Methods and systems for scheduling of sensing platform nodes

1. A method for observing a predetermined monitoring area, wherein one or more sensing platform nodes are employed to observe a predetermined number of sub-areas of the monitoring area, the method comprising:

observing the sub-areas of the monitoring area using the sensing platform nodes so as to collect measuring data for the sub-areas,

providing at least one prediction model for analyzing predictability of measuring data for the sub-areas based on the collected measuring data,

calculating future measuring data for the sub-areas and calculating uncertainty of the future measuring data over time using the prediction model, and

scheduling the sensing platform nodes for observation of the sub-areas according to a scheduling mechanism, wherein the scheduling of the sensing platform nodes is dependent on the calculated uncertainty of the future measuring data predicted for the sub-areas,

wherein the scheduling is performed on the basis of a cost function J(t) according to:

J

⁡

(

t

)

=

∑

i

⁢

P

i

·

uncertainty

i

⁡

(

t

)

⁢

⁢

∀

i

∈

sub

⁢

-

⁢

areas

⁢

⁢

to

⁢

⁢

observe

wherein uncertainty i (t) is the uncertainty of the calculated future measuring data for a respective one of the observed sub-areas i at time t, wherein parameter P i is 0 in a case that a respective one of the sensing platform nodes observes the sub-area i at time t, and wherein parameter P i is 1 in a case that the respective one of the sensing platform nodes does not observe the sub-area i at time t.

2. The method according to claim 1 , wherein the sensing platform nodes include static sensing platform nodes that are located on the sub-areas of the monitoring area, wherein the static sensing platform nodes are accessed/actuated for observing the sub-areas according to the scheduling mechanism.

3. The method according to claim 1 , wherein the sensing platform nodes include mobile sensing platform nodes that are movable to the sub-areas of the monitoring area, wherein the mobile sensing platform nodes are moved to the sub-areas for observing the sub-areas according to the scheduling mechanism.

4. The method according to claim 1 , wherein the sensing platform nodes sense a numerical quantity.

5. The method according to claim 1 , wherein the measuring data of a respective one of the observed sub-areas is provided as a time series of values, wherein by learning from the provided values the prediction model is derived for predicting future values representing the future measuring data of the sub-area.

6. The method according to claim 1 , wherein the uncertainty of the calculated future measuring data is calculated by determining and/or considering confidence intervals pertaining to the calculated future measuring data.

7. The method according to claim 1 , wherein the uncertainty of the calculated future measuring data of a respective one of the observed sub-areas is calculated by considering the uncertainty and/or the measuring data of one or more nearby ones of the observed sub-areas.

8. The method according to claim 1 , wherein the scheduling mechanism considers an aggregated uncertainty of the future measuring data with regard to the observed monitoring area, and wherein the scheduling is performed in such a way that the aggregated uncertainty is reduced and/or minimized.

9. The method according to claim 1 , wherein the scheduling of the sensing platform nodes is determined using an optimization algorithm such that the uncertainty is minimized.

10. The method according to claim 1 , wherein the scheduling mechanism is individually implemented in the sensing platform nodes such that the sensing platform nodes are able to make autonomous scheduling decisions.

11. A system for observing a predetermined monitoring area, the system comprising:

at least one sensing platform node for observing a predetermined number of sub-areas of the monitoring area, the system being configured to observe the sub-areas of the monitoring area using the sensing platform node so as to collect measuring data for the sub-areas,

at least one prediction model configured to analyze predictability of measuring data for the sub-areas based on the collected measuring data, to calculate future measuring data for the sub-areas and to calculate uncertainty of the future measuring data over time, and

at least one planner configured to schedule the sensing platform node for observation of the sub-areas, wherein the scheduling of the sensing platform node is dependent on the calculated uncertainty of the future measuring data predicted for the sub-areas;

wherein the at least one planner is configured to perform the scheduling on the basis of a cost function J(t) according to:

J

⁡

(

t

)

=

∑

i

⁢

⁢

P

i

·

uncertainty

i

⁡

(

t

)

⁢

⁢

∀

i

∈

sub

-

areas

⁢

⁢

to

⁢

⁢

observe

wherein uncertainty i (t) is the uncertainty of the calculated future measuring data for a respective one of the observed sub-areas i at time t, wherein parameter P i is 0 in a case that a respective one of the sensing platform nodes observes the sub-area i at time t, and wherein parameter P i is 1 in a case that the respective one of the sensing platform nodes does not observe the sub-area i at time t.

12. The system according to claim 11 , wherein one or both of the prediction model and the planner are implemented at the sensing platform node.

13. The system according to claim 11 , wherein one or both of the prediction model and the planner are implemented centralized.

14. The system according to claim 13 , wherein one or both of the prediction model and the planner are implemented as a receiver which receive the measuring data collected by the sensing platform node.

15. The method according to claim 1 , comprising aggregating, across the sub-areas, the calculated uncertainty and scheduling the sensing platform nodes based on the aggregated uncertainty.

16. The method according to claim 1 , wherein the uncertainty of the future measuring data is calculated as a function of time.

17. The method according to claim 16 , wherein the scheduling of the sensing platform nodes is dependent on the calculated uncertainty of the future measuring data predicted for the sub-areas at multiple different points in time.