Sensor fusion

1 . A computer-implemented method for correcting remote sensor data of one or more agricultural fields, the method comprising:

sensing remote sensor data for a first agricultural field using a remote sensor, the remote sensor data comprising at least one remote measurement value corresponding to a location of the first agricultural field at a point in time, wherein the remote measurement value is associated with a hyperspectral index or a biomass index;

receiving local sensor data for the first agricultural field from at least one local sensor, the local sensor data comprising at least one local measurement value corresponding to a location of the local sensor and a point in time correlating to the location and point in time of the remote measurement value;

determining a correction model based on the previously received local sensor data and the previously received remote sensor data;

applying the correction model to current remote sensor data to generate corrected current remote sensor data.

2 . The method of claim 1 , wherein the local sensor is non-stationary.

3 . The method of claim 1 , wherein the local measurement value is associated with a hyperspectral index or a biomass index.

4 . The method of claim 1 , further comprising the steps after having received the remote sensor data and the local sensor data:

determining the remote measurement value corresponding to the location of the at least one local sensor; and

determining a difference between the remote measurement value corresponding to the location of the at least one local sensor and the local measurement value of the at least one local sensor for a plurality of points in time,

wherein determining the correction model is further based on the determined difference.

5 . The method of claim 4 , wherein determining the difference between the remote measurement value corresponding to the location of the at least one local sensor and the local measurement value of the at least one local sensor for a plurality of points in time comprises the steps:

receiving local time series data of the local sensor data (DL) from the at least one local sensor, wherein the local time series data comprises a plurality of location data of a plurality of points in time corresponding to the location, where the at least one local sensor is located at a specific point in time;

determining a cluster of local sensor data based on a maximal distance between the respective locations of the local sensors over time;

determining a clustered location of the at least one local sensor based on the determined cluster; and

determining the difference between the remote measurement value corresponding to the clustered location of the at least one local sensor and the local measurement value of the at least one local sensor for a plurality of points in time,

wherein determining the correction model is further based on the determined difference.

6 . The method of claim 5 , wherein the clustered location is determined by determining a center of the cluster.

7 . The method of claim 1 , wherein the remote sensor data comprises at least one remote image, which is based on the at least one remote measurement value;

wherein determining the remote measurement value corresponding to the location of the local sensor comprises the step:

extracting a pixel from the remote image that is closest to the location of the local sensor or extracting a mean of pixels within a predefined distance of the pixel closest to the location of the local sensor; and

determining the remote measurement value based on the extracted pixel,

wherein determining the correction model is also based on the extracted pixel on which the remote measurement value bases.

8 . The method of claim 1 , wherein the method comprises the steps:

if received remote time series data associated with the remote sensor data (DR) comprises at least one gap, where the remote sensor data (DR) at an expected point in time in the time series of remote time series data (DR) are missing; then

receiving local sensor data (DL) for the point in time of the gap; and

determining predicted remote sensor data (DP) for a point in time of the gap based on the received local sensor data (DL).

9 . The method of claim 8 , wherein determining the predicted remote sensor data (DP) comprises:

receiving remote sensor data (DR) of a point in time just before the gap; and

determining the predicted remote sensor data (DP) based on the received remote sensor data (DR) of the point in time just before the gap.

10 . The method of claim 1 , wherein the correction model comprises a projection function depending on historical data sets of remote sensor data and local sensor data, and wherein the remote sensor data is determined based on the projection function.

11 . A non-transitory computer readable storage medium having instructions encoded thereon that, when executed by a processing device, cause the processing device to perform the method of claim 1 .

12 . The method of claim 1 , wherein the remote sensor data (DR) and the local sensor data (DL) are used for monitoring the agricultural field.

13 . A processing device, configured to: sense remote sensor data for a first agricultural field by using a remote sensor, wherein the remote sensor data is received and comprises at least one remote measurement value corresponding to at least one location that is measured by the remote sensor at at least one point in time of obtaining the remote measurement value, corresponding to a location of the first agricultural field at a point in time, wherein the remote measurement value is associated with a hyperspectral index or a biomass index;

receive local sensor data for the first agricultural field from at least one local sensor, wherein the local sensor data comprises at least one local measurement value corresponding to at least one location of the at least one local sensor and corresponding to at least one point in time of obtaining the local measurement value correlating to the location and point of time of obtaining the remote measurement value, wherein the local measurement value corresponds to a location of the local sensor and a point in time correlating to the location and point in time of the remote measurement value;

determine a correction model based on the previously received local sensor data and the previously received remote sensor data;

apply the correction model to current remote sensor data to generate corrected current remote sensor data.

14 . A system for correcting remote sensor data of one or more agricultural fields, comprising:

a remote sensor, configured for providing remote sensor data for the first agricultural field from a remote sensor, wherein the remote sensor data comprises at least one remote measurement value corresponding to at least one location that is measured by the remote sensor at at least one point in time of obtaining the remote measurement value, corresponding to a location of the first agricultural field at a point in time, wherein the remote measurement value is associated with a hyperspectral index or a biomass index;

a local sensor, configured for providing local sensor data for the first agricultural field from at least one local sensor, wherein the local sensor data comprises at least one local measurement value corresponding to at least one location of the at least one local sensor and corresponding to at least one point in time of obtaining the local measurement value correlating to the location and point of time of obtaining the remote measurement value, wherein the local measurement value corresponds to a location of the local sensor and a point in time correlating to the location and point in time of the remote measurement value; and

the processing device of claim 13 .

15 . A method for correcting remote sensor data of one or more agricultural fields, the method comprising:

sensing, by a remote sensor, remote sensor data including at least one remote measurement value corresponding to at least one location that is measured by the remote sensor at at least one point in time of obtaining the remote measurement value, for a first agricultural field using the remote sensor, the remote sensor data comprising at least one remote measurement value corresponding to a location of the first agricultural field at a point in time, wherein the remote measurement value is associated with a hyperspectral index or a biomass index;

receiving local sensor data for the first agricultural field from at least one local sensor, wherein the local sensor data comprises at least one local measurement value corresponding to at least one location of the at least one local sensor and corresponding to at least one point in time of obtaining the local measurement value correlating to the location and point of time of obtaining the remote measurement value, wherein the local measurement value corresponds to a location of the local sensor and a point in time correlating to the location and point in time of the remote measurement value;

determining a correction model based on the previously received local sensor data and the previously received remote sensor data;

applying the correction model to current remote sensor data to generate corrected current remote sensor data.