System, method and device for predicting navigational decision-making behavior

• US 8,478,642 B2
• Filed: 10/20/2009
• Issued: 07/02/2013
• Est. Priority Date: 10/20/2008
• Status: Expired due to Fees

• Current Assignee: Carnegie Mellon University
• Original Assignee: Carnegie Mellon University
• Inventors: Dey, Anind K., Bagnell, James Andrew, Ziebart, Brian D.
• Primary Examiner(s): Sittner, Matthew T
• Application Number:
  US12/581,879
• Publication Number:
  US 20100106603A1
• Time in Patent Office: 1,351 Days
• Field of Search: None
• US Class Current: 705/14.4
• CPC Class Codes: G01C 21/3484 Personalized, e.g. from learned user behaviour or user-defined profiles
• CPC Class Codes: G01C 21/3617 using user history, behaviour, conditions or preferences, e.g. predicted or inferred from previous use or current movement
• CPC Class Codes: G06Q 30/0266 based on the position of the vehicle
• CPC Class Codes: G08G 1/096811 where the route is computed offboard
• CPC Class Codes: G08G 1/096838 where the user preferences are taken into account or the user selects one route out of a plurality
• CPC Class Codes: G08G 1/096844 where the complete route is dynamically recomputed based on new data
• CPC Class Codes: G08G 1/096888 where input information is obtained using learning systems, e.g. history databases

Abstract

A system method and device for predicting navigational decision-making behavior is provided. The location of a device is used to generate the current location of the device, and a history of past locations of the device. A behavior predicting application predicts the route a drive is going to travel based on driver preferences, such as left-hand turns, avoidance of bridges, etc., as well as the preferred destinations and/or routes traveled in the past. The preferences of a particular driver can be learned by the system to be implemented in the behavior predicting application, or manually input by the driver. The system can employ GPS or other positioning technology to determine the location of the device. The device can be a standard GPS-enabled in-car device, or other portable device such as a cell phone or PDA.

Claims

1. A system for predicting navigational decision-making behavior comprising:

a route predicting device having a location unit that determines a current location of the device; and

a behavior predicting application that generates a predicted route a current driver is intending to travel based on the current location of the device and a plurality of predetermined factors that predict the route the current driver is intending to travel, the predetermined factors including other driver data for a plurality of other drivers;

wherein the behavior predicting application generates the predicted route by gathering the other driver data from the plurality of other drivers and comparing the other driver data from the plurality of other drivers to those of the current driver to select a subset of drivers from the plurality of other drivers having behavior patterns that best match the current driver.

2. The system of claim 1 wherein the predetermined factors include driver preferences that indicate a discrete path the current driver is intending to travel based on preferred modes of operating a vehicle, and a frequency of a state of the device based on previously stored locations of the device.

3. The system of claim 1 wherein the location unit is a GPS receiver that communicates with a GPS satellite system to determine the current location of the device.

4. The system of claim 1 wherein the location unit comprises at least one of a gyro and an accelerometer that determines the current location of the device.

5. The system of claim 1 wherein routes gathered from each of the plurality of other drivers are grouped together to provide a single cost weight that is used in generating the predicted route.

6. A method for predicting navigational decision-making behavior of a driver comprising the steps of:

obtaining driver preference data containing driver preferences for a current driver;

obtaining routes traveled data based upon a current location of a device and at least one past location of the device; and

generating, with a processor, a predicted route based on the driver preference data and the routes traveled data and by gathering other driver data from a plurality of other drivers and comparing the other driver data from the plurality of other drivers to those of the current driver to select a subset of drivers from the plurality of other drivers having behavior patterns that best match the current driver.

7. The method of claim 6 wherein the driver preference data is obtained manually from the driver.

8. The method of claim 6 wherein the driver preference data is obtained by monitoring a driving style and behavior pattern of the driver employing the device to learn the driver preferences.

9. The method of claim 6 further comprising analyzing the driver preference data and the routes traveled data to improve the predicted route.

10. The method of claim 9 wherein the step of analyzing is performed using an inverse reinforcement learning approach.

11. The method of claim 9 wherein the step of analyzing is performed using a maximum entropy principle.

12. The method of claim 6 wherein a driver'"'"'s navigation is guided by directions based on the predicted route.

13. The method of claim 6 further comprising aggregating data to leverage increased statistical power for a group of drivers.

14. The method of claim 6 wherein a goal location of the driver is known, and is manually entered by the current driver to generate the predicted route.

15. The method of claim 6 wherein the predicted route is generated based upon an unknown goal location.

16. The method of claim 6 further comprising providing the driver with additional data relevant to the predicted route of the current driver.

17. The method of claim 16 wherein the additional data includes advertisements based upon the predicted route of the current driver.

18. The method of claim 6 wherein the device includes a processor that obtains the current location received from the GPS receiver and transmits the current location to a behavior predicting application to generate the predicted route.

19. The method of claim 6 further comprising the step of grouping together routes from each of the plurality of other drivers to provide a single cost weight that is user in generating the predicted route.

20. A display screen for displaying a predicted route based on navigational decision-making behavior, the display screen comprising:

a route display screen showing a map of an area surrounding a current location of a driver;

an intended route image showing the predicted route that the driver intends to travel, as determined by a prediction application; and

a plurality of buttons allowing the user to interface with the prediction application that determines the route that the driver intends to travel;

wherein at least one of the plurality of buttons allows the user to interface with the prediction application to generate the predicted route by gathering the other driver data from the plurality of other drivers and comparing the other driver data from the plurality of other drivers to those of the current driver to select a subset of drivers from the plurality of other drivers having behavior patterns that best match the current driver.

21. The display screen of claim 20 wherein the display screen is a touch-screen and the plurality of buttons are displayed and can be selected by touching the corresponding portion of the display screen.