Event prediction system, event prediction method, program, and recording medium having same recorded therein
An event prediction system includes an accumulation unit and a generator. The accumulation unit accumulates a plurality of pieces of learning data each including history information representing a situation of a mobile object upon occurrence of an event associated with driving of the mobile object. The generator generates a prediction model for prediction of relative coordinates of an occurrence place of the event relative to the mobile object by using the plurality of pieces of learning data. Each of the plurality of pieces of learning data further includes label information representing the relative coordinates of the occurrence place of the event relative to the mobile object.
1. An event prediction system comprising:
a server device configured to
accumulate a plurality of pieces of learning data each including history information representing a situation of a mobile object at occurrence of an event associated with driving of the mobile object, and
generate a first prediction model for prediction of relative coordinates of an occurrence place of the event relative to the mobile object by using the plurality of pieces of learning data; and
a processor installed in the mobile object, the processor configured to
predict the relative coordinates of the occurrence place of the event relative to the mobile object by using the first prediction model for prediction of relative coordinates of an occurrence place of the event relative to the mobile object and prediction information concerning the mobile object,
wherein the processor calculates a distance from the mobile object to the place indicated by the relative coordinates predicted, estimates the occurrence place of the event when the distance is equal to or less than a specified distance, and does not estimate the occurrence place of the event when the distance is greater than the specified distance.
2. The event prediction system according to claim 1 , wherein the history information includes at least one of information concerning an object adjacent to the mobile object, information concerning a condition of the mobile object, and information concerning a position of the mobile object.
3. The event prediction system according to claim 1 , further comprising a display configured to notify of a prediction result on the event based on the relative coordinates predicted by the processor.
4. The event prediction system according to claim 3 , wherein the display notifies of the occurrence place of the event estimated from the relative coordinates as a prediction result on the event.
5. The event prediction system according to claim 3 , wherein the display is configured to provide a display of the prediction result on the event.
6. The event prediction system according to claim 1 , wherein the processor is configured to use the first prediction model different for an attribute of each driver who drives the mobile object.
7. An event prediction method comprising:
accumulating a plurality of pieces of learning data each including history information representing a situation of a mobile object at occurrence of an event associated with driving of the mobile object;
generating a first prediction model for prediction of relative coordinates of an occurrence place of the event relative to the mobile object by using the plurality of pieces of learning data; and
predicting the relative coordinates of the occurrence place of the event relative to the mobile object by using the first prediction model for prediction of relative coordinates of an occurrence place of the event relative to the mobile object and prediction information concerning the mobile object,
wherein calculating a distance from the mobile object to the place indicated by the relative coordinates predicted, estimating the place where the event when the distance is equal to or less than a specified distance, and not estimating the occurrence place of the event when the distance is greater than the specified distance.
8. A non-transient recording medium storing a program for causing a computer system to execute
an accumulation process of accumulating a plurality of pieces of learning data each including history information representing a situation of a mobile object at occurrence of an event associated with driving of the mobile object,
a generation process of generating a first prediction model for prediction of relative coordinates of an occurrence place of the event relative to the mobile object by using the plurality of pieces of learning data, and
a coordinate prediction process of predicting the relative coordinates of the occurrence place of the event relative to the mobile object by using the first prediction model for prediction of relative coordinates of an occurrence place of the event relative to the mobile object and prediction information concerning the mobile object,
wherein the coordinate prediction process of calculating a distance from the mobile object to the place indicated by the relative coordinates predicted, estimating the place where the occurrence place of the event when the distance is equal to or less than a specified distance, and not estimating the place where the occurrence place of the event when the distance is greater than the specified distance.