External data collection based on predicted data insufficiency

1 . A computer-implemented method comprising:

directing, for a first autonomous vehicle, an action in response to external data;

collecting contextual data about an environment in which the first autonomous vehicle is operating;

analyzing historical actions associated with prior similar contextual situations of the first autonomous vehicle;

correlating analysis of the historical actions with both the action and the collected contextual data;

predicting, based on the correlation, an insufficiency of the external data for performing the action; and

responsive to prediction of the insufficiency of the external data, directing the first autonomous vehicle to collect, from a second data source, additional external data to mitigate the insufficiency, wherein the second data source is a data collection device comprising a second autonomous vehicle, and wherein directing collection of the additional external data comprises deploying the second autonomous vehicle from a base station located on or within the first autonomous vehicle and receiving the additional external data from the second autonomous vehicle.

2 . The computer-implemented method of claim 1 , wherein predicting the insufficiency of the external data for performing the action comprises:

analyzing the action to be performed in response to the external data and similar historical actions;

correlating, based on the analyzing, the action with the contextual data about the environment that the first autonomous vehicle is operating;

generating, based on the correlating, a digital twin simulation of the first autonomous vehicle performing the action; and

determining, based on the digital twin simulation, that the action cannot be safely completed based on the external data and contextual data.

3 . The computer-implemented method of claim 1 , wherein directing collection of additional external data from the second data source comprises:

determining, based on the predicting, a type of data required for mitigating the insufficiency of external data; and

identifying, based on the determined type of data, the second data source from a plurality of data sources.

4 . The computer-implemented method of claim 1 , further comprising:

correlating the collected additional external data with the external data used to determine the action;

modifying, based on the correlating, the action to a second action, wherein the second action is different than the action; and

directing the first autonomous vehicle to perform the second action.

5 . The computer-implemented method of claim 4 , further comprising:

sharing the additional external data from the second data source and/or the second action with a third autonomous vehicle, wherein the third autonomous vehicle modifies one or more actions based on the shared data.

6 . A system comprising:

a processor; and

a computer-readable storage medium communicatively coupled to the processor and storing program instructions which, when executed by the processor, cause the processor to perform a method comprising:

directing, for a first autonomous vehicle, an action in response to external data;

collecting contextual data about an environment in which the autonomous vehicle is operating;

analyzing historical actions associated with prior similar contextual situations of the first autonomous vehicle;

correlating analysis of the historical actions with both the action and the collected contextual data;

predicting, based on the correlation, an insufficiency of external data for performing the action; and

responsive to prediction of the insufficiency of the external data, directing the first autonomous vehicle to collect, from a second data source, additional external data to mitigate the insufficiency, wherein the second data source is a data collection device comprising a second autonomous vehicle, and wherein directing collection of the additional external data comprises deploying the second autonomous vehicle from a base station located on or within the first autonomous vehicle and receiving the additional external data from the second autonomous vehicle.

7 . The system of claim 6 , wherein predicting the insufficiency of the external data for performing the action comprises:

analyzing the action to be performed in response to the external data and similar historical actions;

correlating, based on the analyzing, the action with the contextual data about the environment that the first autonomous vehicle is operating;

generating, based on the correlating, a digital twin simulation of the first autonomous vehicle performing the action; and

determining, based on the digital twin simulation, that the action cannot be safely completed based on the external data and contextual data.

8 . The system of claim 6 , wherein directing collection of additional external data from the second data source comprises:

determining, based on the predicting, a type of data required for mitigating the insufficiency of external data; and

identifying, based on the determined type of data, the second data source from a plurality of data sources.

9 . The system of claim 6 , wherein the method performed by the processor further comprises:

correlating the collected additional external data with the external data used to determine the action;

modifying, based on the correlating, the action to a second action, wherein the second action is different than the action; and

directing the first autonomous vehicle to perform the second action.

10 . The system of claim 9 , wherein the method performed by the processor further comprises:

sharing the additional external data from the second data source and/or the second action with a third autonomous vehicle, wherein the third autonomous vehicle modifies one or more actions based on the shared data.

11 . A computer program product comprising a computer-readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising:

directing, for a first autonomous vehicle, an action in response to external data;

collecting contextual data about an environment in which the first autonomous vehicle is operating;

analyzing historical actions associated with prior similar contextual situations of the first autonomous vehicle;

correlating analysis of the historical actions with both the action and the collected contextual data;

predicting, based on the correlation, an insufficiency of external data for performing the action; and

responsive to prediction of the insufficiency of the external data, directing the first autonomous vehicle to collect, from a second data source, additional external data to mitigate the insufficiency, wherein the second data source is a data collection device comprising a second autonomous vehicle, and wherein directing collection of the additional external data comprises deploying the second autonomous vehicle from a base station located on or within the first autonomous vehicle and receiving the additional external data from the second autonomous vehicle.

12 . The computer program product of claim 11 , wherein predicting the insufficiency of the external data for performing the action comprises:

analyzing the action to be performed in response to the external data and similar historical actions;

correlating, based on the analyzing, the action with the contextual data about the environment that the first autonomous vehicle is operating;

generating, based on the correlating, a digital twin simulation of the first autonomous vehicle performing the action; and

determining, based on the digital twin simulation, that the action cannot be safely completed based on the external data and contextual data.

13 . The computer program product of claim 11 , wherein directing collection of additional external data from the second data source comprises:

determining, based on the predicting, a type of data required for mitigating the insufficiency of external data; and

identifying, based on the determined type of data, the second data source from a plurality of data sources.

14 . The computer program product of claim 11 , wherein the method performed by the processor further comprises:

correlating the collected additional external data with the external data used to determine the action;

modifying, based on the correlating, the action to a second action, wherein the second action is different than the action; and

directing the first autonomous vehicle to perform the second action.