1 + (ΔEIi + Δ PIi) is a function to which the inference result of rule 1 is given. Below, the same calculation will be made for other rules as well. V2 + (ΔEIi + ΔPIi) = min (V + (ΔEIi), V0 (ΔPIi)) = 0.3 V30 (ΔEIi + ΔPIi) = min (V + (ΔEIi), V- (ΔPIi)) = 0.0 V4 + (ΔEIi + ΔPIi) = min (V0 (ΔEIi), V + (ΔPIi)) = 0.22 V50 (ΔEIi + ΔPIi) = min (V0 (ΔEIi), V0 (ΔPIi)) = 0.6 V6- (ΔEIi + ΔPIi) = min (V0 (ΔEIi), V- (ΔPIi)) = 0.0 V70 (ΔEIi + ΔPIi) = min (V- (ΔEIi), V + (ΔPIi)) = 0.0 V8- (△ EIi + △ PIi) = min (V- (△ EIi), V0 (△ PIi)) = 0.0 V9- (△ EIi + △ PIi) = min (V- (△ EIi), V- (△ PIi)) = 0.0 In this way, the actual change rate of inflow △ EIi and statistical change rate of inflow △ P
Use the minimum value of the relationship according to Ii's rule of thumb. Then, when all the rules are executed, three types of fitness candidates related to ΔFEIi, that is, increase, constant, and decrease are calculated. The maximum value is set as the candidate of the predicted inflow amount from each of these suitability candidates. That is, V + (ΔFEIi) = max (Vj + (ΔEIi + ΔPIi)) = max (V1 + (ΔEIi + ΔPIi), V2 + (ΔEIi + ΔPIi), V4 + (ΔEIi + ΔPIi)) = 0.3 V0 (ΔFEIi) = max (V30, V50, V70) = 0.6 V- (ΔFEIi) = max (V6-, V8-, V9-) = 0.0 Therefore, U (ΔFEIi) = (0.3,0.6,0.0) ).

The final predicted inflow ΔFEIi is determined by the following method with the evaluation set U (ΔFEIi) calculated by the above calculation as an input. FIG. 12 is a membership function of the inflow change rate created based on empirical knowledge. If U (△ FEIi) ＝ (0.3,0.6,0.0),
The graph is cut as shown in FIG. 14 by using each value of these evaluation sets as an evaluation value corresponding to each evaluation of increase, constant, and decrease. In FIG. 14, the shaded portion shows the evaluation effectiveness, and the center of gravity is calculated from this shaded portion, and the value projected from the center of gravity to the inflow change rate axis is used as the predicted inflow amount. In this example, it is 8 (units / hour), and the value obtained by adding 8 units to the current inflow amount is the inflow unit predicted to inflow per unit time. The predicted outflow amount ΔFEOi is also calculated by performing the series of fuzzy processing as described above.

Based on these predicted inflow and outflow, the estimated number of registered vehicles FEi (= EOi + ΔFEIi−ΔF
EOi, EOi = number of registered vehicles at present) is calculated,
The required area number NEi of the registered vehicle can be determined by adding the calculated predicted number of registered vehicles and the spare space RSi preset for each time.

The above is the details of the fuzzy prediction method.
Next, a learning method for minimizing the required number of registered vehicle areas NEi will be described. Calculator 1 regularly (1
Weekly / monthly), the daily time-series actual number of registered vehicles PjEO stored in the storage device 2 as statistical data,
Predicted vehicle number FE obtained at each point in the past as described above
i and the number of required areas NEi are stored as time-series data. Based on the time-series predicted number of vehicles in use PjFE and the number of time-series required areas PjNE, learning is performed for each variable factor (month, day, weather, etc.). FIG. 15 is an example of the time-series actual vehicle in-vehicle number PjEO, the time-series predicted vehicle in-vehicle number PjFE, and the time-series required area number PjNE.
Next, Calculator 1 obtains two deviations D1i = PEOi−PFEi D2i = PNEi−PEOi from the time-series actual number of vehicles in-vehicle PjEO, the time-series predicted number of vehicles in-vehicle PjFE, and the number of time-series required areas PjNE, and calculates these deviations. The spare space RSi is calculated by fuzzy processing as an input. Here, PFEi and PNEi are values at each time ti of the time-series predicted vehicle number PjFE and the time-series required area number PjNE. 16-18 show the membership functions for the deviations D1i and D2i and the spare space RSi for use in the fuzzy processing. The qualitative rules for these are shown in FIG. This FIG. 19 is written as the following: (Rule 1) if D1i> 0 and D2i = Large then RSi decreases (Rule 2) if D1i> 0 and D2i = Medium then RSi is 0 (Rule 3) if D1i > 0 and D2i = small then RSi increases (rule 4) if D1i = 0 and D2i = large then RSi is 0 (rule 5) if D1i = 0 and D2i = medium then RSi is 0 (rule 6) if D1i = 0 and D2i = small then RSi increases (rule 7) if D1i <0 and D2i = large then RSi decreases (rule 8) if D1i <0 and D2i = medium then RSi decreases (rule 9) if D1i <0 and D2i = Small then RSi increases Therefore, deviations D1i and D from the membership functions of Figs.
The evaluation sets U (D1i) and U (D2i) of 2i are obtained, fuzzy processing is performed on the membership function of FIG. 18 in the same manner as the method of predicting the inflow / outflow amount performed in FIG. 14, and the spare space RSi To calculate. Further, the spare space RSi calculated here is compared with the RSi previously stored for each variation factor, and if the deviation is within a predetermined range, the average is updated. If it is outside the range, it is stored in the storage device 2 as another pattern.

Finally, a guide display method using the outflow amount prediction information will be described below. FIG. 20 shows an example of the waiting time guidance display method by the waiting time guidance display plates 71 to 7n using the prediction information. The waiting time guide display plates 71 to 7n are installed in a general vehicle at an appropriate distance from the entrance gate, and display the waiting time to the general vehicle when the parking lot is full. Calculator 1 calculates the predicted outflow amount ΔFEOi by inputting the outflow amount information and the statistical information of the actually registered vehicle and the general vehicle by the same predictive information calculation process as described above. The waiting time guide display plate 71 installed near the entrance gate displays the entrance waiting time ITi = 1 / ΔFEOi (hour / vehicle) which is the reciprocal of the predicted outflow ΔFEOi (vehicle / hour) calculated by the computer 1. Admission waiting time WTi
Converted to and displayed. Here, the rate of change in the number of units per unit time is assumed to change uniformly. For the waiting time guide display plate 72 next to the entrance gate 71, the number n of parkable parking spaces between the guide plates 71 and 72 is used, and the waiting time displayed on the waiting time guide display plate 71 at the entrance gate is multiplied by n. nWTi (minutes)
Is displayed. The same applies to the waiting time guide plates 73 to 7n.

In order to effectively carry out the present invention, it is preferable to use different entrances for both registered users and general users, as shown in FIG. Further, in the above-mentioned embodiment, the parking position of the registered vehicle is free, but if a certain parking block is a dedicated block for both registrants, the guiding guidance can be omitted when parking in this block.

[0021]

According to the present invention, by predicting the inflow amount and the outflow amount of the registered vehicle, it is possible to open the vacant area of the registered vehicle to a general vehicle and improve the utilization efficiency of the parking area. In addition, the learning function can minimize the amount of space required for the registered vehicle. Furthermore, by predicting the outflow amount, it becomes possible to display the waiting time guidance for general vehicles, and it is possible to reduce the distraction of driving due to anxiety and discomfort due to uncertain waiting conditions in the parking lot.

[Brief description of drawings]

FIG. 1 is a flow chart showing an embodiment of the method of the present invention.

FIG. 2 is a diagram showing an embodiment of a predictive parking management system using the method of the present invention.

FIG. 3 is an explanatory diagram near a parking lot entrance.

FIG. 4 is a diagram showing an example of a parking position in a parking lot.

FIG. 5 is a diagram showing an example of statistical vehicle presence patterns and actual vehicle presence data.

FIG. 6 is a diagram showing an example of an actual inflow amount.

FIG. 7 is a diagram showing an example of an actual outflow amount.

FIG. 8 is a diagram showing an example of a statistical inflow pattern.

FIG. 9 is a diagram showing an example of a statistical outflow pattern.

FIG. 10 is a diagram showing a membership function with respect to the actual inflow change rate.

FIG. 11 is a diagram showing a membership function with respect to a performance outflow change rate.

FIG. 12 is a diagram showing rules for estimating a predicted inflow amount.

FIG. 13 is a diagram showing rules for estimating a predicted outflow amount.

FIG. 14 is a diagram showing a membership function with respect to a predicted inflow amount.

FIG. 15 is an example of statistical information and actual data regarding the number of vehicles in the vehicle.

FIG. 16 is a diagram showing a membership function with respect to a deviation D1i.

FIG. 17 is a diagram showing a membership function with respect to a deviation D2i.

FIG. 18 is a diagram showing a membership function for a spare space.

FIG. 19 is a diagram showing rules for estimating a spare space.

FIG. 20 is an explanatory diagram of a waiting time guidance method.

[Explanation of symbols]

 1 Computer 2 Storage device 3 Vehicle ID identification device 4 Entrance gate device 5 Exit gate device 6 Parking status guide display plate 7 Waiting time guide display plate 8 Guidance guide plate 9 In-vehicle sensor

Claims (8)

[Claims] 1. A parking prediction management method for controlling an inflow amount of a vehicle into a parking lot accommodating a registered vehicle registered in advance and a general vehicle other than the registered vehicle, when the vehicle enters the entrance, It automatically identifies whether the vehicle is a registered vehicle or a regular vehicle, and if the vehicle is a registered vehicle, it is automatically admitted immediately. If it is a regular vehicle, the number of registered vehicles remains constant for a certain period of time. Calculate the predicted number of registered vehicles by adding the number of registered vehicles to the estimated amount of change in the number of registered vehicles, which indicates how much it will change later. When the value added to the predicted number of registered vehicles calculated above is smaller than the allowable parking number of the parking lot, the general vehicle is allowed to enter the parking prediction management method. 2. A parking prediction management method for controlling a vehicle inflow amount to a parking lot accommodating a registered vehicle registered in advance and a general vehicle other than the registered vehicle, wherein an inflow vehicle and an outflow vehicle of the parking lot are registered. Vehicle identifying means for identifying a vehicle or a general vehicle, inflow / outflow number measuring means for measuring the number of inflows and outflows of a parking lot, and registration for each time from the output of the means and the vehicle identification means Storage means for obtaining and storing the number of inflows, outflows, and in-vehicle numbers of vehicles and general vehicles, and predicting the amount of change in the in-vehicle number that indicates how much the in-vehicle number of registered vehicles will change after a certain time from each time point When a vehicle enters the entrance, the vehicle identifying means identifies whether the vehicle is a registered vehicle or a general vehicle, and when the vehicle is a registered vehicle, the vehicle is allowed to enter immediately. vehicle When it is, the predicted vehicle number of registered vehicles is calculated by adding the stored vehicle number of registered vehicles to the variation amount of vehicle number obtained by the prediction means,
Further, if the value obtained by adding the predetermined number of spares and the stored number of existing general vehicles to the predicted number of registered vehicles calculated above is smaller than the allowable parking number of the parking lot, the general vehicle is not allowed to enter. A parking prediction management method characterized by permitting. 3. A parking prediction management device for controlling the inflow amount of vehicles into a parking lot accommodating pre-registered registered vehicles and general vehicles other than the registered vehicles, the inflowing vehicles and outflowing vehicles of the parking lot being registered. Vehicle identifying means for identifying a vehicle or a general vehicle, inflow / outflow number measuring means for measuring the number of inflows and outflows of a parking lot, and registration for each time from the output of the means and the vehicle identification means Storage means for obtaining and storing the number of inflows, outflows, and in-vehicle numbers of vehicles and general vehicles, and predicting the amount of change in the in-vehicle number that indicates how much the in-vehicle number of registered vehicles will change after a certain time from each time point Predicting means for doing so, and when the vehicle enters the entrance, the identification result from the vehicle identifying means as to whether the vehicle is a registered vehicle or a general vehicle is fetched, and when the vehicle is a registered vehicle, entry is allowed immediately. , General When the vehicle is a vehicle, the predicted vehicle number of registered vehicles is calculated by adding the stored vehicle number of registered vehicles to the variation amount of vehicle number obtained by the predicting means, and further, a predetermined spare number and When the value obtained by adding the stored number of stored ordinary vehicles to the calculated estimated number of registered vehicles is less than the allowable parking number for the parking lot, the vehicle is allowed to enter the vehicle. Parking prediction management device characterized by. 4. The actual vehicle presence pattern for each variation factor is obtained by dividing the number of registered vehicles in the storage unit stored in the storage unit for each variation factor including part or all of the day of the week, the month, and the weather. 4. The amount of change in the number of vehicles in use, which is stored in the storage means and is calculated by the prediction means, using the actual vehicle presence pattern according to the variation factor of the day on which the prediction is made. Parking prediction management device. 5. The predicting means calculates a statistical inflow change rate and a statistical outflow change rate from the past inflowing vehicle numbers and outflowing vehicle numbers of registered vehicles stored in the storage means, and the inflowing vehicle number and outflowing numbers of registered vehicles on the day. Calculate the actual rate of change of the number of inflow vehicles and the rate of change of the actual number of outflow vehicles at that time from the number of vehicles, and make a prediction using the membership function and the inference rule that are predetermined for the statistical rate of change of the inflow vehicle number and the rate of change of the actual inflow vehicle number. Calculate the inflow amount, calculate the estimated outflow amount using the membership function and the inference rule that are predetermined for the statistical outflow number change rate and the actual outflow number change rate, and then calculate the estimated outflow amount from the estimated inflow rate. 4. The parking prediction management device according to claim 3, wherein the amount of change in the number of registered vehicles is calculated by subtracting the amount. 6. The estimated number of vehicles in use, which is obtained by adding the number of registered vehicles at the time point, to the change amount of the number of registered vehicles obtained by the predicting means, and the required number of areas by adding the number of spares to the number. Is obtained for each time and stored in the storage means, and the predicted number of vehicles is subtracted from the number of registered vehicles at each time in the past at a predetermined cycle.
Difference and the required area, minus the number of existing vehicles
Is calculated, and the vacancy of the parking lot is always minimized while securing the spare space for the registered vehicle by using the membership function and the inference rule determined in advance for the first and second differences. The parking prediction management device according to claim 3, wherein the number of spares is determined. 7. A part or all of the parking lot is used as a common area for general vehicles and registered vehicles, and an in-vehicle sensor is provided at each parking position, and an empty parking position is checked using the sensor to permit entry. The parking prediction management device according to claim 3, wherein the guide to the vacant parking position is performed on the vehicle using a guide plate. 8. The parking prediction management according to claim 5, wherein the predicted waiting time for a vehicle waiting to enter is displayed according to the calculated predicted outflow number and the number of vehicles waiting to enter the entrance gate. apparatus.