Generating attribute models for use in adaptive navigation systems

• US 7,996,345 B2
• Filed: 03/15/2010
• Issued: 08/09/2011
• Est. Priority Date: 11/02/2006
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Rasmussen, Jens Eilstrup, Golding, Andrew R.
• Primary Examiner(s): STARKS, WILBERT L
• Application Number:
  US12/724,073
• Publication Number:
  US 20100174479A1
• Time in Patent Office: 512 Days
• Field of Search: 706/14, 706/45
• US Class Current: 706/14
• CPC Class Codes: G01C 21/3484 Personalized, e.g. from learned user behaviour or user-defined profiles

Abstract

Adaptive navigation techniques are disclosed that allow navigation systems to learn from a user'"'"'s personal driving history. As a user drives, models are developed and maintained to learn or otherwise capture the driver'"'"'s personal driving habits and preferences. Example models include road speed, hazard, favored route, and disfavored route models. Other attributes can be used as well, whether based on the user'"'"'s personal driving data or driving data aggregated from a number of users. The models can be learned under explicit conditions (e.g., time of day/week, driver ID) and/or under implicit conditions (e.g., weather, drivers urgency, as inferred from sensor data). Thus, models for a plurality of attributes can be learned, as well as one or more models for each attribute under a plurality of conditions. Attributes can be weighted according to user preference. The attribute weights and/or models can be used in selecting a best route for user.

Claims

1. A machine-readable medium encoded with instructions, that when executed by one or more processors, cause the processor to carry out a process for generating attribute models for use in a navigation system in a vehicle, the process comprising:

for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

collecting sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associating the sensor data with at least one of a plurality of conditions, wherein at least one of the plurality of conditions is from a group consisting of time of day and day of week, each condition descriptive of the driving session at the time the user travelled on the road segment;

for each of the plurality of road segments, determining a value of a road speed of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

for each of the plurality of conditions, storing the values of the road speed in a conditional variant model associated with the condition.

2. The machine-readable medium of claim 1, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

3. The machine-readable medium of claim 1, wherein storing the values of the road speed in a conditional variant model associated with the condition comprises:

clustering values of the road speed based on a similarity metric; and

forming a conditional variant model for an implicit condition based on the clustered values.

4. A method of generating attribute models for use in a navigation system in a vehicle, the method comprising:

for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

collecting sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associating the sensor data with at least one of a plurality of conditions, wherein at least one of the plurality of conditions is from a group consisting of time of day and day of week, each condition descriptive of the driving session at the time the user travelled on the road segment;

for each of the plurality of road segments, determining a value of a road speed of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

for each of the plurality of conditions, storing the values of the road speed in a conditional variant model associated with the condition.

5. The method of claim 4, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

6. The method of claim 4, wherein storing the values of the road speed in a conditional variant model associated with the condition comprises:

clustering values of the road speed based on a similarity metric; and

forming a conditional variant model for an implicit condition based on the clustered values.

7. A system for generating attribute models for use in a navigation system in a vehicle, the method comprising:

an attribute model learning module that, for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

receives sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associates the sensor data with at least one of a plurality of conditions, each condition descriptive of the driving session at the time the user travelled on the road segment;

a hazard modeler that, for each of the plurality of road segments, determines a value of a road safety of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

a model storage that, for each of the plurality of conditions, stores the values of the road safety in a conditional variant model associated with the condition.

8. The system of claim 7, wherein the model storage partitions values of the road safety into separate data buckets by explicit conditions, and wherein the attribute model learning module forms a conditional variant model for each respective explicit condition based on the values of the road safety in that bucket.

9. The system of claim 7, wherein the model storage clusters values of the road safety based on a similarity metric, and wherein the attribute model learning module forms a conditional variant model for an implicit condition based on the clustered values.

10. A machine-readable medium encoded with instructions, that when executed by one or more processors, cause the processor to carry out a process for generating attribute models for use in a navigation system in a vehicle, the process comprising:

for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

collecting sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associating the sensor data with at least one of a plurality of conditions, wherein at least one of the plurality of conditions is from a group consisting of time of day and day of week, each condition descriptive of the driving session at the time the user travelled on the road segment;

for each of the plurality of road segments, determining a value of road safety of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

for each of the plurality of conditions, storing the value of road safety in a conditional variant model associated with the condition.

11. The machine-readable medium of claim 10, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

12. The machine-readable medium of claim 10, wherein storing the values of road safety in a conditional variant model associated with the condition comprises:

clustering values of road safety based on a similarity metric; and

forming a conditional variant model for an implicit condition based on the clustered values.

13. A method of generating attribute models for use in a navigation system in a vehicle, the method comprising:

for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

collecting sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associating the sensor data with at least one of a plurality of conditions, wherein at least one of the plurality of conditions is from a group consisting of time of day and day of week, each condition descriptive of the driving session at the time the user travelled on the road segment;

for each of the plurality of road segments, determining a value of road safety of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

for each of the plurality of conditions, storing the values of road safety in a conditional variant model associated with the condition.

14. The method of claim 13, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

15. The method of claim 13, wherein storing the values of road safety in a conditional variant model associated with the condition comprises:

clustering values of road safety based on a similarity metric; and

forming a conditional variant model for an implicit condition based on the clustered values.

16. A system for generating attribute models for use in a navigation system in a vehicle, the method comprising:

an attribute model learning module that, for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

receives sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associates the sensor data with at least one of a plurality of conditions, each condition descriptive of the driving session at the time the user travelled on the road segment;

a disfavored route modeler that, for each of the plurality of road segments, determines a value of disfavor of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

a model storage that, for each of the plurality of conditions, stores the values of disfavor in a conditional variant model associated with the condition.

17. The system of claim 16, wherein the model storage partitions values of disfavor into separate data buckets by explicit conditions, and wherein the attribute model learning module forms a conditional variant model for each respective explicit condition based on the values of disfavor in that bucket.

18. The system of claim 16, wherein the model storage clusters values of disfavor based on a similarity metric, and wherein the attribute model learning module forms a conditional variant model for an implicit condition based on the clustered values.

19. A machine-readable medium encoded with instructions, that when executed by one or more processors, cause the processor to carry out a process for generating attribute models for use in a navigation system in a vehicle, the process comprising:

for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

collecting sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associating the sensor data with at least one of a plurality of conditions, each condition descriptive of the driving session at the time the user travelled on the road segment;

for each of the plurality of road segments, determining a value of a road familiarity of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

for each of the plurality of conditions, storing the values of the road familiarity in a conditional variant model associated with the condition.

20. The machine-readable medium of claim 19, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

21. The machine-readable medium of claim 19, wherein at least one of the plurality of conditions is from a group consisting of time of day and day of week.

22. The machine-readable medium of claim 19, wherein storing the values of the road familiarity in a conditional variant model, associated with the condition comprises:

clustering values of the road familiarity based on a similarity metric; and

forming a conditional variant model for an implicit condition based on the clustered values.

23. A method of generating attribute models for use in a navigation system in a vehicle, the method comprising:

for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

collecting sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associating the sensor data with at least one of a plurality of conditions, each condition descriptive of the driving session at the time the user travelled on the road segment;

for each of the plurality of road segments, determining a value of a road familiarity of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

for each of the plurality of conditions, storing the values of the road familiarity in a conditional variant model associated with the condition.

24. The method of claim 23, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

25. The method of claim 23, wherein storing the values of the road familiarity in a conditional variant model associated with the condition comprises:

clustering values of the road familiarity based on a similarity metric; and

forming a conditional variant model for an implicit condition based on the clustered values.

26. A machine-readable medium encoded with instructions, that when executed by one or more processors, cause the processor to carry out a process for generating attribute models for use in a navigation system in a vehicle, the process comprising:

for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

collecting sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associating the sensor data with at least one of a plurality of conditions, each condition descriptive of the driving session at the time the user travelled on the road segment;

for each of the plurality of road segments, determining a value of disfavor of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

for each of the plurality of conditions, storing the values of disfavor in a conditional variant model associated with the condition.

27. The machine-readable medium of claim 26, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

28. The machine-readable medium of claim 26, wherein storing the values of disfavor in a conditional variant model associated with the condition comprises:

clustering values of disfavor based on a similarity metric; and

forming a conditional variant model for an implicit condition based on the clustered values.

29. A method of generating attribute models for use in a navigation system in a vehicle, the method comprising:

for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

collecting sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associating the sensor data with at least one of a plurality of conditions, each condition descriptive of the driving session at the time the user travelled on the road segment;

for each of the plurality of road segments, determining a value of disfavor of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

for each of the plurality of conditions, storing the values of disfavor in a conditional variant model associated with the condition.

30. The method of claim 29, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

31. The method of claim 29, wherein storing the values of disfavor in a conditional variant model associated with the condition comprises:

clustering values of disfavor based on a similarity metric; and

forming a conditional variant model for an implicit condition based on the clustered values.

32. A machine-readable medium encoded with instructions, that when executed by one or more processors, cause the processor to carry out a process for generating attribute models for use in a navigation system in a vehicle, the process comprising:

for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

collecting sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associating the sensor data with at least one of a plurality of conditions, each condition descriptive of the driving session at the time the user travelled on the road segment;

for each of the plurality of road segments, determining a value of a road attribute-of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

for each of the plurality of conditions, storing the values of the road attribute in a conditional variant model associated with the condition, wherein storing the values of the road attribute in a conditional variant model associated with the condition comprises;

partitioning values of the road attribute into separate data buckets by explicit conditions; and

forming a conditional variant model for each respective explicit condition based on the values of the road attribute in that bucket.

33. The machine-readable medium of claim 32, wherein the road attribute is road speed.

34. The machine-readable medium of claim 33, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

35. The machine-readable medium of claim 33, wherein at least one of the plurality of conditions is from a group consisting of time of day and day of week.

36. The machine-readable medium of claim 32, wherein the road attribute is road safety.

37. The machine-readable medium of claim 36, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

38. The machine-readable medium of claim 36, wherein at least one of the plurality of conditions is from a group consisting of time of day and day of week.

39. The machine-readable medium of claim 32, wherein the road attribute is road familiarity.

40. The machine-readable medium of claim 32, wherein the road attribute is disfavored roads.

41. A method of generating attribute models for use in a navigation system in a vehicle, the method comprising:

for each of a plurality of driving sessions, during which the user travelled over a plurality of road segments in the vehicle;

collecting sensor data of the vehicle for each of the plurality road segments travelled during a driving session; and

associating the sensor data with at least one of a plurality of conditions, each condition descriptive of the driving session at the time the user travelled on the road segment;

for each of the plurality of road segments, determining a value of a road attribute of the road segment based upon the sensor data collected for that road segment, and associated with at least one of the conditions; and

for each of the plurality of conditions, storing the values of the road attribute in a conditional variant model associated with the condition, wherein storing the values of the road attribute in a conditional variant model associated with the condition comprises;

partitioning values of the road attribute into separate data buckets by explicit conditions; and

forming a conditional variant model for each respective explicit condition based on the values of the road attribute in that bucket.

42. The method of claim 41, wherein the road attribute is road speed.

43. The method of claim 42, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

44. The method of claim 42, wherein at least one of the plurality of conditions is from a group consisting of time of day and day of week.

45. The method of claim 41, wherein the road attribute is road safety.

46. The method of claim 45, wherein at least one of the plurality of conditions is from a group consisting of a level of user hurriedness, an environmental condition, and a level of traffic.

47. The method of claim 45, wherein at least one of the plurality of conditions is from a group consisting of time of day and day of week.

48. The method of claim 41, wherein the road attribute is road familiarity.

49. The method of claim 41, wherein the road attribute is disfavored roads.