AI-optimized harvester configured to maximize yield and minimize impurities
Systems and methods are disclosed herein for optimizing harvester yield. In an embodiment, a controller receives a pre-harvest image from a front-facing camera of a harvester. The controller inputs the pre-harvest image into a model, and receives as output from the model a predicted harvest yield. The controller receives, from an interior camera of the harvester, a post-harvest image including the plants as harvested. The controller inputs the post-harvest image into a second model and receives, as output, an actual harvest yield of the plants as-harvested. The controller determines that the predicted harvest yield does not match the actual harvest yield, and outputs a control signal.
1. A computer-implemented method for optimizing harvester yield, the method comprising:
receiving, from a front-facing camera of a harvester, a pre-harvest image including plants;
inputting the pre-harvest image into a first machine learning model;
receiving as output from the first machine learning model a predicted harvest yield of the plants;
receiving, from an interior camera of the harvester, a post-harvest image including the plants as-harvested;
inputting the post-harvest image into a second machine learning model;
receiving, as output from the second machine learning model, an actual harvest yield of the plants as-harvested;
determining whether the predicted harvest yield matches the actual harvest yield; and
responsive to determining that the predicted harvest yield does not match the actual harvest yield, outputting a control signal.
2. The computer-implemented method of claim 1 , wherein the contents of the receptacle comprise grain.
3. The method of claim 1 , wherein the interior camera is facing at least one of an elevator or a receptacle of the harvester.
4. The method of claim 1 , wherein the interior camera comprises both a camera facing the elevator and a camera facing the receptacle, and wherein the actual harvest yield is determined based on inputting, into the second machine learning model, images from both the camera facing the elevator and the camera facing the receptacle.
5. The method of claim 1 , wherein the output of the first machine learning model is a predicted amount of plants that will be harvested.
6. The method of claim 1 , wherein the output of the second machine learning model is an actual amount of plants that were harvested.
7. The method of claim 1 , wherein determining whether the predicted harvest yield matches the actual harvest yield comprises:
determining whether a difference between the actual harvest yield and the predicted harvest yield is less than a threshold.
8. The method of claim 1 , further comprising:
determining, based on an image from a rear of the harvester, a measure of loss of the plants; and
inputting, in addition to the post-harvest image, the measure of loss into the second machine learning model.
9. The method of claim 1 , wherein the control signal is transmitted to an operator interface that, responsive to receiving the control signal, provides an operator of the harvester with a recommended state change for the harvester.
10. The method of claim 9 , wherein the recommended state change specifies a component of the harvester and a recommended action to take with respect to the specified component.
11. The method of claim 1 , wherein the control signal causes speed of a thresher of the harvester to adjust.
12. The method of claim 11 , wherein the control signal causes the speed of the thresher to decrease responsive to detecting, from an external camera, a loss of quantity of the plants from the thresher.
13. The method of claim 1 , further comprising uploading indicia of the control signals from the harvester to a cloud server.
14. The method of claim 13 , further comprising adjusting the machine learning model based on data downloaded from the cloud server.
15. The method of claim 1 , wherein the control signal causes speed of a cutter of the harvester to adjust.
16. The method of claim 15 , wherein the control signal causes the speed of the cutter to decrease responsive to determining that the actual harvest yield is higher than the predicted harvest yield.
17. The method of claim 15 , wherein the control signal causes the speed of the cutter to increase responsive to determining that the actual harvest yield is lower than the predicted harvest yield.
18. The method of claim 1 , wherein the control signal causes a height of a cutter of the harvester to adjust.
19. The method of claim 18 , wherein the control signal causes the height of the cutter to increase responsive to receiving input, from a camera facing the cutter, indicating lodging of the plants in the cutter.
20. The method of claim 18 , wherein the control signal causes the height of the cutter to decrease responsive to receiving input, from the camera facing the cutter, that there is no further lodging of the plants in the cutter.