Identification of traffic control mechanisms using machine learning

• US 10,210,751 B1
• Filed: 08/04/2017
• Issued: 02/19/2019
• Est. Priority Date: 08/04/2017
• Status: Active Grant

• Current Assignee: Verizon Patent and Licensing Incorporated (Verizon Communications Inc.)
• Original Assignee: Verizon Patent and Licensing Incorporated (Verizon Communications Inc.)
• Inventors: Eno, Joshua D., Storz, Tristan, O'Dwyer, John P., Williams, Lindy, Rasheed, Zeehasham, Brackney, Kevin
• Primary Examiner(s): Paige, Tyler D
• Application Number:
  US15/669,429
• Publication Number:
  US 20190043348A1
• Time in Patent Office: 564 Days
• Field of Search: 701119
• CPC Class Codes: G01C 21/3446 Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
• CPC Class Codes: G01C 21/3461 Preferred or disfavoured areas, e.g. dangerous zones, toll or emission zones, intersections, manoeuvre types, segments such as motorways, toll roads, ferries
• CPC Class Codes: G01C 21/3469 Fuel consumption; Energy use; Emission aspects
• CPC Class Codes: G01C 21/3492 employing speed data or traffic data, e.g. real-time or historical traffic control systems for road vehicles involving transmission of navigation instructions to the vehicle G08G1/0968
• CPC Class Codes: G01C 21/3629 Guidance using speech or audio output, e.g. text-to-speech text to speech systems per se G10L13/00
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G08G 1/012 from other sources than vehicle or roadside beacons, e.g. mobile networks
• CPC Class Codes: G08G 1/0125 Traffic data processing
• CPC Class Codes: G08G 1/0129 for creating historical data or processing based on historical data
• CPC Class Codes: G08G 1/0145 for active traffic flow control
• CPC Class Codes: G08G 1/052 with provision for determining speed or overspeed speed measuring in general G01P
• CPC Class Codes: G08G 1/056 with provision for distinguishing direction of travel
• CPC Class Codes: G08G 1/096811 where the route is computed offboard
• CPC Class Codes: G08G 1/096827 where the route is computed onboard
• CPC Class Codes: G08G 1/096872 where instructions are given per voice
• CPC Class Codes: G08G 1/096888 where input information is obtained using learning systems, e.g. history databases

Abstract

A device can receive a data model that has been trained on base map data and summary statistics data associated with a first geographic region. The device can obtain additional base map data associated with a second geographic region and additional summary statistics data for a set of junctions within the second geographic region. The device can determine traffic control mechanisms associated with the set of junctions by providing the additional base map data and the additional summary statistics data as input for the data model. The device can generate, using output of the data model, a base map that includes information indicating whether the set of junctions include traffic control mechanisms. The device can, after generating the base map, perform one or more actions associated with improving vehicle navigation or traffic management.

Claims

1. A device, comprising:

a memory; and

one or more processors to;

obtain base map data associated with a first geographic region,

the base map data including a set of values indicating attributes of a road map, and

the base map data not including traffic control mechanisms;

determine summary statistics data for a set of junctions within the first geographic region,

the summary statistics data including information associated with a set of vehicles traveling through the set of junctions;

train a data model using the base map data and the summary statistics data,

where the one or more processors, when training the data model, are to;

associate the base map data and the summary statistics data from a junction of the set of junctions with one or more training values,

 

the one or more training values including a likelihood of a traffic control mechanism being located at the junction,

 

the traffic control mechanism including a traffic light, a stop sign, or a yield sign; and

classify the junction as having a particular traffic control mechanism based on associating the base map data and the summary statistics data with the one or more training values;

obtain, after training the data model, additional base map data associated with a second geographic region;

determine additional summary statistics data for a set of junctions within the second geographic region;

determine traffic control mechanisms associated with the set of junctions within the second geographic region by providing the additional base map data and the additional summary statistics data as input for the data model;

generate, using output of the data model, a base map that includes information identifying the traffic control mechanisms at one or more junctions of the set of junctions included within the second geographic region; and

perform, after generating the base map, one or more actions associated with improving vehicle navigation or traffic management.

2. The device of claim 1, where the set of values indicating attributes of the road map include at least one of:

one or more values indicating geographic coordinates of the set of junctions included in the first geographic region,

one or more values indicating geographic coordinates of areas between the set of junctions included in the first geographic region,

one or more values indicating functional road classes,

one or more values indicating speed limits, or

one or more values indicating a number of lanes associated with the set of junctions included within the first geographic region.

3. The device of claim 1, where the summary statistics data for a junction of the set of junctions includes at least one of:

a value indicating an average vehicle stop time at the junction,

a value indicating an average vehicle speed at one or more geographic coordinates associated with the junction,

a value indicating a maximum vehicle speed at the junction,

a value indicating a minimum vehicle speed at the junction, or

a value indicating a total number of vehicles traveling through the junction in a particular direction.

4. The device of claim 1, where the one or more processors, when determining the summary statistics data, are to:

obtain observation data for the set of vehicles traveling through the set of junctions,

the observation data identifying one or more locations of the set of vehicles at one or more time periods,

determine path data for the set of vehicles by analyzing the observation data,

the path data including information indicating vehicle speed and information indicating vehicle direction, and

determine the summary statistics data by analyzing the path data for the set of vehicles.

5. The device of claim 1, where the one or more processors, when determining the traffic control mechanisms, are to:

provide the additional base map data and the additional summary statistics data as input for the data model to cause the data model to output a set of classifications for the set of junctions within the second geographic region,

the set of classifications indicating whether the set of junctions within the second geographic region include traffic control mechanisms; and

where the one or more processors, when generating the base map, are to;

generate a data structure indicative of the base map,

the data structure including metadata identifying the traffic control mechanisms at one or more junctions of the set of junctions within the second geographic region,

the metadata being based on the set of classifications indicating whether the set of junctions within the second geographic region include traffic control mechanisms.

6. The device of claim 1, where the one or more processors, when performing the one or more actions, are to:

receive a request from a user device for a set of navigational directions,

determine a set of navigational directions by using a path-finding algorithm to analyze the base map, and

provide the set of navigational directions to the user device.

7. The device of claim 1, where the one or more processors, when performing one or more actions associated with improving traffic management, are to:

analyze metadata for the set of junctions to determine whether metadata for one or more junctions of the set of junctions satisfies a threshold level of congestion; and

generate a recommendation to install or remove a traffic control mechanism based on the one or more junctions satisfying the threshold level of congestion.

8. A non-transitory computer-readable medium storing instructions, the instructions comprising:

one or more instructions that, when executed by one or more processors, cause the one or more processors to;

obtain base map data associated with a first geographic region,

the base map data including a set of values indicating attributes of a road map, and

the base map data not including traffic control mechanisms;

determine summary statistics data for a set of junctions within the first geographic region,

the summary statistics data including information associated with a set of vehicles traveling through the set of junctions;

train a data model using the base map data and the summary statistics data,

where the one or more instructions, that cause the one or more processors to train data model, cause the one or more processors to;

associate the base map data and the summary statistics data from a junction of the set of junctions with one or more training values,

 

the one or more training values including a likelihood of a traffic control mechanism being located at the junction,

 

the traffic control mechanism including a traffic light, a stop sign, or a yield sign; and

classify the junction as having a particular traffic control mechanism based on associating the base map data and the summary statistics data with the one or more training values;

obtain, after training data model, additional base map data associated with a second geographic region;

determine additional summary statistics data for a set of junctions within the second geographic region;

determine traffic control mechanisms associated with the set of junctions within the second geographic region by providing the additional base map data and the additional summary statistics data as input for the data model;

generate, using output of the data model, a base map that includes information indicating whether the set of junctions included within the second geographic region include traffic control mechanisms; and

perform, after generating the base map, one or more actions associated with improving vehicle navigation or traffic management.

9. The non-transitory computer-readable medium of claim 8, where the one or more instructions, that cause the one or more processors to determine the additional summary statistics data, cause the one or more processors to:

determine path data for the set of vehicles by analyzing observation data,

the observation data identifying one or more locations of the set of vehicles at one or more time periods, and

the path data identifying one or more paths traveled by the set of vehicles, and determine the summary statistics data by analyzing the path data for the set of vehicles.

10. The non-transitory computer-readable medium of claim 8, where the one or more instructions, when executed by the one or more processors, further cause the one or more processors to:

obtain test information associated with a set of junctions with known traffic control mechanisms;

provide the test information as input for the data model; and

determine whether output of the data model satisfies a threshold level of accuracy.

11. The non-transitory computer-readable medium of claim 10, where the one or more instructions, that cause the one or more processors to determine whether the output of the data model satisfies the threshold level of accuracy, cause the one or more processors to:

determine that the output of the data model does not satisfy a threshold level of accuracy; and

where the one or more instructions, when executed by the one or more processors, further cause the one or more processors to;

flag the one or more training values associated with determining the output of the data model as inaccurate training values,

modify the one or more training values that are flagged,

obtain additional base map data, additional path data, and/or additional summary statistics data, and

retrain the data model by associating the additional base map data, the additional path data, and/or the additional summary statistics data with the one or more modified training values.

12. The non-transitory computer-readable medium of claim 10, where the one or more instructions, that cause the one or more processors to perform the one or more actions, cause the one or more processors to:

receive a request from a user device for a set of navigational directions,

determine the set of navigational directions using a path-finding algorithm,

determine a destination arrival time for the set of navigational directions by analyzing base map metadata for one or more junctions used in the set of navigational directions, and

provide the destination arrival time to the user device.

13. The non-transitory computer-readable medium of claim 8, where the one or more instructions, that cause the one or more processors to perform the one or more actions, cause the one or more processors to:

receive a request from a user device for a set of navigational directions,

determine the set of navigational directions using a path-finding algorithm, and

provide the set of navigational directions to the user device to cause the user device to output the set of navigational directions using voice messages,

the voice messages including one or more traffic control mechanisms to be used as landmarks.

14. A method, comprising:

obtaining, by a device, base map data associated with a first geographic region,

the base map data including a set of values indicating attributes of a road map, and

the base map data not including traffic control mechanisms;

determining, by the device, summary statistics data for a set of junctions within the first geographic region,

the summary statistics data including information provided by one or more location aware devices associated with a set of vehicles traveling through the set of junctions;

training, by the device, a data model using the base map data and the summary statistics data,

where training the data model comprises;

associating the base map data and the summary statistics data from a junction of the set of junctions with one or more training values,

the one or more training values including a likelihood of a traffic control mechanism being located at the junction,

 

the traffic control mechanism including a traffic light, a stop sign, or a yield sign; and

classifying, by the device, the junction as having a particular traffic control mechanism based on associating the base map data and the summary statistics data with the one or more training values;

obtaining, by the device and after training the data model, additional base map data associated with a second geographic region;

determining, by the device, additional summary statistics data for a set of junctions within the second geographic region;

determining, by the device, traffic control mechanisms associated with the set of junctions within the second geographic region by providing the additional base map data and the additional summary statistics data as input for the data model;

generating, by the device, a base map that includes information indicating whether the set of junctions included within the second geographic region include traffic control mechanisms; and

performing, by the device, one or more actions associated with improving vehicle navigation or traffic management.

15. The method of claim 14, where determining the summary statistics data comprises:

obtaining observation data for the set of vehicles traveling through the set of junctions within the first geographic region,

the observation data identifying one or more locations of the set of vehicles at one or more time periods,

determining path data for the set of vehicles by analyzing the observation data,

the path data including information indicating vehicle speed and information indicating vehicle direction, and

determining the summary statistics data by analyzing the path data for the set of vehicles.

16. The method of claim 14, where determining the summary statistics data comprises:

determining summary statistics data for one or more vehicles of the set of vehicles,

the one or more vehicles being associated with vehicles traveling through the set of junctions within the first geographic region in a straight line path.

17. The method of claim 14, where training the data model comprises:

assigning weights to one or more values included in the base map data and one or more values included in the summary statistics data, and

where classifying the junction as having the particular traffic control mechanism comprises;

classifying the junction based on associating the one or more values included in the base map data and the one or more values included in the summary statistics data with the one or more training values and based on assigning weights to the one or more values included in the base map data and the summary statistics data.

18. The method of claim 14, where determining the traffic control mechanisms comprises:

determining the traffic control mechanisms by providing the additional base map data and the additional summary statistics data as input for the data model,

the data model to output a set of classifications for the set of junctions within the second geographic region,

the set of classifications indicating whether the set of junctions within the second geographic region include traffic control mechanisms; and

where generating the base map comprises;

generate the base map that includes the information indicating whether the set of junctions included within the second geographic region include traffic control mechanisms,

the information indicating whether the set of junctions within the second geographic region include traffic control mechanisms being based on the set of classifications indicating whether the set of junctions within the second geographic region include traffic control mechanisms.

19. The method of claim 14, where performing the one or more actions comprises:

analyzing metadata for the set of junctions to determine whether metadata for one or more junctions of the set of junctions satisfy a threshold,

generating a recommendation to install, remove, or make a modification to a traffic control mechanism based on determining whether the one or more junctions satisfy the threshold.

20. The method of claim 14, further comprising:

obtaining test information associated with a set of junctions with known traffic control mechanisms;

providing the test information as input for the data model;

determining whether output of the data model satisfies a threshold level of accuracy; and

retraining the data model based on whether the output of the data model satisfies the threshold level of accuracy.