METHOD AND STRUCTURE FOR VEHICULAR TRAFFIC PREDICTION WITH LINK INTERACTIONS

• US 20080175161A1
• Filed: 01/24/2007
• Published: 07/24/2008
• Est. Priority Date: 01/24/2007
• Status: Active Grant

• Current Assignee: Tomtom Global Content B.V (TomTom NV)
• Original Assignee: International Business Machines Corporation
• Inventors: Wynter, Laura, Min, Wanli, Amemiya, Yasuo
• Granted Patent: US 7,953,544 B2
• Time in Patent Office: Days
• US Class Current: 370/252
• CPC Class Codes: G08G 1/0104 Measuring and analyzing of parameters relative to traffic conditions

Abstract

A method and structure for predicting traffic on a network, includes a receiver which receives data related to traffic on at least a portion of a network. A calculator calculates a traffic prediction for at least a part of the network, the traffic prediction being calculated by using a deviation from a historical traffic on the network.

Claims

1. An apparatus, comprising:

a receiver to receive data related to traffic on at least a portion of a network; and

a calculator to calculate a traffic prediction for at least a part of said network,

wherein said traffic prediction is calculated by using a deviation from a historical traffic on said network.

2. The apparatus of claim 1, wherein said network comprises a plurality of interconnected links and a traffic prediction for a link in said network comprises a calculation of a deviation of a historical traffic for said link.

3. The apparatus of claim 2, wherein said traffic prediction for said link is calculated using a relationship vector that defines other links in said network that affect a traffic amount in said link within a specific time duration.

4. The apparatus of claim 2, wherein said calculator further calculates said historical traffic for said link as a calibration for traffic in said link.

5. The apparatus of claim 4, wherein said historical traffic is periodically re-calculated by said calculator.

6. The apparatus of claim 3, wherein said calculator calculates, for each link in said relationship vector, a traffic deviation from a historical traffic for each said link, and said traffic deviation for said link is expressed as a difference vector for said link, said difference vector comprising a vector of deviations of traffic of each link in said relationship vector.

7. The apparatus of claim 6, wherein said difference vector is adjusted by an auto-regressive model that modifies said deviations in said difference vector based upon data of previous time intervals for each link in said relationship vector.

8. The apparatus of claim 2, wherein said prediction comprises a prediction for a first time interval and predictions for subsequent time intervals comprise sequential reiterations of said prediction for said first interval.

9. The apparatus of claim 1, wherein said data related to said traffic prediction comprises one or more of:

traffic speed;

traffic density; and

traffic flow.

10. A method of predicting traffic on a network, said method comprising:

receiving data related to at least a portion of said network; and

calculating a traffic prediction for at least a part of said traffic network by using deviation from a historical traffic on said network.

11. The method of claim 10, wherein said network comprises a plurality of interconnected links and a traffic prediction for a link in said network comprises a calculation of a deviation of a historical traffic for said link.

12. The method of claim 11, wherein said traffic prediction for said link is calculated using a relationship vector that defines other links in said network that affect a traffic amount in said link within a specific time duration.

13. The method of claim 11, further comprising calculating said historical traffic for said link as a calibration for traffic in said link.

14. The method of claim 13, further comprising periodically calculating said historical traffic.

15. The method of claim 12, further comprising, for each link in said relationship vector, calculating a traffic deviation from a historical traffic for each said link, said traffic deviation for said link being expressed as a difference vector for said link, said difference vector comprising a vector of deviations of traffic of each link in said relationship vector.

16. The method of claim 15, further comprising adjusting said difference vector using an auto-regressive model that modifies said deviations in said difference vector based upon data of previous time intervals for each link in said relationship vector.

17. The method of claim 11, wherein said prediction comprises a prediction for a first time interval, said method further comprising re-iterating said prediction of said prediction for said first interval as a prediction for each of a subsequent time intervals for which a future prediction is to be made.

18. The method of claim 10, wherein said data related to said traffic prediction comprises one or more of:

traffic speed;

traffic density; and

traffic flow.

19. A signal-bearing medium tangibly embodying a program of machine-readable instructions executable by a digital processing apparatus to perform a method of predicting traffic on a network, said program comprising:

a receiver module to receive data related to traffic on at least a portion of a network; and

a calculator module to calculate a traffic prediction for at least a part of said network,

wherein said traffic prediction is calculated by using a deviation from a historical traffic on said network.

20. The signal-bearing medium of claim 19, wherein said network comprises a plurality of interconnected links and a traffic prediction for a link in said network comprises a calculation of a deviation of a historical traffic for said link.