Optical flow sensing application in agricultural vehicles
System and techniques for optical flow sensing applications in agricultural vehicles are described herein. A plurality of digital images of an agricultural environment can be obtained from a sensor affixed to agricultural equipment. The plurality of digital images can include a first image and a second image, the second image being captured subsequent to the first image. A transformation of a landmark between the first image and the second image can be identified. A degree of motion for the agricultural equipment relative to an environmental target can be calculated based on the transformation of the landmark.
1. A system for applying optical flow sensing for agricultural vehicles, the system comprising:
an image capture controller to obtain a plurality of digital images of an agricultural environment from a sensor affixed to agricultural equipment (AEQ), the plurality of digital images including a first image and a second image, the second image being captured subsequent to the first image, the agricultural environment including an object that is an implement affixed to the AE at a fixation point;
an optical flow circuit set to identify a transformation of a landmark between the first image and the second image, the landmark being common to the first image and the second image; and
a physical environment circuit set to calculate a degree of motion for the AEQ relative to an environmental target based on the transformation of the landmark, the degree of motion including a degree of motion between the AEQ and the object determined by:
measuring a deformation of a shaft affixed the implement to the AEQ between the first image and the second image from the degree of motion; and
converting the deformation into a bending moment measurement.
2. The system of claim 1 , wherein to calculate the degree of motion includes the physical environment circuit set to determine at least one of rotation, scale, or two dimensional translation of the object relative to the agricultural environment.
3. The system of claim 2 , wherein the sensor is affixed with a static distance and static angle relative to the AEQ, the static distance and angle being invariant with respect to the AEQ as the AEQ moves, and wherein the degree of motion includes a course over ground (COG) and distance applied to a planar representation of a surface upon which the AEQ is moving, the COG and distance determined from the rotation, scale, and two dimensional translation of the object relative to the agricultural environment.
4. The system of claim 3 , comprising a steering input circuit set to:
obtain a track angle error and cross-track distance for an AEQ steering solution, wherein the agricultural environment includes a crop row that is a target of the steering solution; and
provide updated track angle error and cross-track distance values by modifying the track angle error and cross-track distance using the COG and distance.
5. The system of claim 1 , wherein the physical environment circuit set is to use the degree of motion between the AEQ and a second object in the agricultural environment as input into a steering solution to maintain a constant distance between the AEQ and the object.
6. The system of claim 1 , wherein the physical environment circuit set is to:
obtain a target row, wherein the agricultural environment includes a plurality of rows including the target row, wherein the degree of motion includes motion relative to rows in the plurality of rows, and wherein the target row is a different than a row in the plurality of rows down which the AEQ is currently moving;
identify a beginning of a turn towards the target row; and
navigate to the target row through the turn via the calculated degree of motion for the AEQ relative to the agricultural environment.
7. The system of claim 1 wherein the shaft is a rotating shaft.
8. A hardware circuit implemented method comprising:
obtaining a plurality of digital images of an agricultural environment from a sensor affixed to agricultural equipment (AEQ), the plurality of digital images including a first image and a second image, the second image being captured subsequent to the first image, the agricultural environment including an object that is an implement affixed to the AE at a fixation point;
identifying a transformation of a landmark between the first image and the second image, the landmark being common to the first image and the second image; and
calculating a degree of motion for the AEQ relative to an environmental target based on the transformation of the landmark, the degree of motion including a degree of motion between the AEQ and the object determined by:
measuring a deformation of a shaft affixed the implement to the AEQ between the first image and the second image from the degree of motion; and
converting the deformation into a bending moment measurement.
9. The method of claim 8 , wherein calculating the degree of motion includes determining at least one of rotation, scale, or two dimensional translation of the object relative to the agricultural environment.
10. The method of claim 9 , wherein the sensor is affixed with a static distance and static angle relative to the AEQ, the static distance and angle being invariant with respect to the AEQ as the AEQ moves, and wherein the degree of motion includes a course over ground (COG) and distance applied to a planar representation of a surface upon which the AEQ is moving, the COG and distance determined from the rotation, scale, and two dimensional translation of the object relative to the agricultural environment.
11. The method of claim 10 , comprising:
obtaining a track angle error and cross-track distance for an AEQ steering solution, wherein the agricultural environment includes a crop row that is a target of the steering solution; and
providing updated track angle error and cross-track distance values by modifying the track angle error and cross-track distance using the COG and distance.
12. The method of claim 8 , comprising using the degree of motion between the AEQ and a second object in the agricultural environment as input into a steering solution to maintain a constant distance between the AEQ and the object.
13. The method of claim 8 , comprising:
obtaining a target row, wherein the agricultural environment includes a plurality of rows including the target row, wherein the degree of motion includes motion relative to rows in the plurality of rows, and wherein the target row is a different than a row in the plurality of rows down which the AEQ is currently moving;
identifying a beginning of a turn towards the target row; and
navigating to the target row through the turn via the calculated degree of motion for the AEQ relative to the agricultural environment.
14. The method of claim 8 , wherein the shaft is a rotating shaft.
15. A set of non-transitory machine readable media including instructions that, when executed by a machine, cause the machine to perform operation comprising:
obtaining a plurality of digital images of an agricultural environment from a sensor affixed to agricultural equipment (AEQ), the plurality of digital images including a first image and a second image, the second image being captured subsequent to the first image, the agricultural environment including an object that is an implement affixed to the AE at a fixation point;
identifying a transformation of a landmark between the first image and the second image, the landmark being common to the first image and the second image; and
calculating a degree of motion for the AEQ relative to an environmental target based on the transformation of the landmark, the degree of motion including a degree of motion between the AEQ and the object determined by:
measuring a deformation of a shaft affixed the implement to the AEQ between the first image and the second image from the degree of motion; and
converting the deformation into a bending moment measurement.
16. The set of machine readable media of claim 15 , wherein calculating the degree of motion includes determining at least one of rotation, scale, or two dimensional translation of the object relative to the agricultural environment.
17. The set of machine readable media of claim 16 , wherein the sensor is affixed with a static distance and static angle relative to the AEQ, the static distance and angle being invariant with respect to the AEQ as the AEQ moves, and wherein the degree of motion includes a course over ground (COG) and distance applied to a planar representation of a surface upon which the AEQ is moving, the COG and distance determined from the rotation, scale, and two dimensional translation of the object relative to the agricultural environment.
18. The set of machine readable media of claim 17 , wherein the operations further comprise:
obtaining a track angle error and cross-track distance for an AEQ steering solution, wherein the agricultural environment includes a crop row that is a target of the steering solution; and
providing updated track angle error and cross-track distance values by modifying the track angle error and cross-track distance using the COG and distance.
19. The set of machine readable media of claim 15 , wherein the operations further comprise using the degree of motion between the AEQ and a second object in the agricultural environment as input into a steering solution to maintain a constant distance between the AEQ and the object.
20. The set of machine readable media of claim 15 , wherein the operations further comprise:
obtaining a target row, wherein the agricultural environment includes a plurality of rows including the target row, wherein the degree of motion includes motion relative to rows in the plurality of rows, and wherein the target row is a different than a row in the plurality of rows down which the AEQ is currently moving;
identifying a beginning of a turn towards the target row; and
navigating to the target row through the turn via the calculated degree of motion for the AEQ relative to the agricultural environment.
21. The method of claim 15 , wherein the shaft is a rotating shaft.