Systems and methods of optimizing utilization of vehicle onboard storage

1. A method of optimizing utilization of vehicle onboard storage, the method comprising, by a computer system for a vehicle:

monitoring real-time environmental conditions via a plurality of sensors positioned in proximity to the vehicle;

responsive to the monitoring, receiving, from one or more of the plurality of sensors, a first real-time measured value of a first environmental variable and at least one additional real-time measured value of at least one additional environmental variable;

automatically determining, via a machine-learning model, a first expected value of the first environmental variable given the at least one additional real-time measured value of the at least one additional environmental variable;

automatically generating a storage decision in relation to the first real-time measured value based, at least in part, on an evaluation of the first real-time measured value relative to the first expected value; and

causing execution of the storage decision at least with respect to onboard storage of the vehicle.

2. The method of claim 1 , wherein the storage decision indicates at least one of the following:

storage of the first real-time measured value in the onboard storage of the vehicle;

non-storage of the first real-time measured value in the onboard storage of the vehicle; and

transmission of the first real-time measured value over a network for off-board storage.

3. The method of claim 1 , wherein the automatically generating comprises:

computing a difference between the first real-time measured value and the first expected value;

expressing the computed difference as a proportion of the first real-time measured value; and

comparing the proportion to each of at least one threshold for the first environmental variable.

4. The method of claim 3 , wherein the automatically generating comprises, responsive to the proportion satisfying the at least one threshold, automatically determining, as at least a portion of the storage decision, to store the first real-time measured value in the onboard storage.

5. The method of claim 4 , wherein the causing execution comprises storing the first real-time measured value and the at least one additional real-time measured value in the onboard storage.

6. The method of claim 3 , wherein:

the at least one threshold comprises a plurality of thresholds, each threshold of the plurality of thresholds being associated with a distinct priority value of a plurality of priority values;

the automatically generating comprises assigning a particular priority value of the plurality of priority values to the first real-time measured value, wherein the assigning is based, at least in part, on the comparing; and

the causing execution comprises storing the first real-time measured value in the onboard storage in relation to the particular priority value.

7. The method of claim 6 , wherein:

the onboard storage comprises a fixed-size buffer; and

the causing execution comprises overwriting data in the fixed-size buffer in an order that is at least partially based on the plurality of priority values.

8. The method of claim 3 , comprising sending an alert in response to the proportion satisfying the at least one threshold.

9. The method of claim 1 , comprising:

responsive to the monitoring, periodically receiving a plurality of real-time measured values of a plurality of environmental variables from the plurality of sensors;

for each environmental variable of the plurality of environmental variables, for each real-time measured value of the environmental variable that is periodically received:

automatically determining, via the machine-learning model, an expected value of the environmental variable given at least one other real-time measured value of at least one other environmental variable of the plurality of environmental variables;

automatically generating a particular storage decision in relation to the real-time measured value based, at least in part, on an evaluation of the real-time measured value relative to the expected value; and

causing execution of the particular storage decision at least with respect to the onboard storage.

10. The method of claim 9 , comprising retraining the machine-learning model using stored values of the plurality of environmental variables in the onboard storage.

11. The method of claim 1 , wherein the vehicle is an aircraft.

12. A vehicle computer system comprising a processor and memory, wherein the processor and the memory in combination are operable to perform a method comprising:

monitoring real-time environmental conditions via a plurality of sensors positioned in proximity to a vehicle;

responsive to the monitoring, receiving, from one or more of the plurality of sensors, a first real-time measured value of a first environmental variable and at least one additional real-time measured value of at least one additional environmental variable;

automatically determining, via a machine-learning model, a first expected value of the first environmental variable given the at least one additional real-time measured value of the at least one additional environmental variable;

automatically generating a storage decision in relation to the first real-time measured value based, at least in part, on an evaluation of the first real-time measured value relative to the first expected value; and

causing execution of the storage decision at least with respect to onboard storage of the vehicle.

13. The vehicle computer system of claim 12 , wherein the storage decision indicates at least one of the following:

storage of the first real-time measured value in the onboard storage of the vehicle;

non-storage of the first real-time measured value in the onboard storage of the vehicle; and

transmission of the first real-time measured value over a network for off-board storage.

14. The vehicle computer system of claim 12 , wherein the automatically generating comprises:

computing a difference between the first real-time measured value and the first expected value;

expressing the computed difference as a proportion of the first real-time measured value; and

comparing the proportion to each of at least one threshold for the first environmental variable.

15. The vehicle computer system of claim 14 , wherein the automatically generating comprises, responsive to the proportion satisfying the at least one threshold, automatically determining, as at least a portion of the storage decision, to store the first real-time measured value in the onboard storage.

16. The vehicle computer system of claim 14 , wherein:

the at least one threshold comprises a plurality of thresholds, each threshold of the plurality of thresholds being associated with a distinct priority value of a plurality of priority values;

the automatically generating comprises assigning a particular priority value of the plurality of priority values to the first real-time measured value, wherein the assigning is based, at least in part, on the comparing; and

the causing execution comprises storing the first real-time measured value in the onboard storage in relation to the particular priority value.

17. The vehicle computer system of claim 16 , wherein:

the onboard storage comprises a fixed-size buffer; and

the causing execution comprises overwriting data in the fixed-size buffer in an order that is at least partially based on the plurality of priority values.

18. The vehicle computer system of claim 12 , the method comprising:

responsive to the monitoring, periodically receiving a plurality of real-time measured values of a plurality of environmental variables from the plurality of sensors;

for each environmental variable of the plurality of environmental variables, for each real-time measured value of the environmental variable that is periodically received:

automatically determining, via the machine-learning model, an expected value of the environmental variable given at least one other real-time measured value of at least one other environmental variable of the plurality of environmental variables;

automatically generating a particular storage decision in relation to the real-time measured value based, at least in part, on an evaluation of the real-time measured value relative to the expected value; and

causing execution of the particular storage decision at least with respect to the onboard storage.

19. The vehicle computer system of claim 18 , the method comprising retraining the machine-learning model using stored values of the plurality of environmental variables in the onboard storage.

20. A computer-program product comprising a non-transitory computer-usable medium having computer-readable program code embodied therein, the computer-readable program code adapted to be executed to implement a method comprising:

monitoring real-time environmental conditions via a plurality of sensors positioned in proximity to a vehicle;

responsive to the monitoring, receiving, from one or more of the plurality of sensors, a first real-time measured value of a first environmental variable and at least one additional real-time measured value of at least one additional environmental variable;

automatically determining, via a machine-learning model, a first expected value of the first environmental variable given the at least one additional real-time measured value of the at least one additional environmental variable;

automatically generating a storage decision in relation to the first real-time measured value based, at least in part, on an evaluation of the first real-time measured value relative to the first expected value; and

causing execution of the storage decision at least with respect to onboard storage of the vehicle.