Autonomous vehicles implementing user-based decision making

1 . A system comprising:

a processor and a memory coupled thereto, wherein the processor is configured to:

identify at least one entity of a plurality of entities of a self-driving vehicle;

obtain a preference of the at least one entity, wherein the preference is synthesized based on responses by the at least one entity to a questionnaire;

while the self-driving vehicle is driving, select, based on the preference of the at least one entity, a driving action from a plurality of potential actions each associated with a different potential alternative outcome; and

cause the self-driving vehicle to perform the selected driving action automatically, without relying on user input.

2 . The system of claim 1 , wherein said system is configured to provide a personalized ethical decision for the self-driving vehicle that is modified based on the preferences of the at least one entity that is controlling the self-driving vehicle at the time a decision is being made.

3 . The system of claim 1 , wherein the action selection is configured to be invoked in response to an identification of an impending accident by the self-driving vehicle while the self-driving vehicle is driving, wherein the plurality of potential actions are potential actions by the self-driving vehicle that would not prevent the impending accident and would result in different outcomes that are not Pareto improvement of one another.

4 . The system of claim 1 , wherein the preference is a preference function to compare outcomes, the preference function is configured to take into account at least one of:

demographic information of a person that is expected to be injured in at least one of the outcomes;

behavioral information regarding an activity of a person that is expected to be injured in at least one of the outcomes;

information relating to a non-human animal that is expected to be hurt in at least one of the outcomes; or

an expected potential harm to a passenger of the self-driving vehicle in at least one of the outcomes.

5 . The system of claim 1 , wherein the at least one entity is a controlling entity of the self-driving vehicle, wherein said system is configured to identify the controlling entity of the self-driving vehicle and update the preference accordingly.

6 . The system of claim 1 , wherein the at least one entity is a group of passengers of the self-driving vehicle, wherein the preference is a joint preference that is determined based on responses of each individual passenger in the group of passengers to a questionnaire.

7 . The system of claim 1 , wherein the preference for the self-driving vehicle is modified based on pre-defined timeslots, wherein the pre-defined timeslots are associated with different entities.

8 . A method comprising:

identifying at least one entity of a plurality of entities of a self-driving vehicle;

obtaining a preference of the at least one entity, wherein the preference is synthesized based on responses by the at least one entity to a questionnaire;

while the self-driving vehicle is driving, selecting, based on the preference of the at least one entity, a driving action for a self-driving vehicle from a plurality of potential actions, wherein each potential action of the plurality of potential actions is associated with a different potential alternative outcome; and

causing the self-driving vehicle to perform the selected driving action automatically, without relying on user input.

9 . The method of claim 8 further comprises: determining that the at least one entity is a controlling entity of the self-driving vehicle at a time the selection is made.

10 . The method of claim 8 , wherein said selecting is performed in response to an identification of an impending accident by the self-driving vehicle while the self-driving vehicle is driving, wherein the plurality of potential actions are potential actions by the self-driving vehicle that would not prevent the impending accident and would result in different outcomes that are not Pareto improvement of one another.

11 . The method of claim 8 , wherein the preference is a preference function to compare outcomes that is configured to take into account at least one of:

demographic information of a person that is expected to be injured in at least one of the outcomes;

behavioral information regarding an activity of a person that is expected to be injured in at least one of the outcomes;

information relating to a non-human animal that is expected to be hurt in at least one of the outcomes; or

an expected potential harm to a passenger of the self-driving vehicle in at least one of the outcomes.

12 . The method of claim 8 , wherein the at least one entity is a passenger of the self-driving vehicle.

13 . The method of claim 8 further comprises updating the preference for the self-driving vehicle in response to a passenger disembarking from the self-driving vehicle.

14 . The method of claim 8 , wherein the at least one entity is a group of passengers of the self-driving vehicle, wherein the preference for the self-driving vehicle is determined based on a joint preference, wherein the joint preference is determined based on responses of each individual passenger in the group of passengers to a questionnaire.

15 . The method of claim 8 , wherein the preference for the self-driving vehicle is modified based on pre-defined timeslots, wherein the pre-defined timeslots are associated with different entities.

16 . A computer program product comprising a non-transitory computer readable storage medium retaining program instructions, which program instructions when read by a processor, cause the processor to perform a method comprising:

identifying at least one entity of a plurality of entities of a self-driving vehicle;

obtaining a preference of the at least one entity, wherein the preference is synthesized based on responses by the at least one entity to a questionnaire;

while the self-driving vehicle is driving, selecting, based on the preference of the at least one entity, a driving action for a self-driving vehicle from a plurality of potential actions each associated with a different potential alternative outcome; and

causing the self-driving vehicle to perform the selected driving action automatically, without relying on user input.

17 . The computer program product of claim 16 , wherein the program instructions further cause the processor to perform: in response to a determination that a second entity of the plurality of entities is controlling the self-driving vehicle, updating the preference to be used for the selection of the action.

18 . The computer program product of claim 16 , wherein the action selection is performed in response to an identification of an impending accident by the self-driving vehicle while the self-driving vehicle is driving, wherein the plurality of potential actions are potential actions by the self-driving vehicle that would not prevent the impending accident and would result in different outcomes that are not Pareto improvement of one another.