WO2011161679A1

WO2011161679A1 - A method for predicting the arrival times at public transportation stations

Info

Publication number: WO2011161679A1
Application number: PCT/IL2011/000499
Authority: WO
Inventors: Tomer Y. Morad; Shachar Daniel; Guy Cohen
Original assignee: Transspot Ltd.
Priority date: 2010-06-24
Filing date: 2011-06-23
Publication date: 2011-12-29
Also published as: IL206592A0

Abstract

The present invention relates to a method for predicting the arrival of a public transportation vehicle at a preset station comprising the steps of: (a) receiving the route of said public transportation vehicle; (b) receiving the coordinates of a station of said public transportation vehicle; (c) receiving the segments of said route and their related statistics; (d) receiving at least one GPS reading, indicative of the location of said public transportation vehicle; and (e) calculating the predicted travel time required for said public transportation vehicle to reach said preset station based at least on the location of said public transportation vehicle and said statistics of said segments leading to said preset station.

Description

A METHOD FOR PREDICTING THE ARRIVAL TIMES AT PUBLIC TRANSPORTATION STATIONS

Field of the Invention

The present invention relates to the field of public transportation. More particularly, the invention relates to a method for predicting the arrival time of a traveling public transportation vehicle at the preset stations of its route.

Background of the Invention

Today, some of the public transportation vehicles offer their passengers information display systems. These systems typically display general information, such as weather forecasts and news, and location related information, such as a map of the suiroundings and related commercial information. Some of these public transportation systems also display the predicted time for arrival at the remaining stations en route.

Because travel time can be extremely variable, it is useful to provide the passengers estimates of travel time and prediction intervals. However, arrival time prediction of public transportation vehicles, especially traffic dependant vehicles, is extremely complicated.

Different techniques for predicting the arrival time of buses at stations are described in a paper by Ran Hee Jeong titled "The prediction of bus arrival time using automatic vehicle location systems data", Texas A&M University, December 2004. In the paper, a travel time prediction model had been examined over a bus route running in the downtown of Houston, Texas. Three kinds of models were used to predict bus arrival time: a historical based model, regression models, and artificial neural network (ANN) models. Nevertheless, all these models are based on historic information that are extrapolated for predicting the arrival time of future travels.

US 2007/0210936 discloses a system and method for generating imminent arrival alerts of a public transportation vehicle. The determination of the bus's location and the information about its projected route are used to create and transmit predictive alerts. The described method predicts when the bus will arrive at a certain bus stop by knowing the route and the bus's present location. After this determination, bus riders, who wish to board at this certain bus stop, are alerted to the imminent arrival of the bus at their bus stop. Nevertheless, the described system requires utilizing the location determination capabilities of GPS-enabled phones and the cooperation of the users.

Summary of the Invention

It is an object of the present invention to provide a method for predicting the arrival times of a public transportation vehicle at its preset stations.

It is another object of the present invention to provide a method for displaying to the users of public transportation the projected arrival time at the designated stations. It is still another object of the present invention to provide a method for automatically updating, in real time, the predicted arrival times of a pubhc transportation vehicle at its preset stations.

Other objects and advantages of the invention will become apparent as the description proceeds.

The present invention relates to a method for predicting the arrival of a pubhc transportation vehicle at a preset station comprising the steps of: (a) receiving the route of said pubhc transportation vehicle; (b) receiving the coordinates of a station of said pubhc transportation vehicle; (c) receiving the segments of said route and their related statistics; (d) receiving at least one GPS reading, indicative of the location of said pubhc transportation vehicle; and (e) calculating the predicted travel time required for said pubhc transportation vehicle to reach said preset station based at least on the location of said pubhc transportation vehicle and said statistics of said segments leading to said preset station.

Preferably, the method further comprises the step of calculating the predicted travel times required for the pubhc transportation vehicle to reach a number of preset stations on the route.

Preferably, the creation of the segments of the route and their related statistics comprises the following steps: (a) receiving the route of said vehicle; (b) receiving the coordinates of the stations of said pubhc transportation; (c) traveling on said route; (d) gathering statistics on said route during traveling; and (e) dividing said route into segments based on the gathered statistics. Preferably, the statistics comprise at least one of the following: segment length, segment coordinates, average travel time, variance of the travel time, distribution of the travel times, or any combination thereof.

Preferably, the method is apphed more than once, and the collected traveling statistics for each segment is averaged.

In one embodiment, the segments of said route are the route segments between the stations.

In one embodiment, predicting the travel time for traveling a part of a segment is calculated by dividing the length of said part of said segment by the actual measured average speed of the vehicle on any part of said segment.

In one embodiment, predicting the travel time for traveling a part of a segment is calculated using real time updates from other public transportation vehicles traveling in the vicinity of said segment.

Brief Description of the Drawings

In the drawings:

- Fig. 1 is a block diagram of the system for statistically mapping and predicting the arrival time of the public transportation vehicle at a number of preset stations of its route according to an embodiment of the invention. ί

- 5 -

- Fig. 2 is a block diagram of the method for statistically mapping a designated route according to one embodiment.

^■ Fig. 3 is a block diagram of the method for predicting the arrival time, of a public transportation vehicle, at a preset station, while traveling en route, according to one embodiment.

Detailed Description of Preferred Embodiments

For the sake of brevity the following terms are defined explicitly:

- Route— representation of a linear approximation of a predetermined path operated by a public transportation vehicle. The route may belong to any route formation technique or method such as the Directed Acyclic Graph (DAG) method, where each vertex represents a geographic location and each edge represents a linear geographic line between two locations. The first and last vertices in the route have only one edge, and all other vertices have exactly two edges.

- Fragment— a preset portion of the route between two representations of consecutive GPS readings on the route.

- Segment — a portion of the route which contains a number of consecutive fragments according to their approximated speed.

Fig. 1 is a block diagram of the system for statistically mapping and predicting the arrival time of the public transportation vehicle at a number of preset stations of its route according to an embodiment of the invention. In this embodiment, the system 300 is preferably located on a traveling vehicle, which may be a bus, car, ship, boat, plane or any other vehicle. The system 300 has a controller 100, such as a processor, microprocessor, or any other known control-unit for controlling the system 300. The system also comprises a GPS receiver 140 for providing the coordinates of the approximated location, and two databases (DB), 110 and 160, where the two DB 110 and 160 may be stored in the same repository, or any other constellation thereof. The DB 110 is used for storing the route and other route related information, and the DB 160 is used for storing the coordinates of the stations. One of the tasks of the controller 100 is to calculate distance, time, and speed of the various segments of the route using the GPS readings received from GPS receiver 140, and the clock 150. The system 300 may have a display 130 for displaying information, and communication means 120 for communicating information.

Fig. 2 is a block diagram of the method for statistically mapping a designated route according to one embodiment. In step 1 the system on a vehicle, such as system 300 described in relations to Fig. 1, receives a public transportation route and the coordinates of the stations for the designated route. The data may be received in any known way of communicating data, wired or wireless, such as by cellular networking, Blue -Tooth, direct uploading from a repository, etc. There are many methods for finding, forming, presenting and storing the route electronically; such methods may be found in USPTO 6,366,851. Typically, there is no need to store a map of the whole city or area, only the required information concerning the route. In one of the embodiments the received route includes alternating routes for the public transportation vehicle. The received stations are in relations to the received route, where each station may be linked to a specific location on the route. For example, if a route includes traveling on a street back and forth, each of the stations on that street is linked, either with the route of traveling back or the route of travehng forth. The data may be received in any known way of communicating data, wired or wireless, such as by cellular networking, Blue- Tooth, direct uploading from a repository, etc. In step 2 the vehicle travels the received route and stops at the designated stations and collects the traveling statistics on the route. The integrated GPS 140 provides periodic readings for indicating the location of the vehicle, where the system 300 integrates these readings with the readings from the clock 150. Thus the system 300 can interpolate the varying speeds of the vehicle traveling the different fragments of the route. In other words the system can find and store the approximated speed of each fragment of the route. In another embodiment the speed is extracted from the odometer outputs of the vehicle. In another embodiment the GPS readings are adjusted prior to their use by dead reckoning methods. Such dead reckoning methods may use data from the vehicle's odometer, a gyro, an accelerometer, or any combination thereof. In step 3 the data from step 2 is analyzed from at least one journey and the stored route is divided into a number of segments, where each segment contains a number of consecutive fragments according to their approximated speed. Typically, when the route is divided into segments, each segment contains a number of consecutive fragments where the deviation in their speed is under a certain threshold. In one embodiment a preset deviation is designated, where consecutive fragments having a similar speed within the designated deviation are aggregated together into one segment. In an embodiment, the aggregation of fragments into segments is done by first selecting a threshold, then starting from the found speed of the first fragment and aggregating consecutive fragments of similar found speed as long as the deviations of all the approximated speeds of the fragments are within the selected threshold. If, for example, the first 10 fragments have been aggregated and the 11^th fragment's speed has a large deviation from one of the former fragments then a new segment is declared and the consecutive fragments, i.e. 12^th 13^th etc., are aggregated with the 11^th fragment as long as their approximated speed doesn't vary more that the threshold from the other fragments of the segment. Thus the fragments may be aggregated into segments of similar speeds and the total route is divided into segments of different averaged speeds. The threshold may be set according to the nature of the route or the vehicle, or any other requirement, and may also be found according to trial and error.

In one embodiment the threshold used for aggregating the fragments into segments is variable and may depend on the average speed of the segment. For example, a larger threshold may be used for segments having an average speed of 50 Km h, than a threshold for segments having an average speed of 10 Km/h.

In one of the embodiments, step 2 of the method described in relations to Fig. 2, is applied more than once, by the same vehicle or by a number of vehicles, and their collected traveling statistics on the route is averaged. Thus step 3 may be implemented on the averaged fragments for producing more accurate results. In one embodiment, the route is divided into segments from each distinct journey. The final division of the route into segments is calculated by identifying close clusters of segment divisions from different journeys and dividing the route into segments based on these clusters. In another embodiment, the average speed and variance of the speed for each point on the route is calculated. In this embodiment, the route is divided into segments at the points where the average speed varies outside a given threshold, or the variance varies above a given threshold. In one embodiment, segments cannot have a length below a certain threshold, such as a threshold of 25m. In one embodiment, the route segments statistics include: average of the travel time, variance of the travel time, any other segment related information, or any other combination thereof. In one embodiment, the statistics are collected for different times of days and different days of the week.

Fig. 3 is a block diagram of the method for predicting the arrival time of a public transportation vehicle at a preset station while traveling en route, according to one embodiment. In step 11 the system on the public transportation vehicle, such as system 300 described in relations to Fig. 1, receives a public transportation route and the coordinates of the stations for the designated route. The received stations are in relations to the received route, where each station may be linked to a specific location on the route. The data may be received in any known way of communicating data, wired or wireless, such as by cellular networking, Blue -Tooth, direct uploading from a repository, etc. In step 12 the system receives the information and statistics concerning the segments, as described in relations to Fig. 2. In one embodiment the system receives the length and average speed of each segment. In some of the embodiments steps 11 and 12 may be combined and their data may be transferred together. In step 13, when the vehicle is traveling, the system may receive periodic GPS readings from a GPS receiver. The GPS readings are used for determining the present location of the vehicle in relations to the segment it is traveling, and to determine the traveling speed of the vehicle on the partial segment traveled, for calculating the time remaining for traveling the remaining part of the segment. For example when a vehicle is present at the middle of a segment the system may conclude that the time for traveling the rest of the segment will be similar to the time spent on traveling the first half of the segment. In another embodiment, when a vehicle is present at the middle of a segment the system may conclude that the average speed measured so far in the segment will be the average speed in the remaining part of the segment. In another embodiment, the average speed measured in the partial part of the segment and the speed provided in the statistics of the segment are used together to estimate the speed in the remaining part of the segment. Thus the system may predict the time remaining for traveling to the end of the presently traveled segment. In step 14 the system predicts the traveling time for the following segments based on the speed calculations of the former segments. For example, if the system calculates that the traveling vehicle has traveled the first segment in a lower average speed than the one expected, i.e. received in step 12, in relations to that segment, the system may conclude that the traffic is slower at this stage and therefore it may extrapolate that the other segments will be slower as well. In one embodiment, the measured speed of a segment is compared to the average speed and the variance of the speed provided in the statistics of the segment, in order to predict the speed of a following segment according to the following segment's average speed and variance of the speed. Thus the GPS readings are used for concluding the whereabouts of the vehicle and for calculating the speed of the vehicle in relations to the received statistics of the segments. In step 15 the system uses the predictions from steps 13 and 14 in order to predict the arrival time at the next station of the route. In one of the embodiments the system also predicts the arrival time at the following stations as well, based on the prediction found in step 15. In one embodiment, the historical average speed for each of the subsequent segments to the subsequent stations is used to calculate the remaining travel duration for arrival at said stations. In one embodiment these predicted arrival times at the stations of the route are displayed to the users of the vehicle. In one embodiment the new calculated traveling time of the segments may be used to update the segments' statistics. In one embodiment these new updated statistics may be transmitted to the other vehicles designated for traveling the same segments, or other segments in the vicinity, such as, vehicles that will travel within a radius of 100 meters of the said segment. For example, when a vehicle enters a traffic jam, its average speed is reduced. In one embodiment, when the average speed varies outside a preset threshold, a notification is sent to other vehicles designated for traveling the same segments in the near future. When the vehicles receive the notification, they may consider the recent travel speeds in their future predictions. In one embodiment, all recent travel speeds received in the recent past, for example within the last 30 minutes, supersede the historical average travel speeds for the same segments. In another embodiment, the recent travel speeds received in the recent past are averaged with the historical travel speeds.

In one embodiment the route segments are actually the segments between stations.

While some embodiments of the invention have been described by way of illustration, it will be apparent that the invention can be carried into practice with many modifications, variations and adaptations, and with the use of numerous equivalents or alternative solutions that are within the scope of persons skilled in the art, without departing from the invention or exceeding the scope of claims.

Claims

1. A method for predicting the arrival of a pubhc transportation vehicle at a preset station comprising the steps of:

a. receiving the route of said public transportation vehicle; b. receiving the coordinates of a station of said pubhc transportation vehicle;

c. receiving the segments of said route and their related statistics; d. receiving at least one GPS reading, indicative of the location of said pubhc transportation vehicle; and

e. calculating the predicted travel time required for said pubhc transportation vehicle to reach said preset station based at least on the location of said pubhc transportation vehicle and said statistics of said segments leading to said preset station.

2. A method according to claim 1 further comprising the step of calculating the predicted travel times required for the pubhc transportation vehicle to reach a number of preset stations on the route.

3. A method according to claim 1, where the creation of the segments of the route and their related statistics comprises the following steps: a. receiving the route of said vehicle;

b. receiving the coordinates of the stations of said pubhc transportation;

c. traveling on said route;

d. gathering statistics on said route during traveling; and e. dividing said route into segments based on the gathered statistics.

4. A method according to claim 3, where the statistics comprise at least one of the following: segment length, segment coordinates, average travel time, variance of the travel time, distribution of the travel times, or any combination thereof.

5. A method according to claim 3, where the method is applied more than once, and the collected traveling statistics for each segment is averaged.

6. A method according to claim 1, where the segments of said route are the route segments between the stations.

7. A method according to claim 1, where predicting the travel time for traveling a part of a segment is calculated by dividing the length of said part of said segment by the actual measured average speed of the vehicle on any part of said segment.

8. A method according to claim 1, where predicting the travel time for traveling a part of a segment is calculated using real time updates from other public transportation vehicles traveling in the vicinity of said segment.