System and/or method for predicting user action in connection with a vehicle user interface using machine learning
Disclosed are a system, method and system to predict an individual's action in connection with a vehicle user interface using machine learning. One or more models may be developed based, at least in part, on observations of past actions by the individual among the plurality of target actions by the individual. Extracted features of a current context may be applied to the developed one or more models to predict a subsequent action among the target actions by the individual.
1. A system to be disposed in a vehicle, the system comprising:
one or more memory devices; and
one or more processors coupled to the one or more memory devices to:
determine features indicative of a context in which a driver is currently operating the vehicle;
determine a plurality of available vehicle user interface actions for the context;
calculate a posterior probability for each action in the plurality of available vehicle user interface actions;
generate a prediction of a future vehicle user interface action based, at least in part, on the posterior probability for each action in the plurality of available vehicle user interface actions; and
cause a user interface in the vehicle to generate an output based on the prediction of the future vehicle user interface action.
2. The system of claim 1 , wherein the features include a time of day.
3. The system of claim 1 , wherein the features include a day of the week.
4. The system of claim 1 , wherein the features include a longitude value and a latitude value.
5. The system of claim 1 , wherein the prediction of the future vehicle user interface action is further generated based on the features.
6. The system of claim 1 , wherein the prediction of the future vehicle user interface action is further generated based on user action context parameters relating past vehicle user interface actions requested by the driver.
7. The system of claim 6 , wherein the past vehicle user interface actions are based on a reliability weight.
8. The system of claim 7 , wherein the reliability weight is computed based on past predictions.
9. The system of claim 7 , wherein the reliability weight is computed based on actual observed vehicle user interface actions.
10. The system of claim 7 , wherein the reliability weight is conditioned on the features.
11. The system of claim 1 , wherein the prediction of the future vehicle user interface action is further generated based on a sum of probabilities based on an established model weighted according to computed reliability weights.
12. The system of claim 1 , wherein for each of the plurality of available vehicle user interface actions the one or more processors are further configured to:
compute a probability of the features indicative of a contemporaneous context;
sum computed probabilities of the contemporaneous context conditioned on the plurality of available vehicle user interface actions; and
determine the parameters indicative of posterior probability conditioned on the features indicative of the contemporaneous context based, at least in part, on the summed computed probabilities.
13. The system of claim 12 , wherein the one or more processors are further configured to:
identify a plurality of context attributes of the features indicative of the context in which the driver is currently operating the vehicle;
establish a model for each of the plurality of context attributes of the features of the context in which the driver is currently operating the vehicle; and
determine computed probabilities of the features indicative of the context in which the driver is currently operating the vehicle conditioned on the plurality of available vehicle user interface actions based, at least on in part, on the models of the context attributes of the features indicative of the context in which the driver is currently operating the vehicle.
14. The system of claim 13 , wherein the one or more processors are further configured to:
for each established model, compute a reliability weight based, at least in part, on past predictions and associated detected actual observed vehicle user interface actions; and
for at least one of the plurality of available vehicle user interface actions, determine a predicted probability of at least one of the plurality of available vehicle user interface actions based, at least in part, on a sum of probabilities based on the established model weighted according to computed reliability weights.
15. The system of claim 1 , wherein the one or more processors are further configured to:
receive updated features indicative of the context in which the vehicle is currently operating;
determine whether a predetermined amount of time has elapsed since the prediction was generated; and
responsive to such a determination that the predetermined amount of time has elapsed, generate an update of the prediction of the future vehicle user interface action.
16. The system of claim 15 , wherein the one or more processors are further configured to:
provide signals the user interface to present an update of options to the driver based, at least in part, on the update of the prediction of the future vehicle user interface action.
17. The system of claim 16 , wherein the one or more processors are further configured to:
cause an output device of the vehicle to present one or more selectable updated options based, at least in part, on the update of the prediction of the future vehicle user interface action.
18. The system of claim 1 , wherein the posterior probability for each action in the plurality of available vehicle user interface actions is based, at least in part, on a Bayes model.
19. A computer-implemented method for training a model for predicting a future vehicle user interface action, the method comprising:
determine features indicative of a context in which a driver is currently operating the vehicle;
determine a plurality of available vehicle user interface actions for the context;
calculate a posterior probability for each action in the plurality of available vehicle user interface actions;
generate a prediction of the future vehicle user interface action based, at least in part, on the posterior probability for each action in the plurality of available vehicle user interface actions; and
cause a user interface in the vehicle to generate an output based on the prediction of the future vehicle user interface action.
20. An article comprising a non-transitory storage medium comprising computer-readable instructions stored thereon, the instructions to be executable by one or more processors to:
determine features indicative of a context in which a driver is currently operating the vehicle;
determine a plurality of available vehicle user interface actions for the context;
calculate a posterior probability for each action in the plurality of available vehicle user interface actions;
generate a prediction of the future vehicle user interface action based, at least in part, on the posterior probability for each action in the plurality of available vehicle user interface actions; and
cause a user interface in the vehicle to generate an output based on the prediction of the future vehicle user interface action.