A METHOD OF SELECTIVELY TREATING VEGETATION IN A FIELD

Abstract

The present disclosure relates to a method of selectively treating vegetation in a field with a treatment apparatus that is configured to move in the field and to treat vegetation in the field. In particular, the method includes determining a location of target vegetation in the field and having the treatment apparatus, such as a sprayer, move to the target vegetation in the field and treat the tart vegetation in the field with a treatment applicator disposed on the treatment apparatus.

Description

Claims

1. A computer-implemented method of selectively treating vegetation in a field with a treatment apparatus that is configured to move in the field and to treat vegetation in the field, the method including: receiving geo-referenced image data of a field, from one or more image capture devices not coupled to the treatment apparatus, wherein the geo-referenced image data includes a plurality of pixels and geo-reference data associated with each one of the pixels;analysing spectral information of the pixels to classify pixels as target vegetation in the field;determining a location of the target vegetation in the field based on the geo-reference data associated with pixels classified as said target vegetation;determining geo-referenced application map data based on the location of the target vegetation in the field, wherein pixels not classified as said target vegetation are not included in the geo-referenced application map data;providing the geo-referenced application map data to the treatment apparatus and moving the treatment apparatus based on the geo-referenced application map data;receiving ambient data from the treatment apparatus moving in the field;determining treatment instruction data for actuation of at least one treatment applicator, disposed on the treatment apparatus, configured to treat vegetation in the field, based on the geo-referenced application map data correlated with the ambient data;providing the treatment instruction data to the at least one treatment applicator; andthe at least one treatment applicator treating the target vegetation in the field based on the treatment instruction data when the treatment apparatus is moved to the target vegetation in the field.

2. A computer-implemented method according to claim 1, further including determining an optimal route for the treatment apparatus to move in the field based on the location of the target vegetation in the field and recording the optimal route in the geo-referenced application map data.

3. A computer-implemented method according to claim 1, wherein analysing the spectral information of the pixels includes unsupervised and or supervised classification using a classifier trained on a spectral signature of target vegetation in a field.

4. A computer-implemented method according to claim 3, further including embedding a chlorophyll spectral signature in selected pixels of the image data corresponding to vegetation in the field, and analysing the chlorophyll spectral signature of these pixels using the classifier.

5. A computer-implemented method according to claim 1, wherein the geo-referenced image data is recorded as data native to the one or more image capture devices.

6. A computer-implemented method according to claim 5, wherein the data has a resolution of less than or equal to 5 cm.

7. A computer-implemented method according to claim 3, further including classifying the target vegetation in the field as a plurality of grades of target vegetation based on a vegetation index.

8. A computer-implemented method according to claim 3, wherein the geo-referenced image data of the field includes a plurality of channels of data, and the method further includes consolidating the channels to form one channel for the geo-referenced application map data by representing the target vegetation as a vector of intensity values.

9. A computer-implemented method according to claim 8, further including segmenting the geo-referenced application map data by applying one or more threshold values to the intensity values.

10. A computer-implemented method according to claim 1, wherein analysing the spectral information of the pixels includes classifying pixels as areas of the field that are not target vegetation in the field, and determining areas of the field that remain as the target vegetation in the field.

11. A computer-implemented method according to claim 1, wherein the treatment apparatus further includes a plurality of sensors for sensing the ambient data.

12. (canceled)

13. (canceled)

14. A computer-implemented method according to claim 11, wherein the treatment apparatus is a sprayer carrying one or more agrochemicals for the at least one treatment applicator to treat the target vegetation in the field, and the treatment instruction data includes instructions for applying the one or more agrochemicals to the target vegetation in the field with the at least one treatment applicator.

15. A computer-implemented method according to claim 14, wherein the treatment apparatus includes more than one treatment applicator, and the method includes selecting one of the treatment applicators closest to the target vegetation in the field to treat the target vegetation based on the ambient data.

16. A computer-implemented method according to claim 15, wherein the treatment apparatus includes a rigid portion and a flexible suspended boom including the treatment applicators disposed thereon, each having one or more nozzles for applying the one or more agrochemicals to the target vegetation in the field.

17. A computer-implemented method according to claim 16, wherein the plurality of sensors are disposed on the rigid portion and the suspended boom at spaced apart locations, and the sensors sense position data of each of the one or more nozzles relative to the rigid portion of the treatment apparatus.

18. A computer-implemented method according to claim 17, wherein the sensors sense heading data of each of the one or more nozzles, and the method includes determining in real-time a position of each of the one or more nozzles in the field.

19. (canceled)

20. A computer-implemented method according to claim 1, further including kinematic modelling of each of the components of the treatment apparatus in a dynamic apparatus model based on physical constraints of the treatment apparatus, and comparing the dynamic apparatus model to the ambient data sensed by the sensors of the treatment apparatus to resolve ambient influences on the ambient data.

21. (canceled)

22. A computer-implemented method according to claim 1, further including receiving the geo-referenced image data from an unmanned aerial vehicle (UAV), or a manned aerial vehicle, including the one or more image capture devices, configured to fly over the field and to generate the geo-referenced image data.

23. A computer-implemented method according to claim 1, further including receiving the geo-referenced image data from one or more satellites including the one or more image capture devices.

24. A system for selectively treating vegetation in a field, the system including: a treatment apparatus configured to move in the field and to treat vegetation in the field with at least one treatment applicator disposed on the treatment apparatus; andone or more processors configured to: receive geo-referenced image data of a field, from one or more image capture devices not coupled to the treatment apparatus, wherein the geo-referenced image data includes a plurality of pixels and geo-reference data associated with each one of the pixels;analyse spectral information of the pixels to classify pixels as target vegetation in the field;determine a location of the target vegetation in the field based on the geo-reference data associated with pixels classified as said target vegetation;determine geo-referenced application map data based on the location of the target vegetation in the field, wherein pixels not classified as said target vegetation are not included in the geo-referenced application map data;provide the geo-referenced application map data to the treatment apparatus and move the treatment apparatus based on the geo-referenced application map data;receive ambient data from the treatment apparatus moving in the field;determine treatment instruction data for actuation of the at least one treatment applicator based on the geo-referenced application map data correlated with the ambient data; andprovide the treatment instruction data to the at least one treatment applicator, whereinthe at least one treatment applicator is configured to treat the target vegetation in the field based on the treatment instruction data when the treatment apparatus is moved to the target vegetation in the field.

25. (canceled)

26. (canceled)

27. (canceled)

28. (canceled)

29. (canceled)