Systems and methods for distracted driving detection

1. A computer-implemented method for a host vehicle to detect driving distraction, comprising:

receiving proximate vehicle data about a proximate vehicle proximate to the host vehicle;

estimating one or more baselines for a predetermined future time for the proximate vehicle from the proximate vehicle data;

comparing current kinematic data of the proximate vehicle data for the predetermined future time to the one or more baselines;

generating one or more distraction flags associated with the proximate vehicle based on the comparison, wherein a distraction flag, of the one or more distraction flags, is associated with a type of distraction flag, and wherein there are a plurality of types of distraction flag including an offset flag based on a position of the proximate vehicle and a variance flag based on a predicted position of the proximate vehicle and the position of the proximate vehicle at the predetermined future time;

calculating a distraction probability based on the one or more distraction flags; and

controlling one or more vehicle systems of the host vehicle based on the distraction flags in response to the distraction probability satisfying a probability threshold.

2. The computer-implemented method of claim 1 , wherein estimating the one or more baselines includes applying the proximate vehicle data to a motion model to predict behavior of the proximate vehicle.

3. The computer-implemented method of claim 2 , wherein the proximate vehicle data is applied to the motion model using a model fitting method.

4. The computer-implemented method of claim 1 , wherein a distraction flag is identified when the current kinematic data exceeds the one or more baselines by a distraction threshold.

5. The computer-implemented method of claim 4 , wherein the offset flag is based on a distance measurement, the current kinematic data is a position measurement, and the baseline includes a lane offset value.

6. The computer-implemented method of claim 4 , wherein the one or more vehicle systems are controlled to change operation of the host vehicle associated with the current kinematic data to satisfy the distraction threshold.

7. The computer-implemented method of claim 1 , wherein the one or more baselines include a Lane Offset Value (LOV) baseline that predicts a path of the proximate vehicle in a lane.

8. The computer-implemented method of claim 7 , wherein the current kinematic data includes a distance measurement between an observed proximate vehicle and the predicted path.

9. The computer-implemented method of claim 1 , wherein the one or more baselines include a Lane Offset Error (LOE) baseline that predicts variance in predicted behavior of the proximate vehicle, and wherein the variance flag is based on the LOE error.

10. A system for distracted driving detection, comprising:

a processor;

a data receiving module, implemented via the processor, configured to receive proximate vehicle data about a proximate vehicle proximate to a host vehicle;

an estimation module, implemented via the processor, configured to estimate one or more baselines for a predetermined future time for the proximate vehicle from the proximate vehicle data;

a determination module, implemented via the processor, configured to:

generate one or more distraction flags for the proximate vehicle based on current kinematic data about the proximate vehicle at the predetermined future time and the one or more baselines, wherein a distraction flag, of the one or more distraction flags, is associated with a type of distraction flag, and wherein there are a plurality of types of distraction flag including an offset flag based on a position of the proximate vehicle and a variance flag based on a predicted position of the proximate vehicle and the position of the proximate vehicle at the predetermined future time, and

calculate a distraction probability based on the one or more distraction flags; and

a control module, implemented via the processor, configured to control one or more vehicle systems of the host vehicle based on the identified distraction flags in response to the distraction probability satisfying a probability threshold.

11. The system of claim 10 , wherein estimating the one or more baselines includes applying the proximate vehicle data to a motion model to predict behavior of the proximate vehicle.

12. The system of claim 11 , wherein the proximate vehicle data is applied to the motion model using a model fitting method.

13. The system of claim 10 , wherein a distraction flag is identified when the current kinematic data exceeds the one or more baselines by a distraction threshold.

14. The system of claim 13 , wherein the offset flag is based on a distance measurement, the current kinematic data is a position measurement, and the baseline includes a lane offset value.

15. A non-transitory computer-readable storage medium storing instructions that, when executed by a computer, causes the computer to perform a method comprising:

receiving proximate vehicle data about a proximate vehicle proximate to a host vehicle;

estimating a Lane Offset Value (LOV) baseline that predicts a path of the proximate vehicle in a lane;

estimating a Lane Offset Error (LOE) baseline that predicts variance in predicted behavior of the proximate vehicle;

comparing current kinematic data of the proximate vehicle data to the LOV baseline and the LOE baseline;

generating one or more distraction flags associated with the proximate vehicle based on the comparison, wherein a distraction flag, of the one or more distraction flags, is associated with a type of distraction flag, and wherein there are a plurality of types of distraction flag including an offset flag based on a position of the proximate vehicle relative to the LOV baseline and a variance flag based on a predicted position of the proximate vehicle and the position of the proximate vehicle at the predetermined future time based on the LOE baseline;

calculating a distraction probability based on the distraction flags; and

controlling one or more vehicle systems of the host vehicle based on the distraction probability in response to the distraction probability satisfying a probability threshold.

16. The non-transitory computer-readable storage medium of claim 15 , wherein the LOV baseline or the LOE baseline is estimated using a motion model.

17. The non-transitory computer-readable storage medium of claim 16 , wherein the motion model is one of a constant velocity model, constant acceleration model, constant steering angle and velocity model, constant turn rate and velocity model, constant turn rate and acceleration model, and a constant curvature and acceleration model.