DETECTION OF DRIVING ACTIONS THAT MITIGATE RISK

1 . A method comprising:

detecting, by at least one processor and using visual data from a camera coupled to a first vehicle, a second vehicle ahead of the first vehicle;

determining, by the at least one processor, a type of the second vehicle;

estimating, by the at least one processor, a distance between the first vehicle and the second vehicle based on observed dimensions of the second vehicle in the visual data and known average vehicle dimensions for the type of the second vehicle;

monitoring, by the at least one processor over a time interval, whether a driver of the first vehicle maintains a safe following distance from the second vehicle based on the estimated distance;

determining, by the at least one processor, that the driver satisfied one or more positive driving criteria during the time interval, wherein the one or more positive driving criteria comprise maintaining the safe following distance from the second vehicle; and

generating, by the at least one processor, a positive recognition notification based on the determination that the driver satisfied the one or more positive driving criteria.

2 . The method of claim 1 , wherein the estimated distance is based on the type of the second vehicle but not a make or a model of the second vehicle.

3 . The method of claim 2 , wherein the type of the second vehicle is one of: a sedan, a van, a semi-truck, a sport utility vehicle, or a motorcycle.

4 . The method of claim 1 , wherein estimating the distance comprises:

computing the observed dimensions of the second vehicle in pixel coordinates in the visual data; and

estimating the distance based on a relationship between the observed dimensions in the visual data and the known average vehicle dimensions for the type in real-world coordinates.

5 . The method of claim 1 , wherein monitoring whether the driver maintains the safe following distance comprises:

converting the estimated distance from a unit of distance to a unit of time based on a speed of the first vehicle; and

determining whether the estimated distance in the unit of time exceeds a predetermined time threshold.

6 . The method of claim 1 , further comprising:

determining, by the at least one processor, whether the driver maintains proper lane position relative to lane markings; and wherein determining that the driver satisfied the one or more positive driving criteria further comprises determining that the driver maintained proper lane position.

7 . The method of claim 6 , wherein determining whether the driver maintains proper lane position comprises:

determining whether the first vehicle maintains an offset position relative to the second vehicle when the second vehicle is in an adjacent lane.

8 . The method of claim 1 , further comprising:

determining, by the at least one processor, whether the driver uses turn signal indicators when changing lanes; and wherein determining that the driver satisfied the one or more positive driving criteria further comprises determining that the driver used turn signal indicators when changing lanes.

9 . The method of claim 8 , wherein determining whether the driver uses turn signal indicators comprises at least one of:

accessing turn signal data via an OBD-II interface of the first vehicle; or

detecting, using a microphone, a sound of a turn signal indicator of the first vehicle.

10 . The method of claim 1 , wherein the time interval is defined based on satisfaction of the one or more positive driving criteria rather than solely on an absence of negative driving events.

11 . The method of claim 1 , further comprising:

transmitting, by the at least one processor, data to a remote device, wherein the data comprises visual data captured by the camera and an indication that the driver satisfied the one or more positive driving criteria during the time interval.

12 . The method of claim 11 , wherein the visual data is used to train an autonomous vehicle driving system.

13 . A system comprising:

a camera configured to be coupled to a first vehicle;

at least one processor; and

memory storing instructions that, when executed by the at least one processor, cause the system to:

detect, using visual data from the camera, a second vehicle ahead of the first vehicle;

determine a type of the second vehicle;

estimate a distance between the first vehicle and the second vehicle based on observed dimensions of the second vehicle in the visual data and known average vehicle dimensions for the type of the second vehicle;

monitor, over a time interval, whether a driver of the first vehicle maintains a safe following distance from the second vehicle based on the estimated distance;

determine that the driver satisfied one or more positive driving criteria during the time interval, wherein the one or more positive driving criteria comprise maintaining the safe following distance from the second vehicle; and

generate a positive recognition notification based on the determination that the driver satisfied the one or more positive driving criteria.

14 . The system of claim 13 , wherein the estimated distance is based on the type of the second vehicle but not a make or a model of the second vehicle.

15 . The system of claim 14 , wherein the type of the second vehicle is one of: a sedan, a van, a semi-truck, a sport utility vehicle, or a motorcycle.

16 . The system of claim 13 , wherein the instructions further cause the system to:

convert the estimated distance from a unit of distance to a unit of time based on a speed of the first vehicle; and

determine whether the estimated distance in the unit of time exceeds a predetermined time threshold.

17 . The system of claim 13 , further comprising at least one of:

an OBD-II interface for accessing turn signal data; or

a microphone for detecting a sound of a turn signal indicator of the first vehicle.

18 . A non-transitory computer-readable storage medium having instructions stored thereon that, when executed by at least one processor, cause the at least one processor to perform operations comprising:

detecting, using visual data from a camera coupled to a first vehicle, second vehicle ahead of the first vehicle;

determining a type of the second vehicle;

estimating a distance between the first vehicle and the second vehicle based on observed dimensions of the second vehicle in the visual data and known average vehicle dimensions for the type of the second vehicle;

monitoring, over a time interval, whether a driver of the first vehicle maintains a safe following distance from the second vehicle based on the estimated distance;

determining that the driver satisfied one or more positive driving criteria during the time interval, wherein the one or more positive driving criteria comprise maintaining the safe following distance from the second vehicle; and

generating a positive recognition notification based on the determination that the driver satisfied the one or more positive driving criteria.

19 . The non-transitory computer-readable storage medium of claim 18 , wherein the estimated distance is based on the type of the second vehicle but not a make or a model of the second vehicle.

20 . The non-transitory computer-readable storage medium of claim 18 , wherein the operations further comprise:

transmitting data to a remote device, wherein the data comprises visual data captured by the camera and an indication that the driver satisfied the one or more positive driving criteria, and wherein the visual data is used to train an autonomous vehicle driving system.