JPH0423317B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Industrial Application] This invention is a bus operation system that collects transit information based on basic information of route bus operation plans, predicts bus arrival times, and displays operation status. This relates to a management device.

[Conventional technology]

FIG. 4 is a block connection diagram of a conventional specific motor vehicle operation control device disclosed in, for example, Japanese Patent Publication No. 54-11878. In the figure, 41 is a central operation control device, 42a, 42b, 42c are ground receivers, 4
3a, 43b, 43c are the ground receivers 42a,
42b, 42c antenna, 44a, 44b, 4
4c is a central operation control device 41 and a ground receiver 42
a, 42b, 42c, a line connecting 45a,
45b, 45c are cars, 46a, 46b, 46
47a, 47b, 47c are on-vehicle antennas for the on-vehicle radio transmitters 46a, 46b, 46c, 48 is the direction of travel, and 49 is the vehicle 45a, 45.
b, 45c travel route.

Next, the operation will be explained. Since the ground receivers 42a, 42b, 42c are arranged along the travel path 49, the ground receivers 42a, 42b,
The antennas 43a, 43b, 43c of 42c catch the radio waves transmitted from the cars 45a, 45b, 45c, and the ground receivers 42a, 42b, 42c send a line to the central operation control unit 41 to notify that the car has passed in front of the ground receiver. 44a, 44b, 44c
Transmit via. For example, when the car 45a passes the installation position of the antenna 43a, the ground receiver 42a sends passing information of the car 45a to the central operation control device 41. This central operation control device 41 stores the average traveling time (the average value of the time required for the automobile to travel once on a predetermined route) and the average traveling speed of the automobile 45a on the traveling route 49, and receives information from the ground Based on the passage information from the aircraft 42a, the estimated arrival time of the automobile 45a at the ground receiver 42b is calculated as follows, taking into consideration the time at the time of the passage, average travel time, and variations in travel speed.

Vehicle passing time at ground receiver 42a + ground receiver 42b
It is calculated as distance 42a/past average traveling speed of cars on that route...(400).

Similarly, the estimated time of arrival at the next receiver 42c is calculated, making it possible to track the arrival time of the vehicle on its route.

[Problem that the invention seeks to solve]

Conventional specific vehicle operation control devices are configured as described above, so when calculating the required time for a unit section, the past average traveling speed of the vehicle on that route (one lap on that route) is used. By doing so, if there is traffic congestion or a congested section during a certain time period, the average traveling speed of the automobile 45a on that route will not necessarily be close to the traveling speed of that unit section, and the following required It is a relationship between time and minutes.

Distance between ground receivers 42b and 42a/Past average traveling speed of the vehicle on that route ≠ Distance between ground receivers 42b and 42a/travel speed for the unit section Therefore, there is a large error from the actual travel time. It was bound to occur. Furthermore, even when using the past average traveling speed of the cars 45a, 45b, and 45c in that section, the above-mentioned error also depends on how far back in time the required travel time (hereinafter referred to as actual value) is used. On the other hand, no matter what the service interval of a route is, or even if there are variations in travel speed between sections, it is necessary to predict the estimated arrival time and display the operation status of the unit section quickly and accurately. However, due to the above-mentioned ambiguities, the conventional prediction calculation formula 400 is not a general-purpose prediction calculation formula that can be applied under any conditions, and is therefore not practical, and prediction accuracy cannot be expected. There was a problem.

This invention was made to solve the above-mentioned problems, and uses a central processing unit to store the planned travel time required for a unit section on the route of a route bus and the actual value of the vehicles that have already traveled. By applying a series of predictive calculation formulas based on these planned and actual values, the required time for a unit section can be consistently calculated for any traffic flow and any route with any service interval. A route bus operation management device that can make predictions, calculate the accurate arrival time of the target vehicle at a specific point on the route, and also track and manage bus operation status and display the arrival time. The purpose is to obtain.

[Means for solving problems]

The route bus operation management device according to the present invention stores the actual value of the travel time required for passing through a unit section in the past in the central processing unit, extracts the actual value to be used as necessary, and The data is first processed into data (delay coefficient) that allows relative comparison of the degree of delay for each route, and then weighted according to whether the actual value is new or old, and these values are used to calculate the existing value for each unit section. The central processing unit calculates the moving average value of traveling vehicles and sample values of vehicles to be predicted, and totals the predicted travel time for each unit section.

[Effect]

Calculation of the estimated arrival time in this invention involves storing the basic information of the operation plan and the standard operation time required in advance in the storage unit of the central processing unit, and also storing the actual required time of vehicles that have already traveled (before the prediction target vehicle). The values are collected and stored in the storage unit, and this stored information is extracted and subjected to primary processing, focusing on the latest actual values that are considered to best reflect the traffic conditions at that time. By using data to perform predictive calculations, we can improve the accuracy of predicting arrival times on routes with any service interval, and also make the tracking of bus operation status and display of arrival times more accurate.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings. In FIG. 1, 1 is a central processing unit, 2a,
2b and 2c are terrestrial radios installed at point A, point B, and point C, respectively; 43a, 43b, and 43c
are antennas of terrestrial radio equipment 2a, 2b, 2c, 44
a, 44b, 44c are lines connecting the central processing unit 1 and the terrestrial radios 2a, 2b, 2c; 5a, 5;
b, 5c are route buses, 6a, 6b, 6c are on-board radios of route buses 5a, 5b, 5c, 47a, 47
b, 47c are the antennas of the on-board radios 6a, 6b, 6c, 48 is the travel direction of the route buses 5a, 5b, 5c, and 49 is the travel route of the route buses 5a, 5b, 5c.

FIG. 2 shows details of the central processing unit 1. In the figure, numeral 10 mainly consists of a microprocessor, which controls reading and writing data to storage units 11 to 20, reads out necessary data from the storage units 11 to 20, and performs calculations. It processes the data and writes the results to the storage unit. Reference numeral 11 denotes a basic information storage section on the operation plan provided for each point and each route, and this operation plan basic information storage section 11 stores vehicle numbers, passing times, timetable numbers, etc. Reference numeral 12 indicates a standard operation time storage section provided for each route and route; 13 indicates a passage information storage section for storing the vehicle number, passing time, and timetable number of passing buses; and 14 indicates a storage section provided for each route and location. 15 is an actual operation interval storage section provided for each route; 16 is a storage section for delay coefficients in which the actual operation time is normalized by the standard operation time; and 17 is a storage section for the actual operation time required for actual operation. A weighting storage unit, 18 a sample value storage unit that stores a moving average value (sample value) of vehicles that have already traveled in a unit section, 19 a parameter storage unit that stores parameters used when calculating weighting and sample values, and 20 a A display output unit that outputs the vehicle number, route diagram number, arrival time, or departure time, which are the results calculated and determined by the processing unit 10, and is displayed on a display device (not shown) for the operation manager and at the terminal or main stop. It drives a device (not shown), etc.

Next, the operation will be explained. Route bus 5a, 5
On-board radios 6a, 6b, 6c are installed on b, 5c,
On-board radios 6a, 6b,
terrestrial radio device 2a, which transmits and receives bus information to and from 6c;
2b and 2c are arranged as described above, and the central processing unit 1 is installed in a business office or the like, the lines 44a,
44b and 44c, the central processing unit 1 collects bus passage information from the terrestrial radios 2a, 2b, and 2c, processes it in the processing unit 10 shown in FIG. 2, and stores it in the passage information storage unit 13. That is, the passage information for each bus passing through points A, B, and C (by vehicle number, by point, by timetable number) is compared with the operation plan basic information storage unit 11 shown in FIG. The information is stored in the section 13. In addition, the passage information storage unit 1
3, necessary data are read out and processed by the processing section 10, and the results are stored in the actual travel time storage section 14.

In FIG. 1, this device determines when the bus 5a will arrive at point B after passing point A.
Determine as follows. In order to predict the time required to travel between points A and B, we use existing vehicles that have passed point B before the prediction target vehicle 5a (vehicle in front, vehicle in front,
Calculations are performed using processed relatively new actual values of the previous car, etc.), and using the "delay coefficient", "weighting", and "moving average value = sample value" defined below.

Normally, the standard travel time between points for route buses is determined in advance, and varies depending on the time of operation and the route.

Therefore, since the actual value of vehicles that have already traveled between points A and B on the route bus also differs depending on the time of operation and the route, it is necessary to use a value that is compared with some reference value. In this device, this standardized required time is defined as a delay coefficient D _i as follows.

D _i = r _i /T _S (i=0, -1, -2,...)... (100) However, r _i : Actual operation time T _S : Standard operation time Front car 5b, the car in front 5 which is a running vehicle
c, Actual travel time r for the car in front (not shown)
〓, r _-1 and r _-2 are read out from the actual operation required time storage section 14 shown in Fig. 2, and the standard operation required time T _S is read out from the standard operation required time storage section 12, As a result of calculation processing using both values in step 10, D ₀ = r ₀ /T _S , D _-1 = r _-1 /T _S and D _-2 = r _-2 /T _S are the delay coefficients of each of the vehicles that have already traveled. is stored in the delay coefficient storage unit 16 as follows. In the conventional method, when predicting the travel time for a unit section (corresponding to points A-B and B-C in Figure 1), the average value of past actual values is simply used, resulting in an unclear expression. ing.

In the present invention, the target time is set by paying attention to which past performance value should be used and how far back in time it should be used. In other words, route buses operate at different intervals depending on the route and section, for example, some run every 3 minutes, others every 30 minutes, and the number of actual values used, that is, the number of samples. Occurs to some extent. Therefore, actual values should be used depending on the characteristics of the route and section. Furthermore, since road traffic changes from moment to moment, it is considered that the use of old performance values that go back too far into the past will not necessarily match current values. That is, since the latest actual value most reflects the traffic situation at that time, in the present invention, when extracting the actual value by limiting the time period,
The actual values are weighted. Here, the newer the actual value is, the larger it is, and the older the actual value is, the smaller it is.
This weight W _i is defined as a function of the service interval as follows.

W _i =a+s _i -s _i-1 /b (maximum W _i =1.0)...(101) a≦W _i ≦1, i=0, i=-1, i=-2,
... (Vehicle in front) (Vehicle in front) (Vehicle in front) However, a: Weight correction coefficient b: Upper limit set value of service interval s: Arrival time of route bus at point A Here, s ₀ , s _-1 is the vehicle in front 5 at point A in Figure 1.
b, the arrival time of the car 5c in front, and a, b are parameters.

a is s ₀ =s ₋₁ , that is, the weight of the arriving car when the car in front and the car in front arrive at the same time in Dango operation, and b is the upper limit set value of the travel interval at which the data of the car in front is taken in. For example, if b=30 (minutes), when a=1/3 and s ₀ −s ₋₁ ≧20 (minutes), the weight of the vehicle in front 5b becomes W ₀ =1. As for weighting, when the service interval is short, the number of samples is large and the weight is distributed, and when the service interval is long, the number of samples is small and the weight becomes larger for the actual value that passed immediately before the target vehicle to predict. Weighting for each vehicle that has already traveled can be calculated using formula (101), and the weighting determined in this way is stored in the weighting storage section 17. Also, point A-
Even in the same time period, actual values between B vary depending on the number of passengers, waiting at traffic lights, and other changes in traffic conditions. ) to calculate the predicted arrival time.

Here, the sample values of the vehicle in front 5b and the vehicle in front 5c are defined as follows.

l ₀ = W ₀ D ₀ + (1-W ₀ ) l _-1 ...(102a) l _-1 = W _-1 D _-1 + (1-W _-1 ) l _-2 ... (102b) However, l ₀ : Sample value of the vehicle 5b in front, l _-1 : Sample value of the vehicle 5c in front, l _-2 : Sample value of the vehicle two in front (not shown in Figure 1) W ₀ , W _-1 : Front Weights of the vehicle and the vehicle in front D ₀ , D ^- ₁ : Delay coefficients of the vehicle in front and the vehicle in front The sample values of the vehicles that have already traveled are obtained by reading out the weighting from the weighting storage unit 17 in FIG. The delay coefficient is read out from the storage unit 16, and the processing unit 1
0, calculations are performed using equations (102a) and (102b), and the results are stored in the sample value storage unit 18.

The prediction calculation of the arrival time of the prediction target vehicle at point B is performed using the sample value l of the prediction target vehicle derived from the sample values of the vehicle in front and the vehicle in front and the passing time at point A of the vehicle in front and the vehicle in front of the vehicle. ₁ (predicted value).

FIG. 3 is a graph showing the situation of sample value vs. section entry time, where the section entry time means the passing time of the prediction target vehicle 5a at point A in FIG. and the section entry time of the predicted vehicle,
The vertical axis shows sample values of vehicles that have already traveled and vehicles to be predicted. From FIG. 3, the sample value l ₁ of the prediction target vehicle is derived as follows.

l ₁ −l ₀ /s ₁ −s ₀ =k ₁・l ₀ −l ₋₁ /s ₀ −s ₋₁ , that is, l ₁ =l ₀ +k ₁・l ₀ −l ₋₁ /s ₀ −s _{− 1}・(s ₁ − s ₀ )……(103
) However, k ₁ : Slope of straight line Here, the value of k ₁ is the slope of the straight line, but here it is assumed that it is approximated by the slope of a quadratic curve, which is easy to calculate. Therefore, k ₁ becomes:

When s _i < s ₀ + c, k ₁ = 1-s _i - s ₀ /2c (i = 1, 2, 3,...) ...(104a) When s _i ≧ s ₀ + c, k ₁ = 0.5 (s _i = s ₀ + c)...
...(104b) However, c: Prediction upper limit setting value The travel time (predicted value) of the prediction target vehicle 5a between points A and B is based on the sample value _l1 .
It is multiplied by the standard travel time T _S between.
In other words, the time required to travel between points A and B = l ₁ × T _S
becomes.

Therefore, the time when the prediction target vehicle 5a passes point B is:
Since the time required to travel between points A and B is added to the time of passing point A, the time of passing point B = s ₁ +l ₁ ×T _S (105). Therefore, in order to predict the passing time of the prediction target vehicle 5a at point B, the sample values of the previously traveled vehicles shown in FIG. Call from the passage information storage unit 13, read the parameter c from the parameter storage unit 19, and calculate the equations (103) and (104).
The processing unit 10 calculates the equation and stores the sample value l ₁ of the prediction target vehicle 5a in the sample value storage unit 18, and also calculates the equation (105) in the processing unit 10 to determine the passing time of the prediction target vehicle. Derive. In a similar manner, the passing time of the prediction target vehicle at point C can also be determined by calculating and summing the required travel time for each unit section AB and BC.

Therefore, the passing time at points beyond point C, which are not shown in Fig. 1, can also be calculated by summing up the travel time (prediction etc.) for each unit section using the actual values of vehicles that have already traveled. Predictive calculations can be made. Processing section 10 shown in FIG.
Then, if the result of calculating the predicted passing time of the vehicle to be predicted is read out from the operation plan basic information storage section 11, the actual value is read out from the passing information storage section 13, compared, and outputted to the display output section 20, the route will be determined. The planned passing time, actual value, and predicted time of each point along the travel route of the buses 5a to 5c can be displayed.

As a result, the operation status of each bus can be displayed for each point on a CRT screen, etc. at a business office, etc., so tracking management is possible. The predicted arrival time or scheduled departure time can also be displayed on display devices along the travel route other than at the destination.

In addition, in the above embodiment, when deriving the sample value (predicted value) of the prediction target vehicle using equation (103), the slope k ₁ of the straight line is approximated by a quadratic curve and the calculation process is performed ( Although the formula 104) is set, the same effect as in the above embodiment can be obtained even if it is approximated by another easy-to-calculate function.

〔Effect of the invention〕

As described above, according to the present invention, the operation plan value and the actual value are stored in the storage unit of the central processing unit, and the arrival time can be calculated by the processing unit based on the latest actual value by limiting the time period. With this configuration, it can be applied to routes with any operating interval, and highly accurate prediction results can be obtained. This allows the bus operator to accurately display the estimated time of arrival on the route, allowing the bus operator to monitor the status of the bus without leaving the office. It also becomes possible to display the information at major bus stops, which has the effect of contributing to customer service.

[Brief explanation of drawings]

Fig. 1 is a block connection diagram of a route bus operation control device according to an embodiment of the present invention, Fig. 2 is a block connection diagram of a central processing unit according to an embodiment of the invention, and Fig. 3 is an embodiment of the invention. A graph showing how predicted sample values are derived from actual sample values according to an example, and FIG. 4 is a block connection diagram of a conventional automobile operation control system. 1 is a central processing unit, 2a, 2b, 2c are ground radios, 5a, 5b, 5c are route buses, 6a, 6
b, 6c are on-board radios, 10 is a processing unit, 11 is an operation plan basic information storage unit, 12 is a standard operation required time storage unit, 13 is a passage information storage unit, 14 is an actual operation required time storage unit, 15 is an actual running interval storage section, 16 is a delay coefficient storage section, 17 is a weighting storage section, 18 is a sample value storage section, 19 is a parameter storage section, 2
0 is a display output unit, 41 is a central operation control device, 42
a, 42b, 42c are ground receivers, 43a, 43
b, 43c are antennas, 44a, 44b, 44c
is the line, 45a, 45b, 45c are specific cars,
46a, 46b, 46c are on-board radio transmitters, 47
a, 47b, and 47c are on-vehicle antennas, 48 is a running direction, and 49 is a running route. In addition, in the figures, the same reference numerals indicate the same or equivalent parts.

Claims (1)

[Claims] 1 Consists of an on-board radio installed on a track bus, a ground radio that sends and receives information to and from the on-board radio, and a central processing unit installed at a business office, etc. In the route bus operation management device, which collects bus passing information from , the processing unit of the central processing unit for calculating the arrival time by adding the total estimated travel time of each unit section into which the untraveled section is divided, and the basic information of the operation plan, passing information, and the actual time of the unit section. It is equipped with a storage unit that stores the required travel time, the standard travel time, and the actual travel interval, and limits the target time of the past travel time used for prediction, and uses the actual travel time as the standard travel time. The moving average value of a plurality of vehicles that have already passed through the unit section before the prediction target vehicle is calculated using the delay coefficient normalized by the required time and the weighted value for the actual travel time. The processing unit calculates, further limits the entry time of each unit section of the prediction target vehicle and the already running vehicle, and the prediction target time, and calculates the value of each unit interval of the prediction target vehicle from the calculated moving average value. The sample value is calculated by the processing unit, and the predicted travel time for each unit section is calculated as the product of the predicted sample value and the standard operation time, and in the same way, the predicted travel time for each unit section in the untraveled section is calculated. Calculate the arrival time at a specific point by accumulating the minutes,
A route bus operation management device characterized by displaying the tracking management of bus operation status and arrival time on a display output section.