System and method for providing an RNN-based human trust model

1. A computer-implemented method for providing an RNN-based human trust model, comprising:

receiving a plurality of inputs related to an autonomous operation of a vehicle and a driving scene of the vehicle;

analyzing the plurality of inputs to determine automation variables and scene variables, wherein crowd-sourced data associated with surveys that pertain to a driver's self-reported trust and the driver's self-reported reliability with respect to the autonomous operation of the vehicle is collected and analyzed;

outputting a short-term trust recurrent neural network state, from a short-term trust recurrent neural network, that captures an effect of the driver's experience with respect to an instantaneous vehicle maneuver and a long-term trust recurrent neural network state, from a long-term trust recurrent neural network, that captures the effect of the driver's experience with respect to the autonomous operation of the vehicle during a traffic scenario based on the automation variables, the scene variables, and the crowd-sourced data; and

predicting a take-over intent of the driver to take over control of the vehicle from an automated operation of the vehicle during the traffic scenario based on the short-term trust recurrent neural network state and the long-term trust recurrent neural network state,

wherein the output of the short-term trust recurrent neural network state from the short-term trust recurrent neural network is directly input to the long-term trust recurrent neural network that outputs the long-term trust recurrent neural network state.

2. The computer-implemented method of claim 1 , wherein receiving the plurality of inputs includes receiving dynamic data, image data, and LiDAR data, wherein the dynamic data, the image data, and the LiDAR data are analyzed to determine the traffic scenario in which the vehicle is operated.

3. The computer-implemented method of claim 1 , wherein the traffic scenario includes data that pertains to at least one traffic maneuver of the vehicle and at least one traffic configuration of the driving scene of the vehicle that takes place at a current time stamp.

4. The computer-implemented method of claim 1 , wherein the automation variables include a level of automation transparency that pertains to augmented reality cues that indicate information that are associated with particular autonomous functions that are occurring during the traffic scenario.

5. The computer-implemented method of claim 1 , wherein the automation variables include a level of automation reliability that is associated with a level of manual control the driver applies during the traffic scenario.

6. The computer-implemented method of claim 1 , wherein the scene variables include a level of risk and a level of scene difficultly, wherein the level of risk is based on a classification and a location of dynamic objects that are located within the driving scene during the traffic scenario and the level of scene difficultly includes environmental factors that influence the autonomous operation of the vehicle during the traffic scenario.

7. The computer-implemented method of claim 1 , wherein the scene variables include a previous take-over intent of the driver during a previous traffic scenario that is similar to the traffic scenario at a current time stamp, wherein the previous take-over intent includes data that pertains to a previous traffic scenario that includes at least one matching traffic maneuver of the vehicle and at least one matching traffic configuration of the driving scene of the vehicle to the traffic scenario in which the vehicle is currently operating.

8. The computer-implemented method of claim 1 , wherein the self-reported trust indicates a subjective indication of a level of trust with respect to the autonomous operation of a vehicle and the self-reported reliability indicates a subjective reliability that is associated with a level of manual control that the driver applies to operate the vehicle during the traffic scenario.

9. The computer-implemented method of claim 1 , wherein the prediction regarding the take-over intent is utilized to provide at least one of: a level of control to at least one system of the vehicle during the traffic scenario and a level of automation transparency during the traffic scenario.

10. A system for providing an RNN-based human trust model comprising:

a memory storing instructions when executed by a processor cause the processor to:

receive a plurality of inputs related to an autonomous operation of a vehicle and a driving scene of the vehicle;

analyze the plurality of inputs to determine automation variables and scene variables, wherein crowd-sourced data associated with surveys that pertain to a driver's self-reported trust and the driver's self-reported reliability with respect to the autonomous operation of the vehicle is collected and analyzed;

output a short-term trust recurrent neural network state, from a short-term trust recurrent neural network, that captures an effect of the driver's experience with respect to an instantaneous vehicle maneuver and a long-term trust recurrent neural network state, from a long-term trust recurrent neural network, that captures the effect of the driver's experience with respect to the autonomous operation of the vehicle during a traffic scenario based on the automation variables, the scene variables, and the crowd-sourced data; and

predict a take-over intent of the driver to take over control of the vehicle from an automated operation of the vehicle during the traffic scenario based on the short-term trust recurrent neural network state and the long-term trust recurrent neural network state,

wherein the output of the short-term trust recurrent neural network state from the short-term trust recurrent neural network is directly input to the long-term trust recurrent neural network that outputs the long-term trust recurrent neural network state.

11. The system of claim 10 , wherein receiving the plurality of inputs includes receiving dynamic data, image data, and LiDAR data, wherein the dynamic data, the image data, and the LiDAR data are analyzed to determine the traffic scenario in which the vehicle is operated.

12. The system of claim 10 , wherein the traffic scenario includes data that pertains to at least one traffic maneuver of the vehicle and at least one traffic configuration of the driving scene of the vehicle that takes place at a current time stamp.

13. The system of claim 10 , wherein the automation variables include a level of automation transparency that pertains to augmented reality cues that indicate information that are associated with particular autonomous functions that are occurring during the traffic scenario.

14. The system of claim 10 , wherein the automation variables include a level of automation reliability that is associated with a level of manual control the driver applies during the traffic scenario.

15. The system of claim 10 , wherein the scene variables include a level of risk and a level of scene difficultly, wherein the level of risk is based on a classification and a location of dynamic objects that are located within the driving scene during the traffic scenario and the level of scene difficultly includes environmental factors that influence the autonomous operation of the vehicle during the traffic scenario.

16. The system of claim 10 , wherein the scene variables include a previous take-over intent of the driver during a previous traffic scenario that is similar to the traffic scenario at a current time stamp, wherein the previous take-over intent includes data that pertains to a previous traffic scenario that includes at least one matching traffic maneuver of the vehicle and at least one matching traffic configuration of the driving scene of the vehicle to the traffic scenario in which the vehicle is currently operating.

17. The system of claim 10 , wherein the self-reported trust indicates a subjective indication of a level of trust with respect to the autonomous operation of a vehicle and the self-reported reliability indicates a subjective reliability that is associated with a level of manual control that the driver applies to operate the vehicle during the traffic scenario.

18. The system of claim 10 , wherein the prediction regarding the take-over intent is utilized to provide at least one of: a level of control to at least one system of the vehicle during the traffic scenario and a level of automation transparency during the traffic scenario.

19. A non-transitory computer readable storage medium storing instructions that when executed by a computer, which includes a processor perform a method, the method comprising:

receiving a plurality of inputs related to an autonomous operation of a vehicle and a driving scene of the vehicle;

analyzing the plurality of inputs to determine automation variables and scene variables, wherein crowd-sourced data associated with surveys that pertain to a driver's self-reported trust and the driver's self-reported reliability with respect to the autonomous operation of the vehicle is collected and analyzed;

outputting a short-term trust recurrent neural network state, from a short-term trust recurrent neural network, that captures an effect of the driver's experience with respect to an instantaneous vehicle maneuver and a long-term trust recurrent neural network state, from a long-term trust recurrent neural network, that captures the effect of the driver's experience with respect to the autonomous operation of the vehicle during a traffic scenario based on the automation variables, the scene variables, and the crowd-sourced data; and

predicting a take-over intent of the driver to take over control of the vehicle from an automated operation of the vehicle during the traffic scenario based on the short-term trust recurrent neural network state and the long-term trust recurrent neural network state,

wherein the output of the short-term trust recurrent neural network state from the short-term trust recurrent neural network is directly input to the long-term trust recurrent neural network that outputs the long-term trust recurrent neural network state.

20. The non-transitory computer readable storage medium of claim 19 , wherein the prediction regarding the take-over intent is utilized to provide at least one of: a level of control to at least one system of the vehicle during the traffic scenario and a level of automation transparency during the traffic scenario.