Road network optimization based on vehicle telematics information

1 . A computer-implemented method comprising:

obtaining geo-rectified aerial images representing real-world locations using an unmanned aerial vehicle (UAV);

generating a road network model from the aerial images and one or more of 3-dimensional models, point cloud data, and geo-rectified imagery;

generating one or more area segments based on the road network model;

obtaining historical event information associated with one or more uniquely identified vehicles, the historical event information comprising sensor-derived telematics data including at least vehicle speed, braking, fuel consumption, and suspension state, for vehicles that have traversed the one or more area segments;

correlating the historical telematics event information with the corresponding area segments of the UAV-derived road network model;

determining from the correlated telematics event information whether traversal of the one or more area segments satisfies a vehicle traversal criteria;

generating, based on the traversal criteria, a UAV flight plan configured to direct the UAV to perform an inspection of one or more area segments; and

transmitting the flight plan to the UAV for execution.

2 . The method of claim 1 , further comprising:

determining the vehicle traversal criteria for one or more of the one or more area segments.

3 . The method of claim 2 , wherein the unique identifier comprises a vehicle ID associated with a telematics event stream.

4 . The method of claim 1 , wherein the historical event information for a vehicle includes one or more of the following: time of an event, geo-spatial location of the event, duration of the event, speed of the vehicle, fuel consumption of the vehicle, gear position of the vehicle, brake application, vehicle load information, shock absorber data, acceleration data, deceleration data.

5 . The method of claim 1 , further comprising:

determining the vehicle traversal criteria based on area geometry for a particular area segment of the one or more area segments.

6 . The method of claim 5 , wherein the area geometry for a particular area segment is defined by one or more parameters including: cross-falls, center line, superelevation, road width, grade, radius, switch backs, 2-ways, n-ways, or rimpull curves.

7 . The method of claim 1 , further comprising:

generating a flight plan for a UAV, the flight plan based upon an inspection area associated with a area segment or with a location of a determined area condition; and

providing the flight plan to the UAV or to a ground control station operable to control the UAV and perform the flight plan.

8 . The method of claim 1 , further comprising:

generating instructions for the one or more vehicles to obtain sensor data describing actual physical area conditions;

transmitting the instructions to the one or more vehicles;

receiving sensor data obtained from the one or more vehicles, wherein the sensor data is digital images, or LIDAR data; and

updating portions of the area network from the received sensor data.

9 . The method of claim 1 , further comprising:

training a visual classifier with a data set of images of actual physical area conditions;

receiving images different from the data set of images, wherein the different images depict a portion of an area associated with an area network; and

identifying, with the visual classifier, a type of area condition found in the different images.

10 . The method of claim 2 , wherein the traversal criteria comprises a predetermined traversal time for a particular type of vehicle, and wherein each type of vehicle has a particular traversal time for a segment.

11 . The method of claim 1 , wherein the one or more vehicles have traversed the one or more area segments, and the one or more vehicles have obtained operational data, the operational data comprising geo-spatial positions, time, and event data of respective ones of the one or more vehicles.

12 . The method of claim 1 , wherein executing the UAV flight plan comprises autonomously navigating the UAV along a predefined flight path over the one or more area segments and capturing inspection imagery or LIDAR data during execution of the flight plan.

13 . A system for optimizing road network analysis, the system comprising:

an unmanned aerial vehicle (UAV) including one or more cameras configured to capture geo-rectified aerial images of real-world locations;

a processing subsystem configured to:

generate a road network model from the aerial images and one or more of three-dimensional models, point cloud data, and geo-rectified imagery;

segment the road network model into a plurality of area segments;

obtain historical telematics event information associated with a plurality of uniquely identified vehicles that have traversed the area segments, the telematics event information including at least speed, braking, fuel consumption, and suspension state;

correlate the telematics event information with the corresponding area segments of the UAV-derived road network model;

determine whether traversal of the area segments satisfies one or more traversal criteria;

generate a UAV flight plan based on a detected road condition; and

transmit the flight plan to the UAV for execution; and

a communications interface configured to provide updated road network data, traversal criteria, or UAV flight plans to a ground control station or vehicle system.

14 . The system of claim 13 , wherein the UAV further comprises a LIDAR sensor, and wherein the road network model is generated based in part on LIDAR data.

15 . The system of claim 13 , wherein the traversal criteria comprises a predetermined traversal time associated with a particular vehicle type, and wherein different vehicle types are associated with different traversal times.

16 . The system of claim 13 , wherein the processing subsystem is further configured to train a classifier with images of road conditions and apply the classifier to UAV aerial imagery to automatically identify area conditions.

17 . An apparatus comprising:

a computer readable storage media; and

program instructions stored on the computer readable storage media which, when executed by a processing system, direct the processing system to:

obtain geo-rectified aerial images representing real-world locations using an unmanned aerial vehicle (UAV);

generate a road network model from the aerial images and one or more of 3-dimensional models, point cloud data, and geo-rectified imagery;

generate one or more area segments based on the road network model;

obtain historical event information associated with one or more uniquely identified vehicles, the historical event information comprising sensor-derived telematics data including at least vehicle speed, braking, fuel consumption, and suspension state, for vehicles that have traversed the one or more area segments;

correlate the historical telematics event information with the corresponding area segments of the UAV-derived road network model;

determine from the correlated telematics event information whether traversal of the one or more area segments satisfies a vehicle traversal criteria;

generate based on the traversal criteria, a UAV flight plan configured to direct the UAV to perform an inspection of one or more area segments;

transmit the flight plan to the UAV for execution.

18 . The apparatus of claim 17 , wherein the program instructions, when executed by the processing system, further direct the processing system to:

determine the vehicle traversal criteria for one or more of the one or more area segments.

19 . The apparatus of claim 18 , wherein the unique identifier comprises a vehicle ID associated with a telematics event stream.

20 . The apparatus of claim 17 , wherein the program instructions, when executed by the processing system, further direct the processing system to:

determine the vehicle traversal criteria based on area geometry for a particular area segment of the one or more area segments.