专利摘要:
SYSTEM OF CONTROL AND MEASUREMENT OF THE CUTTING HEIGHT FOR A BASIC HARVESTER, METHOD, AND, HARVESTER. A system is provided so that it can include a 3D sensor. The 3D sensor can be configured to detect an elevator area in a combine. The 3D sensor can also be configured to transmit a first signal associated with the area. The system also includes a processing device when communicating with the 3D sensor. The system may further include a memory device in which executable instructions are stored by the processing device to make the processing device receive the first signal and determine a volume of material in the elevator based on the first signal.
公开号:BR102014027364B1
申请号:R102014027364-6
申请日:2014-10-31
公开日:2021-01-12
发明作者:Matthew J. Darr;Daniel Joseph Corbett;Herman Herman;Bryan E. Dugas;Herman Badino;Carlos Vallespi-Gonzalez
申请人:Iowa State University Research Foundation, Inc.;Deere & Company;Carnegie Mellon University;
IPC主号:
专利说明:

[0001] [0001] This disclosure relates generally to devices for use in agricultural operations. More specifically, but not by way of limitation, this disclosure refers to producing cutting height measurement and control systems for a basic harvester. Fundamentals of the invention
[0002] [0002] A harvester or reaper can be used to collect material (eg, sugar cane, vegetables, crops, etc.) from a farm and loading the material onto a cart pulled by a tractor. As with many other types of agricultural operations, it may be desirable for a farm operator to know the yield for each section of the farm. This can allow the farm operator to optimize the cultivation process. Income data can be difficult to obtain. summary
[0003] [0003] In one example, a system is provided that includes a 3D sensor. The 3D sensor can be configured to detect an elevator area on a combine. The 3D sensor can also be configured to transmit a first signal associated with the area. The system also includes a processing device when communicating with the camera system. The system may further include a memory device in which executable instructions are stored by the processing device to cause the processing device to receive the first signal and to determine a volume of a material in the elevator based on the first signal.
[0004] [0004] In another example, a method is provided that may include receiving, by a processing device, a data signal from a 3D sensor. The 3D sensor can be positioned on a combine harvester lift. The method can also include generating, through the processing device, a three-dimensional (3D) map based on the data signal. The method can include more determination, the processing device, a volume of material on the elevator according to the 3D map.
[0005] [0005] In another example, a combine is provided. The combine may include an elevator. A 3D sensor can be coupled with the combine. The 3D sensor can be configured to detect an area associated with the elevator. The additional 3D sensor can be configured to transmit a data signal associated with the area to generate a 3D map associated with a material in the elevator, determine a composition of the material and determine a volume of cultures included in the material based on the composition and the 3D map.
[0006] [0006] In one example, a system is provided that includes a 3D sensor. The 3D sensor can be configured to detect an elevator area on a combine. The 3D sensor can also be configured to transmit a first signal associated with the area. The system also includes a processing device when communicating with the camera system. The system may further include a memory device in which executable instructions are stored by the processing device to cause the processing device to receive the first signal and to determine a volume of a material in the elevator based on the first signal.
[0007] [0007] In another example, a method is provided that may include receiving, by a processing device, a data signal from a 3D sensor positioned on a combine. The method may also include generating, through the processing device, a three-dimensional (3D) map based on the data signal. The method may further include determining, by the processing device, a lateral position of a material row or a height of the material line based on the 3D map. Brief Description of the Figures
[0008] [0008] Fig. 1 is a front perspective view of an example of a combine harvester for use with a yield measurement system and a base cutter height control system according to an example.
[0009] [0009] Fig. 2 is a perspective view of a yield measurement system mounted on a combine harvester according to an example.
[0010] [0010] Fig. 3 is a perspective view of a camera system for use with a performance measurement system according to an example.
[0011] [0011] Fig. 4 is a perspective view of an elevator for use with a performance measurement system according to an example.
[0012] [0012] Fig. 5 is a top view of a paddle lift for use with a performance measurement system according to an example.
[0013] [0013] Fig. 6 is a block diagram, representing an example of a computing device for use with a performance measurement system according to an example.
[0014] [0014] Fig. 7 is a block diagram of an example of a process for measuring the yield of a material transported by the elevator according to an example.
[0015] [0015] Fig. 8 is a segmented image for use in determining the material composition of the image according to an example.
[0016] [0016] Fig. 9 is an example of a process for determining the composition of a material in an image according to an example.
[0017] [0017] Fig. 10 shows examples of output yield maps by a yield measurement system according to an example.
[0018] [0018] Fig. 11 is a side perspective view of an example of a combine for use with a height control base cut system according to an example.
[0019] [0019] Fig. 12 is a front perspective view of an example of a combine harvester for use with a height control base cut system according to an example.
[0020] [0020] Fig. 13 is a perspective view of an example of a camera system mounted on the underside of a combine device for use with a height control base cut system according to an example.
[0021] [0021] Fig. 14 is a perspective view of an example of a camera system mounted at an angle to the underside of a culture device and for use with a height control base cut system according to An example.
[0022] [0022] Fig. 15 is a perspective view of an example of a 1502 lighting system for use with a height control base cut system according to an example.
[0023] [0023] Fig. 16 is a block diagram of an example of a process for determining a new base cutter level according to an example.
[0024] [0024] Fig. 17 is a graph showing an example of a 3D map for use with a height-based base cut system according to an example.
[0025] [0025] Fig. 18 is a graph for use in determining the lateral line position, line height, and material height according to an example.
[0026] [0026] Fig. 19 is an example of a geographic map output by a height control base cut system according to an example. Detailed Description Yield Measurement System
[0027] [0027] Certain aspects and characteristics of the present disclosure are directed to a real time, to produce measurement system for a harvester (for example, a harvester). The yield measurement system may include a 3D sensor (for example, a stereo camera system) coupled to a combine lift. The 3D sensor can detect a material (for example, sugar cane, potatoes, beets, tomatoes or other crops) passing over part of the elevator. The yield measurement system can process the 3D sensor data to measure the net volume of materials transported by the elevator in real time (for example, substantially simultaneously with culture).
[0028] [0028] In some examples, the yield measurement system can distinguish between crops and garbage (for example, foreign materials) carried by elevator. The yield measurement system can distinguish between crops and waste, determining one or more characteristics (for example, sizes, colors, shapes and textures) of the material transported by elevator. The yield measurement system can determine the composition of the material (for example, how most of the material is the crop and the amount of waste material is) based on the characteristics. For example, the 3D sensor can include two or more cameras. The performance measurement system can analyze the images of the visible portion of the elevator through the captured cameras. The captured images can capture the surface area of the material in the elevator. Based on the captured images, the yield measurement system, it can be estimated that 80% of the material's surface area includes sugarcane and 20% of the material's surface area includes garbage. Based on the waste culture ratio, the yield measurement system can apply a scale factor to measure liquid volume to estimate the volume of material associated with the culture. For example, the yield measurement system can multiply the net volume measurement by 80% (ie 0.8) to determine the actual volume of the material that is associated with sugarcane. In this way, the measurement system can determine the yield (for example, estimate) from the captured images the volume of the culture on the elevator portion visible by the chambers.
[0029] [0029] In some examples, the yield measurement system can measure the volume of material in overlapping sections of the elevator. The Measurement System can adjust the volume measurement based on the amount of overlap. This can allow the yield measurement system to provide more accurate volume measurements.
[0030] [0030] In some examples, the yield measurement system may include sensors for detecting parameters associated with the combine. Parameters can include ground speed, Global Positioning System (GPS) location, primary extractor speed, secondary extractor speed, primary fan speed, secondary fan speed, engine rpm, a parameter associated with a cutter tips, direction, crop processing hitch and the operational status of the harvester The yield measurement system can integrate volume measurement with harvester parameters to generate a real-time yield map.
[0031] [0031] These illustrative examples are given to introduce the reader to the general subject discussed here and are not intended to limit the scope of the concepts disclosed. The following sections describe several additional resources and examples with reference to drawings in which numerals indicate how the elements, and directional descriptions are used to describe illustrative aspects, but, as illustrative aspects, should not be used to limit disclosure. gift.
[0032] [0032] Fig. 1 is a front perspective view of an example of a combine 100 for use with a yield measurement system and a height cutter height control system according to an example. Harvester 100 can be used to collect material 102 (for example, sugar cane 506 shown in Fig. 5) from a field. The harvester 100 can be any harvester 100 for collecting any type of crop. Material 102 can pass through the prospector 100 on the elevator 104. Material 102 can be transported by elevator 104 and loaded (e.g., dumped) onto a cart 108 pulled by a tractor 106.
[0033] [0033] Fig. 2 is a perspective view of a yield measurement system 200 mounted on a harvester elevator 104 according to an example. The performance measurement system 200 may include a 3D sensor 202 to capture the three-dimensional (3D) profile of material in the elevator 104. The 3D sensor may include one or more cameras, a 3D laser scanner, an ultrasonic transducer, an range, a structured light camera, a 3D camera and a LIDAR flash.
[0034] [0034] In some examples, the 3D sensor 202 may include two or more cameras (for example, color cameras or grayscale cameras) mounted in the same housing or in different boxes. Cameras can have a known geometric or spatial relationship to each other. For example, as shown in Fig. 3, the 3D sensor 202 can include stereo cameras 302a, 302b mounted on a frame 304. Frame 304 can include one or more mounting devices 306a, 306b (for example, holes, clips or clips ) for mounting the 3D 202 sensor for the combine. Frame 304 may include holes positioned in front of stereo cameras 302a, 302b such that frame 304 does not block the field of view of stereo cameras 302a, 302b. Stereo cameras 302a, 302b may have a known spatial relationship to each other.
[0035] [0035] The 3D sensor 202 can be mounted above the elevator 104 and oriented towards the elevator 104. This can allow the 3D sensor 202 to detect materials carried by the elevator 104 and pass through the detectable zone of the 3D sensor 202 (for example, the field of view of the cameras, associated with the 3D sensor 202). The 3D sensor 202 can transmit signals associated with the detectable area to a computing device (for example, computing device 612 of Fig. 6).
[0036] [0036] The computing device (not shown) can be positioned in any suitable location within the combine. For example, the computing device can be positioned inside the combine's passenger cabin. The computing device can receive the 3D sensor signals 202 and apply the data associated with the signals to one or more algorithms (for example, data processing algorithms). stereo vision algorithms). The computing device can use the algorithms to generate a 3D profile of the scene detected by the 3D sensor 202 (for example, the elevator part 104 and the material within the elevator part 104). In examples where the 3D sensor 202 includes cameras, as the number of cameras that make up the 3D sensor 202 increases, the computing device may be able to generate a more detailed or accurate 3D profile of the scene.
[0037] [0037] As illustrated in Fig. 8, in the examples where the 3D sensor 202 includes cameras, the computing device can generate a 3D profile of the scene based on images captured from cameras. For example, the computing device can apply triangulation to captured images. The triangulation can include a pixel from a first image taken by a first camera to an associated pixel in a second image taken by a second matching camera. The difference between the pixel location in the first image and the corresponding pixel location in the second image may be the disparity. The computing device can generate an 804 image depicting the disparities between the pixels in captured images. Based on each pixel location in the first image, the associated disparity and a distance in real space (for example, in meters) between the first camera and the second camera, the computing device can determine a 3D location in the associated real space with each pixel. The computing device can use 3D locations in real space to generate a 3D 806 profile corresponding to the camera data.
[0038] [0038] In some examples, the computing device can use data from a camera on the 3D sensor to determine a material characteristic. For example, the computing device can determine one or more characteristics associated with the material (for example, color and texture) and one or more characteristics associated with the elevator. In some examples, the computing device may determine that the material is one color (for example, brown) and that the elevator is a different color (for example, silver).
[0039] [0039] The computing device can use a material characteristic to determine the composition of the material. For example, the computing device can use the feature associated with the material and / or the feature associated with the elevator to divide a captured image 800 into different segmented regions 802. In some examples, the computing device may use the 3D profile 806 to improve the division of 800 images captured in different 802 segmented regions. As shown in Fig. 9, after dividing the image into 802 segmented regions, the computing device can generate corrected 904 segmented regions. The 904 corrected segmented regions can include versions of each of the 802 targeted regions that have been corrected for lighting, scale and perspective to improve the invariance of these factors. The computing device may include a feature extraction module 906 configured to extract features 908 from each of the 904 corrected targeted regions. In some instances, the feature extraction module 906 may apply one or more filters to the corrected targeted region, change the dimensions of a corrected targeted region, or be responsible for non-linearities in the corrected targeted region. The computing device can classify (for example, by applying a classifier or regressor to) characteristics 908 to determine garbage 910 in each of the targeted regions 802 of content. In this way, the computing device can analyze the segmented regions 802 in the image to determine the amount of waste (and thus the amount of material desired) in the material in the elevator. For example, the computing device can determine whether the material includes 40% waste and 60% of a desired crop. Based on the relationship, the computing device can apply a scale factor to the net volume of the material. The result can provide a more accurate determination of the desired crop volume, being transported by the 104 elevator.
[0040] [0040] Returning to fig. 2, in some examples, the computing device may operate an exhaust fan (not shown) based on the composition of the material. The extractor can be part of a primary extractor or a secondary extractor in a crop yield system associated with the combine. The material collected by the harvester can be harvested, cleaning system before moving towards the elevator, 104. The hood can rotate to explode in the material, which can remove trash and material debris. In some instances, based on the amount of waste in the material in the elevator 104, the computing device can change the rate at which the exhaust fan rotates. If the amount of waste is above a limit, the computing device can cause the speed of the extractor fan to increase. If the amount of waste is below a limit, the computing device causes the speed of the extractor fan to decrease. In some instances, if the hood turns at a higher speed, the hood can blow more waste out of the collected material, producing a higher crop-to-waste ratio. In this way, the computing device can determine the composition of the material and modulate the speed of the extractor fan to control the amount of waste in the material.
[0041] [0041] In some examples, the combine may include a primary extractor cleaning system. The primary extractor cleaning system can move the primary extractor (for example, primary extractor 110 in Fig. 1) to clean the primary extractor. The computing device can cause the primary extractor, cleaning system to disengage (for example, turn off) if the cleaning system of the primary extractor gets too close to the 3D sensor 202 (for example, a camera associated with the 3D sensor 202). This can reduce the accumulation of dust and debris on the 3D 202 sensor.
[0042] [0042] In some examples, the performance measurement system 200 can measure the characteristics of the elevator 104 when the elevator 104 does not include any material. The resulting measurement can act as a baseline or calibration measurement that can be used to account for the accumulation of dirt and debris and deformations in the form of an elevator 104, which can affect the measurement of the liquid volume. For example, the calibration measurement can be subtracted from the liquid volume measurement to determine the actual volume of material in the elevator 104. A low-pass filter or a probabilistic filter (for example, a Kalman filter) can be applied to the volume material in elevator 104 to increase accuracy.
[0043] [0043] In some examples, the yield measurement system 200 can determine if there is no material in the elevator 104 (for example, to start calibration) by monitoring the volume of the material in the elevator 104 and one or more parameters associated with the combine (for example, ground speed, elevator speed, engine load or combine hitch). For example, the yield measuring system 200 can determine that there is no material in the elevator 104 if the speed at the combine's ground is zero km / h for more than seven seconds, while the elevator 104 is run. As another example, the performance measuring system 200 can determine that there is no material in the elevator 104 if the performance measuring system is disengaged for more than five seconds, while the elevator 104 is executed. Upon the performance measurement system 200 determining that there is no material in the elevator 104, the performance measurement system 200 can automatically measure the characteristics of the elevator 104 (for example, as described above) to generate or update the calibration measurements. This dynamic calibration process can account for accumulation of dirt or deformation in the elevator 104 over time, increasing the accuracy of the performance measurement system 200.
[0044] [0044] There are numerous possible mounting positions for the 3D sensor 202 for displaying the material carried by the elevator 104. In some instances, the 3D sensor 202 can be centered over the entire width of the elevator 104. This can eliminate or reduce the slope in relation to the uneven shape of the material that can accumulate in the elevator 104 blades (for example, blades 502 shown in Fig. 5). In some instances, 3D sensor 202 may be positioned as far away as possible from elevator 104 (for example, while avoiding collision with the primary extractor, when elevator 104 is raised and rotated). The 3D sensor 202 can also be positioned at an angle (for example, 38 °) with respect to the elevator plane that is not perpendicular to the elevator plane. For example, the 3D sensor 202 can be positioned at an angle less than 90 ° to the plane of the elevator. Positioning and tilting the 3D sensor 202 in such a way can improve the results of the image processing algorithms, as described below.
[0045] [0045] For example, the 3D sensor 202 can include at least two cameras. To generate the 3D profile of the scene captured by the cameras, the computing device can apply an algorithm that corresponds to a pixel from a first image from the first camera to a corresponding pixel in a second image from the second camera. This process can be repeated until the pixels in the first image are mapped to corresponding pixels in the second image. This process can be negatively affected, however, by a pattern of holes (for example, holes 504 shown in Fig. 5) that can be on the back plate of elevator 104. For example, instead of a single pixel of the first image of a single pixel from the second mapping image, a pixel from the first image that is associated with a hole in the back plate can be compared to the countless pixels from the second image associated with several holes in the back plate. This can generate errors in the output of the algorithm. In some instances, positioning the cameras at an angle can cause a distortion of perspective in the images captured by the cameras. This can allow the computing device to more easily find original, matching pixels between camera images, thereby minimizing correlation errors. In other examples, the output of the algorithms can be improved, making the pattern of holes in the rear plate irregular, which can reduce correlation errors.
[0046] [0046] In some examples, the performance measurement system 200 may include a lighting system, 204. The lighting system 204 may include one or more lights (for example, incandescent lamps or light-emitting diodes). In some examples, the lighting system 204 may include one or more sensors (for example, a photoelectric cell) to detect an amount of ambient light. The computing device can operate one or more lights based on the ambient light level of one or more sensors detected.
[0047] [0047] In the examples where the 3D sensor 202 includes a camera, the lighting system 204 can illuminate the cameras field of view. This can allow the 200 performance measurement system to be used in environments with little or no ambient light. For example, the lighting system 204 may allow the performance measurement system 200 to be used at dawn, night, dusk, or during adverse weather conditions (for example, cloudy days).
[0048] [0048] Also, in the examples where the 3D sensor 202 includes a camera, the lighting system 204 can reduce the exposure time of the cameras. This can allow the exposure time to be fast enough to reduce or eliminate motion blur in images captured by the cameras, even if elevator 104 (and material in elevator 104) is moving at a high speed. For example, to minimize the exposure time of the cameras, lighting system 204 can provide additional light to the scene captured by the cameras. The additional light can sufficiently illuminate the scene, such that cameras can capture images using short exposure times.
[0049] [0049] The performance measurement system 200 may include one or more blade sensors 206 (for example, Hall effect transducers). In some examples, paddle sensors 206 can be positioned on the side of the elevator, 104. Other examples (for example, the example shown in Fig. 4), paddle sensors 206 can be positioned below the elevator 104. The paddle sensors paddle 206 can detect when a paddle passes the paddle 206 sensors and transmit a sensor signal associated with the computing device.
[0050] [0050] In some examples, the computing device can receive and use the paddle sensor signals to trigger data capture by the 3D 202 sensor. For example, the computing device can receive paddle sensor signals associated with a paddle , passing a paddle sensor 206 and substantially causing the 3D sensor 202 to simultaneously capture data (e.g., causing a camera associated with the 3D sensor 202 to capture an image). Using paddle sensors 206 to trigger data capture, paddles can be in substantially the same position within the detectable area of 3D sensor 202 each time data is captured. This can improve the accuracy of the algorithms used to determine the volume of the material in the elevator 104.
[0051] [0051] In some examples, the computing device may use the signals from the paddle sensor to determine the speed of the elevator 104. For example, a single paddle sensor 206 can be positioned on the side of the elevator, 104. As the paddles pass paddle sensor 206, paddle sensor 206 can detect the paddle and transmit sensor signals associated with the computing device. If the paddles on elevator 104 are spaced evenly, the computing device can divide the amount of spacing between the time period between two sensor signals to determine the speed of the elevator 104. For example, if the time period between two signals from the sensor sensor is 200 ms, and the pad spacing is 20 cm, so the computing device can divide 200 ms by 20 cm to determine that the speed is 1 m / s.
[0052] [0052] As another example, two paddle sensors 206 can be positioned on the side of the elevator, 104. The computing device can determine the speed of the lift based on signals from the paddle sensor of the two paddle sensors 206, even if the blades are not evenly spaced. For example, the computing device can determine the amount of time between when a paddle passes the first paddle sensor 206 (activating the first paddle sensor 206) and when the paddle passes the second paddle sensor 206 (activating the second paddle sensor) shovel 206). If the spacing between the two-paddle sensors 206 is known, the computing device can calculate the elevator speed by dividing the amount of spacing by the amount of time between 206 paddle sensor activations. For example, if the spacing between the two blade sensors 206 is 10 cm, and the time difference between the activation of the two blade sensors 206 is 0.1 s, then the computing device can divide 10 cm by 0.1 s to determine that the speed is 1 m / s.
[0053] [0053] Other methods can additionally or alternatively be used to determine the speed of the elevator 104. For example, an encoder can be positioned on a motor (for example, a hydraulic motor) or gear that moves the elevator 104. The computing device can use signals from the encoder sensor to determine the elevator speed.
[0054] [0054] In the examples where the 3D sensor 202 includes a camera, the computing device can use the position of the paddles in images from the camera to determine the speed of the elevator 104. For example, as shown in Fig. 5, the dashed lines 508 can represent the locations of the 502 paddles, as determined by the camera-based computing device. The computing device can use the locations of the 502 blades and the time between each image to be taken, to determine the speed of the elevator. For example, the computing device can determine the distance that a 502 blade has moved (for example, 10 cm) between two images. The computing device can divide the distance the blade 502 moved by the amount of time between taking the two images (0.1 s) to determine the speed of the elevator (1 m / s). In some instances, the speed of the paddles 502 determined based on the image information may be more accurate than the speed of the paddles 502 determined based on the paddle sensors 206. The computing device may also receive one or more parameters associated with the combine (for example, engine speed) and use the parameters to increase the accuracy of the given lift speed.
[0055] [0055] Returning to fig. 2, the performance measurement system 200 can use the data from the 3D sensor 202 and the elevator speed 104 to determine the material yield. For example, the computing device can use 3D sensor data 202 to generate a 3D profile of the scene within the detectable area of 3D sensor 202. The computing device can use the 3D profile to determine the instantaneous volume of the desired culture in the elevator 104 The computing device can multiply the instantaneous volume by the speed of the elevator to estimate the total volume of the desired culture carried by the elevator 104. The computing device can also receive data (for example, via a CAN bus connection) associated with the combine, such as speed, GPS location and operational status of the combine. You can determine the computing device, based on the total desired harvest volume transported by the 104 elevator and the data associated with the combine, a measure of instant yield of the desired crop.
[0056] [0056] Fig. 6 is a block diagram, representing an example of a 612 computing device for use with a yield measurement system according to an example. The computing device 612 may include a network interface 620. The network interface 620 may represent one or more components that facilitate network connection. Examples include, but are not limited to, wired interfaces such as Ethernet, USB, IEEE 1394, and / or wireless interfaces such as IEEE 802.11, Bluetooth or radio interfaces for accessing cellular networks (for example, transceiver / antenna to access a CDMA, GSM, UMTS or other mobile communications network).
[0057] [0057] The computing device 612 can include an input / output interface 610. The I / O interface 610 can be used to facilitate a connection to the hardware used for data entry (for example, buttons, switches, joysticks, surfaces touch sensitive or other user input devices) or data output (for example, monitors, touch screen monitors or other user interface devices). For example, the 610 I / O interface can be in wired or wireless communication with a 3D 202 sensor (for example, a camera, a 3D laser scanner, an ultrasonic transducer, an interval sensor, a structured light camera, a 3D camera or a LIDAR flash), 206 paddle sensors (for example, Hall effect sensors), a lighting system 204 (for example, to detect low light and illuminate a camera's field of view that is associated with the 3D sensor 202) and other 618 sensors (for example, an axis encoder). In some examples, computing device 612 may be in wired or wireless communication with 3D sensor 202, paddle sensors 206, lighting system 204 and other sensors 618 via network interface 620.
[0058] [0058] The I / O interface may be in communication with, or include, an area controller network interface 611 (CAN). In some examples, the CAN 611 interface may be directly communicating with the 604 bus or another component of the 612 computing device. The CAN 611 interface may be wired (for example, via a CAN bus) or wireless communication with sensors harvester 614 through a CAN bus and configured to receive data from harvester 614 sensors. Harvest 614 sensors can detect various harvester parameters. For example, combine 614 sensors can detect ground speed, GPS location, primary extractor speed, secondary extractor speed, primary fan speed, secondary fan speed, engine rpm, steering, crop processing hitch, operational status, a parameter associated with a tip cutter and other combine parameters. In some instances, harvester 614 sensors can detect external environmental characteristics for the harvester, such as crop width and farm condition (eg, burnt sugar cane and green sugar cane). The sensors of the combine 614 can transmit signals from the sensor to the CAN 611 interface.
[0059] [0059] The computing device 612 includes a processing device 602. The processing device 602 can perform one or more operations for the determination of a measurement of yield of a material. Processing device 602 can execute instructions 608 stored in memory device 606 to perform operations. Processing device 602 may include processing device for one or more processing devices. Non-limiting examples of processing device 602 include a Programmable Farm Door Matrix ("FPGA"), an application specific integrated circuit (“ASIC”), a microprocessor, etc.
[0060] [0060] Processing device 602 can be communicatively coupled to memory device 606 via bus 604. Nonvolatile memory device 606 can include any type of memory device that retains stored information when turned off. Non-limiting examples of the electrically erasable memory device 606 and programmable read-only memory ("EEPROM"), flash memory or any other type of non-volatile memory. In some respects, at least part of the memory device 606 may include a means from which the processing device 602 can read instructions 608. A computer-readable medium may include electronic, optical, magnetic or other storage devices capable of providing the 602 processing device with computer-readable instructions or other program code. Non-limiting examples of computer-readable media include (but are not limited to) magnetic disk (s), memory chip (s), ROM, random access memory ("RAM"), an ASIC, a configured processor, optical storage or any other means so that a computer processor can read the instructions. Instructions can include specific processor instructions generated by a compiler or code interpreter written in any suitable computer programming language, including, for example, C, C ++, C #, etc. In some examples, memory 606 may include an image processing module (not shown). The processing device 602 can use the image processing module to perform one or more image processing tasks on the data received from the 3D meter 202.
[0061] [0061] In some examples, the processing device 602 can receive data from the 3D sensor 202. The processing device 602 can apply the data to algorithms to determine an instantaneous volume of material in a part of the elevator. The processing device 602 can also receive data from the sensors of the paddle sensors 206. The processing device 602 can apply the sensor data from the paddle sensors 206 to algorithms to determine the speed of the elevator. The processing device 602 can multiply the instantaneous volume by the speed of the elevator to estimate the total volume of material in the elevator. The processing device 602 can also receive data from the combine harvester 614. The processing device 602 can integrate the data from the combine with the estimate of the total volume of material to generate a yield map in real time.
[0062] [0062] Fig. 7 is a block diagram of an example of a 700 process for measuring the volume of material transported by elevator according to an example. Some examples may include more, less than or different steps from those described in Fig. 7. The following steps are described with reference to components described above, with respect to the computing device 612 shown in Fig. 6.
[0063] [0063] In block 702, processing device 602 processes data from 3D sensor 202. Processing device 602 can receive data via a wired or wireless connection with 3D sensor 202. Data can be associated with a portion detectable lift from the 3D 202 sensor area.
[0064] [0064] In some examples, the processing device 602 can apply one or more algorithms to the data to process the data. For example, processing device 602 can rectify, perform disparity calculations and perform spot filtering on the data.
[0065] [0065] In block 704, the processing device 602 generates a 3D map (for example, 3D topological map) based on the processed data. The processing device 602 can execute one or more algorithms to produce the 3D map. In some examples, the processing device 602 may apply triangulation to generate a 3D profile of the processed data. The triangulation can include a pixel from a first image taken by a first camera to an associated pixel in a second image taken by a second matching camera. The difference between the pixel location in the first image and the corresponding pixel location in the second image may be the disparity. Based on the location of the pixel in the first image, the disparity and distance in real space (for example, in meters) between the first camera and the second camera, the computing device can determine a 3D location in the real space associated with the pixel . This process can be repeated for all pixels in the first image and the second image to generate a 3D profile corresponding to the processed camera data.
[0066] [0066] In some examples, after generating the 3D profile, the processing device 602 can change the coordinate system of the 3D profile to a frame aligned with the plane of the elevator, such that the y axis follows along the paddle direction, the z axis points from the elevator to the 3D sensor 202 (for example, a camera on the 3D sensor), and X is orthogonal to the y axis and the z axis. The frame source can be centered in the detectable area of the 3D 202 sensor (for example, in the display mode of a 3D 202 sensor camera). The processing device 602 can divide the 3D profile into a grid of squares (for example, 1.25 cm by 1.25 cm in size) associated with the elevator plane. The processing device 602 can discard points that fall outside the grid of squares. For each square in the grid of squares, the processing device 602 can determine the robust maximum height (for example, using a median filter) in relation to the plane of the elevator. Based on the grid of squares and the robust maximum heights, the processing device 602 can generate the 3D map. Each cell on the 3D map can have a robust maximum height associated with it.
[0067] [0067] In block 706, processing device 602 filters the 3D map. The processing device 602 can filter the 3D map to remove noise, erroneous data and irrelevant data. This can result in a more accurate 3D map. For example, processing device 602 can filter out, using a spatial filter, parts of the 3D map associated with calibration data (for example, as discussed with respect to block 709). For example, the processing device 602 can filter portions of the 3D map associated with the elevator blades. This may allow the processing device 602 to generate an instantaneous volume measurement unadulterated by the paddle volume.
[0068] [0068] In block 708, processing device 602 determines whether the yield measurement system is in a calibration mode. For example, processing device 602 can query a location in memory 606 to determine whether the yield measurement system is in calibration mode. If the performance measurement system is in calibration mode, process 700 can continue up to block 709.
[0069] [0069] In block 709, processing device 602 generates calibration data. The processing device 602 can generate the calibration data by analyzing the 3D map to determine the location of the elevator rear plate and deformations (for example, folds or ridges) on the elevator rear plate. The processing device 602 can also analyze the 3D map to determine the characteristics of the blades. The location of the rear lift plate, rear lift deformation plate, and paddle characteristics can be included in the calibration data.
[0070] [0070] In block 710, the processing device 602, stores the calibration data. In some examples, the processing device 602 can store the calibration data in memory 606. In other examples, the processing device 602 can store the data elsewhere, such as in a remote database or a 612 computing device. (for example, by communicating with the remote database or device 612 through the compute network interface 620).
[0071] [0071] If the yield measurement system is not in a calibration mode, process 700 continues until block 712. In block 712, processing device 602 determines the volume of the material based on the filtered 3D map. For example, the processing device 602 can subtract a calibration value associated with the plane of the rear plate of the elevator from the robust maximum height (for example, determined in block 704) of the cell. This can produce the height of the material in the cell. Because each cell can have a known area (for example, 1.25 cm x 1.25 cm = 1.5625 cm2), the processing device 602 can multiply the height of the material in the cell by the cell area to determine the volume of material in the cell. Processing device 602 can repeat this process for all cells to determine the instantaneous total volume of material in the elevator.
[0072] [0072] In block 714, the processing device 602 can estimate the composition of the material detectable by the 3D sensor 202. For example, data from the 3D sensor 202 can be associated with images of the material surface in the elevator. The processing device 602 can determine the characteristics (for example, the sizes, shapes, textures and colors) associated with the surface area of the material. Based on the characteristics, the processing device 602 can determine the relationship between a desired material for foreign materials on the material surface. For example, the processing device 602 can determine that the surface area of the material includes 80% of the desired material and 20% waste. Based on the desired material to foreign material ratio, processing device 602 can apply a scale factor for instantaneous volume measurement (e.g., determined in block 712) to estimate the volume of material associated with the desired material.
[0073] [0073] In block 716, processing device 602 determines the speed of the elevator. In some instances, the processing device 602 may rely on sensor signals from the paddle sensors 206 to determine the speed of the elevator. In other examples, processing device 602 can apply 3D sensor data algorithms 202 to determine the speed of the elevator. In still other examples, the processing device 602 can use sensor signals from other sensors 618 (e.g., an axis encoder) to determine the speed of the elevator.
[0074] [0074] In block 718, the processing device 602 estimates the total instantaneous volume of the desired material in the elevator. In some examples, processing device 602 may apply a scale factor (based on the composition determined in block 714) to the estimated instantaneous volume of block 707. This can generate an estimate of the instantaneous volume of the desired material. The processing device 602 can multiply this instantaneous volume estimate of the elevator speed to determine the total instantaneous volume of the desired material in the elevator.
[0075] [0075] In block 720, processing device 602 determines the instantaneous yield of the desired material. The processing device 602 can determine the instantaneous yield of the desired material based on the total instantaneous volume of the desired material and combine parameters. For example, the processing device 602 can receive data associated with the speed, GPS location, the combine direction from the 614 combine sensors. The processing device 602 can integrate the total instantaneous volume of the desired material (for example, from block 718 ) with the GPS location to determine the instant yield of the desired material at that GPS location. In this way, the processing device 602 can generate a yield map in real time that represents the yield of the desired material at various locations on the farm. In some examples, instantaneous yield can be integrated over time, so that measurement of the accumulated yield of the harvested material can be generated. Instant yield can be further tuned based on the weight of the material, which can be determined when the material arrives at a processing plant or using density or weight measurement methods on the farm.
[0076] [0076] Fig. 10 shows output yield maps 1002a-d by a yield measurement system according to an example. Yield map 1002a depicts a yield map generated using gross yield data. In some examples, one or more algorithms can be applied to the gross yield data to increase the accuracy of the yield map.
[0077] [0077] For example, the collected material may not always flow steadily through the combine. On the contrary, material can accumulate in the combine before being passed to elevator 104. If material accumulates and is passed en masse to elevator 104, yield measurement system 200 may erroneously measure a high yield value. The computing device can account for these "spikes" in yield by tracking the time and space area covered during the accumulation period and spatially, redividing the high yield values. For example, the computing device can apply the times when peak yields and one or more harvester parameters (for example, GPS location or harvester mincer pressure) occurred to algorithms spatially redistribute high yield values. The algorithms can be selected dynamically by a user from a number of options. For example, options may include algorithms to determine: (i) an average flow during the accumulation region, or (ii) a standard flow adjustment rate based on a previous combine pass through the same region on the farm. This can reduce errors on the yield chart due to material accumulation, increasing the precision and accuracy of the yield chart.
[0078] [0078] As another example, the computing device can eliminate data points collected by the harvester when the harvester is not gathering crops. The computing device can determine when the harvester is not gathering crops based on one or more parameters associated with the harvester. For example, the computing device can determine if the harvester is not harvesting crops based on the height of the base cutter (for example, if the height of the base cutter is above or below a threshold), the lift drive hitch , the instantaneous volume of material in the elevator (for example, if the instantaneous volume is above or below a limit), or a GPS location (for example, if the combine is positioned in an area known to harvest or not outside predetermined farm boundary). Yield map 1002b describes an example of a yield map, in which the gross yield data collected by the combine when no crop collection has been eliminated.
[0079] [0079] Yet another example, there may be a time delay between when the material is actually collected from the farm by the combine and when the GPS receiver determines the GPS location associated with the material. This delay time can result in the GPS receiver marking the collected material with a GPS location that is incorrect for a certain distance. The computing device can apply one or more algorithms to correct such inaccuracies. For example, based on data from the 614 combine sensors (for example, the combine's ground speed) and the elevator speed, the computing device can determine the distance the GPS location can be turned off. The computing device can correct the material's GPS location based on this distance. The corrected GPS location can more accurately reflect the precise spatial location on the farm from which the material was collected. Yield map 1002c shows an example of a yield map generated using corrected GPS locations.
[0080] [0080] As another example, spatial filtering can be used to improve the visual representation of performance data (for example, the performance map). Spatial filtering, instead of time series filtering, can allow multidimensional filtering of yield data. The characteristics of spatial filtering can be adjusted to meet the end user's objectives. For example, Kriging filtering or Weighted Distance Inverse Filtering can be used to create a smoothed response of yield data across various zones on the farm where crops are being harvested. The yield map 1002d depicts an example of a yield map generated using corrected GPS locations and spatial filtering.
[0081] [0081] In some examples, the 612 computing device can also produce real-time measurements of mass productivity of machines. The mass productivity of machines can represent the amount of material (for example, in tons) produced per hour. The computing device 612 can determine the mass productivity of machines based on the volume (or yield) of the material and the amount of time required to produce the volume (or yield) of the material. Mass productivity of machines can be used to assess the utilization of capital assets and to optimize support equipment and crop logistics.
[0082] [0082] In some examples, the mass yield and productivity data on machines and can be presented to the operator through a user interface (for example, a dynamic user interface). For example, measurements of mass productivity on machines and a yield map can be presented through a user interface that is the output on a display that is communicating with the 612 computing device. This can facilitate data collection and operational feedback. In some instances, yield data can be used to optimize the transport of harvested crops and the logistics of support equipment associated with the crop. The mass productivity of machines can guide the business rules associated with service and maintenance, operator training production teams (for example, sugar cane production teams). Base Cutter Height control system
[0083] [0083] Certain aspects and characteristics of the present disclosure are directed to a height cutter height control system to measure the height of a base cutter in a combine. The base cutter height control system can include a camera system of at least two cameras. Cameras can capture images of material (for example, stubble) cut by the base cutter and passing under the harvester. The base cutter height control system can apply image, image processing to determine the height of the cutter base. Based on the determined height of the base cutter, the height control system of the base cutter can change the height of the base cutter (for example, to an ideal or preferred height).
[0084] [0084] In some examples, the combine may include both the yield measurement system (described above) and the height cutter height control system (described below).
[0085] [0085] Fig. 11 is a side perspective view of an example of a combine 100 for use with an 1100 base cutter height control system according to an example. The base cutter height control system 1100 can include a 3D sensor 1102 mounted on a combine 100. The 3D sensor can include one or more of a camera, a 3D laser scanner, an ultrasonic transducer, an interval sensor, a structured light camera, a 3D camera and a LIDAR flash.
[0086] [0086] The 3D sensor 1102 can be mounted in any suitable location for the detection of one or more characteristics of an area of land that is being processed, or has been previously processed by the base cutter blades. For example, the 3D sensor 1102 can be positioned under the harvester 100, in front of the harvester 100, or at the back of the harvester 100. In the example shown in Fig. 12, the 3D sensor 1102 is mounted under the harvester 100 and between the rear wheels of the combine. In this example, 3D sensor 1102 can detect area 1202 immediately behind and between the rear wheels of the combine 100.
[0087] [0087] In some instances, the base cutter height control system 1100 may include a debris barrier 1104. Debris barrier 1104 can be positioned near, near or in front of 3D sensor 1102 to prevent dirt or debris from entering the detectable area (for example, the field of view) of 3D sensor 1102. In some instances, debris barrier 1104 may include curtains or partitions (for example, plastic or metal partitions).
[0088] [0088] Fig. 13 is a perspective view of an example of a 3D sensor 1102 mounted on the bottom of a 1300 combine for use with a height cutter height control system according to an example. The 3D sensor 1102 can include two or more cameras. The cameras can be configured in the same way as the cameras described in relation to Fig. 2-3.
[0089] [0089] The 3D sensor 1102 can be mounted on the combine at an angle (for example, an angle of inclination). The angle can be between 0 degrees and 90 degrees in relation to the plane (for example, the geometric plane) of the ground. In some examples, the angle can be configured to allow 3D sensor 1102 to display an area that is not directly perpendicular to the ground. For example, as shown in Fig. 14, 3D sensor 1102 can be mounted to the combine at an angle of 13.30 degrees.
[0090] [0090] Fig. 15 is a perspective view of an example of a 1502 lighting system for use with a base cutter height control system according to an example. The lighting system 1502 can include one or more lights 1504a, 1504b and one or more sensors to detect the amount of ambient light (not shown). For example, the lighting system 1502 may include a light 1504a positioned on one side end of the 3D sensor 1102 and another 1504b light positioned on the other side end of the 3D sensor 1102. The lights 1504a, 1504b can be separated or integrated with a housing of the 3D sensor 1102. In some examples, lights 1504a, 1504b may include a protective case configured to protect lights 1504a, 1504b from dust and debris. The protective case can include plastic, rubber, acrylic or metal.
[0091] [0091] The base cutter height control system can operate the lights 1504a, 1504b. In some instances, the base cutter height control system can operate lights 1504a, 1504b based on the detected ambient light level of one or more ambient light sensors. This can allow the base cutter height control system to be used in times of low to no ambient light. This can also allow cameras associated with the 3D 1102 sensor to have exposure times short enough to reduce or eliminate motion blur, even if the collector is moving at a high speed.
[0092] [0092] The base cutter height control system can include a computing device (for example, similar to the 612 computing device in Fig. 6). In some instances, the computing device can receive signals from the ambient light sensor sensors in the lighting system and, based on the sensor signals, operate the lights.
[0093] [0093] The computing device can also receive data from the 3D sensor 1102. The data can be associated with the cut material (for example, stubble) passing under or near the harvester. The computing device can apply one or more algorithms (for example, stereo vision algorithms) to the data (or a filtered version of the data) to generate 2.5-dimensional (3D) soil map. Based on the 3D map, the computing device can determine a soil profile.
[0094] [0094] In some examples, the computing device may use the soil profile to determine the (i) lateral position line; (ii) height of the line above the ground; (iii) material height; (iv) lateral position of the material; and (v) height adjustment of the desired base cutter. In some instances, the computing device may generate a geographic map that represents material height data associated with various locations in the field.
[0095] [0095] Fig. 16 is a block diagram of an example of a 1600 process for controlling the height of the base cutter according to an example. Some examples may include different steps, more or less than the steps described in Fig. 16. The following steps are described with reference to components described above, with respect to the computing device 612 shown in Fig. 6.
[0096] [0096] In block 1602, processing device 602 processes data from 3D sensor 202. Processing device 602 can receive data through a wired or wireless connection with 3D sensor 202. Data can be associated with material within the detectable area of the 3D sensor 202. The data can be, or can be converted, to the 3D profile by the processing device 602. For example, the processing device 602 can convert the 3D profile data using triangulation or any other method .
[0097] [0097] The processing device 602 can apply one or more algorithms to the data to process the data. In some instances, the processing device 602 can filter the data to remove noise and erroneous data. For example, the computing device can apply a filter to the data to reduce the effect of a material in the air (for example, a piece of sugar cane) obscuring the camera's field of view associated with the 3D 202 sensor.
[0098] [0098] In some examples, the processing device 602 can calibrate the data. For example, the combine can be driven on level ground. Processing device 602 can use data from 3D sensor 202 to determine the position and orientation of 3D sensor 202 in relation to the ground. The processing device 602 can store the position and orientation and use the position and orientation to calibrate the data.
[0099] [0099] In block 1604, processing device 602 generates a 3D map based on processed data. The 3D map can include multiple cells. Each cell can include a rectangular area, corresponding to an associated area in real space. For example, as shown in Fig. 17, 3D map 1702 can include multiple cells. The center of each cell can be represented with a dot 1706. In some examples, each cell on the 3D map can be associated with parameters associated with the cell area in real space. For example, a parameter can include the height of the material (for example, the height of the stubble) and the standard deviation of the height of the material, associated with the cell area in real space. The 1704 limits of the 3D map can define the limits of the 3D map.
[0100] [00100] To generate the 3D map, the processing device 602 can map the 3D data points to the corresponding cells. For example, a 3D data point can be mapped to a specific 1708 cell. In some examples, multiple 3D data points can be mapped to the same particular cell 1708.
[0101] [00101] In block 1606, processing device 602 determines an upper 3D map and a lower 3D map. The bottom 3D map can include a 3D map associated with features on the plane or close to ground level. The top 3D map can include a 3D map associated with features in the plane along or near the top of the material (for example, the top of the stubble). The processing device 602 can determine the top 3D map and the bottom 3D map based on the 3D map (e.g., block generated in 1604). The processing device 602 can determine the top 3D map using a high percentile (e.g., 90%) of the height of the 3D data points in relation to the ground. Processing 602 can determine the bottom 3D map using a low percentile (for example, 10%) of the height of the 3D data points in relation to the ground.
[0102] [00102] In block 1608, processing device 602 determines the position of the lateral line and the height of the line (for example, the lateral position and height of an area in real space, associated with a line). The processing device 602 can determine the line position based on the bottom 3D map. The processing device 602 can fit a model (for example, a parabola) with the bottom 3D map. For example, as shown in Fig. 18, processing device 602 can fit a parabola 1810 with the bottom map 3D 1802. Processing device 602 can determine the lateral position of the line, based on the location of the vertex of parabola 1810 in the 1808 axis. In some examples, if the 3D sensor 202 is positioned in the middle of the wheels (or tracks) of the combine, the center of the x-axis can represent the lateral mean of the wheels (or tracks) of the combine.
[0103] [00103] In some examples, processing device 602 can determine the height of line 1806 based on the vertex 1808 of parable 1810. For example, processing device 602 can determine the height of line 1806 based on the height of vertex 1808 (for example, 0.1 m) of the parable.
[0104] [00104] In block 1610, processing device 602 determines the height of the material. The processing device 602 can determine the height of the material based on the top 3D map. For example, as shown in Fig. 18, processing device 602 can generate the upper 3D map 1804. To determine the height of the material, processing device 602 can subtract the height of the lower 3D map 1802 from the height of the map. Superior 3D 1804 at the vertex 1808 (for example, 1.5 m). The difference between the height of the lower 3D map 1802 and the height of the upper 3D map 1804 at vertex 1808 may be the height of the material.
[0105] [00105] In block 1612, processing device 602 determines a new height of the cutter base (or orientation). The processing device 602 can determine the new height of the base cutter based on the height of a given material (for example, the 1610 block), the current height setting of the cutter base and user input data. For example, a user can enter a desired material height on the 612 computing device (for example, via a keyboard or touch screen monitor). The processing device 602 can determine the new base cutter height (or orientation) based on the difference between the desired material height and the determined material height. In some examples, the processing device 602 may use proportions, hysteresis or other algorithms to generate a new base cutter height. For example, the processing device 602 can determine the new height of the base cutter as a proportion of the difference between the desired height of the material and the determined height of the material.
[0106] [00106] In some examples, the processing device 602 can determine a new height of the base cutter based on a calibration factor. For example, a user can manipulate the height of the base cutter until the material cut by the combine reaches a desired height. The user can have the processing device 602 determine the height of the material (for example, using the method described in blocks 1602-1610) and store the height of the material. This height can be used as a calibration factor. The processing device 602 can compare the height of the material cut by the harvester later for the calibration factor. Processing device 602 can determine the new height of the base cutter based on the difference between the height of the material and the calibration factor.
[0107] [00107] In some examples, when determining a new height of the base cutter, the processing device 602 causes the height of the base cutter to change to the new height of the base of the cutter. In this way, the height control system of the base cutter can act as a feedback loop.
[0108] [00108] In block 1614, processing device 602 can determine the number of pieces of material fallen (for example, cut), the length of pieces of material fallen and the number of roots of the material based on data from the 3D sensor 202. For example, the processing device 602 can extract characteristics from the data. The processing device 602 can extract characteristics from the filtering data, determine associated texture data, and determine the 2D and 3D edges associated with the data. The processing device 602 can also determine the entropy, frequency and first order and derivatives associated with the highest order data. Processing device 602 classifies (for example, by applying a classifier or regressor to) the characteristics for detecting and determining (i) the number of fallen pieces of material, (ii) the length of the fallen pieces of material and (iii ) the number of visible roots of the material.
[0109] [00109] In some examples, the processing device 602 can estimate the damage resulting from a base cutter height that is very low (for example, with respect to a desired or ideal base cutter height). If the height of the cutter base is too low, the base cutter can penetrate the material and extract line or cause damage to the root of the material. Processing device 602 can estimate damage based on one or more of the various fallen pieces of material, the length of the fallen pieces of material and the number of visible roots of the material.
[0110] [00110] Fig. 19 is an example of a 1900 geographic map output by a base cutter height control system according to an example. In some examples, the 1900 geographic map can be presented through a user interface that is output on a screen. This can facilitate data collection and operational feedback.
[0111] [00111] The 1900 geographic map can include different colored regions 1902, 1904, 1906. In some examples, each of the different colored regions 1902, 1904, 1906 on the map can be associated with different material heights. For example, the colored region 1902 can be associated with geographical areas with high material. The colored region 1904 can be associated with geographical areas with low material. The colored region 1906 can be associated with geographical areas with medium-height material. In other examples, each of the different colored regions 1902, 1904, 1906 on the map can be associated with the number of pieces of fallen material, the length of the pieces of fallen material, or the number of roots of the material. For example, colored region 1902 can be associated with geographical areas with a high number of pieces of fallen material. The colored region 1904 can be associated with geographical areas with a low number of fallen pieces of material. The colored region 1906 can be associated with geographical areas with an average number of pieces of fallen material.
[0112] [00112] The height control system of the base cutter can generate the colored regions, 1902, 1904, 1906, based on the (i) height of the material, a number of fallen pieces of material, a length of a fallen piece of the material, or a number of roots of the material; and (ii) data from one or more harvester sensors (for example, harvester sensors 614 of Fig. 6). For example, on the base cut height control system to determine the material height for a given location, the base cutter height control system can receive GPS data (for example, a latitude and longitude) associated with that particular location. The base cutter's height control system can locate the GPS on the 1900 geographic map and update the color associated with the GPS location based on material height. As another example, on the base cutter height control system by determining the number of fallen pieces of material for a specific location, the base cutter height control system can receive GPS data (for example, a latitude and longitude ) associated with that particular location. The base cutter's height control system can locate the GPS on the 1900 geographic map and update the color associated with the GPS location based on the number of pieces of material fallen.
[0113] [00113] The above description of certain modalities, including illustrated modalities, has been presented for purposes of illustration and description only and is not intended to be exhaustive or limit the dissemination of the precise forms disclosed. Numerous modifications, adaptations and respective uses will be evident for those skilled in the art without departing from the scope of the disclosure.
权利要求:
Claims (37)
[0001]
Cutting height control and measurement system for a basic harvester, characterized by the fact that it comprises: a stereo camera system (302a, 302b) configured to detect an elevator area (104) in a combine (100) and to transmit images associated with the area, the elevator having an elevator rear plate with an irregular pattern of openings configured to reduce volume determination errors; a processing device (602) in communication with the stereo camera system (302a, 302b); and a memory device (606) in which executable instructions are stored by the processing device (602) to make the processing device: receive the images; generate a three-dimensional (3D) profile (806) of the area, at least in part, by analyzing the images using triangulation, the 3D profile including a grid of squares, each square in the grid of squares having a predefined area and a corresponding height in relation to an elevator plan; filtering, from the 3D profile, data points corresponding to a shovel (502) in the elevator to produce a filtered 3D profile; and determine a volume of a material (102) in the elevator based on the filtered 3D profile.
[0002]
System according to claim 1, characterized by the fact that it comprises a sensor configured to detect a combine parameter (100) and transmit a signal associated with the parameter to the processing device (602), in which the memory device ( 606) further comprises instructions executable by the processing device to make the processing device: receive the signal; and generate a three-dimensional (3D) map based on the signal; determine the volume of the material (102) based on the 3D map; and determine a material yield (102) based on the volume of the material and the signal.
[0003]
System according to claim 2, characterized by the fact that the parameter consists of a GPS location, speed, direction or crop processing hitch.
[0004]
System according to claim 2, characterized in that the memory device (606) further comprises instructions executable by the processing device (602) to make the processing device: generate a corrected GPS location by modifying a GPS location associated with distance yield; spatially filter the yield to generate a filtered yield value; and generate a yield map based on the filtered yield value and the corrected GPS location.
[0005]
System according to claim 1, characterized in that the memory device (606) still comprises instructions executable by the processing device (602) to make the processing device: determine the composition of the material (102) by analyzing a color and texture of the material in the images, where the composition is comprised of a first percentage of the material that is attributable to a target material and a second of the material that is attributable to a foreign material ; and apply a scale factor to the material volume (102) based on the composition to determine a calibrated volume.
[0006]
System according to claim 5, characterized in that the memory device (606) still comprises instructions executable by the processing device (602) to make the processing device: change a fan speed associated with a primary extractor (110) or a secondary extractor based on composition.
[0007]
System according to claim 5, characterized in that the memory device (606) still comprises instructions executable by the processing device (602) to make the processing device: operate the stereo camera system (302a, 302b) when the elevator does not include the material (102); receive images from the stereo camera system; determine a baseline volume based on the images; and modify the calibrated volume based on the base volume.
[0008]
System according to claim 1, characterized by the fact that it still comprises a lighting system (204) that includes: a sensor configured to detect an amount of ambient light, and at least one light positioned to illuminate the elevator area (104), wherein the memory device (606) further includes instructions that are executable by the processing device (602) for cause the processing device to operate at least one light to illuminate the elevator area based on the detected amount of ambient light.
[0009]
System according to claim 5, characterized in that the memory device (606) further includes instructions that are executable by the processing device to make the processing device determine the composition of the material (102) by: determining that a first part of an image corresponds to the elevator (104) and a second part of the image corresponds to the material based on a first color of the elevator being different from a second color of the material.
[0010]
System according to claim 1, characterized in that it comprises a paddle sensor (206) configured to detect the position of a paddle (502) in the elevator (104) and transmit a signal to the processing device (602), in which the memory device (606) further comprises instructions executable by the processing device to make the processing device: receive the signal; determine the speed of the elevator (104) based on the signal; and operate the stereo camera system (302a, 302b) based on the signal.
[0011]
System according to claim 1, characterized in that the memory device (606) still comprises instructions executable by the processing device (602) to make the processing device: determine the position of a paddle (502) in the elevator (104) based on images from the stereo camera system (302a, 302b) by identifying the paddle in an image from the stereo camera system (302a, 302b); and determine the speed of the elevator (104) based on the position of the blade.
[0012]
System according to claim 1, characterized by the fact that triangulation involves correlating a first set of pixels in a first image of a first camera in the stereo camera system (302a, 302b) to a second set of pixels in a second image of a second camera in the stereo camera system (302a, 302b).
[0013]
Method, characterized by the fact that it comprises: receive, through a processing device (602), images from a stereo camera system (302a, 302b), with images of an area of an elevator (104) of a combine (100), the elevator having a rear plate of the elevator with an irregular pattern of openings configured to reduce errors in determining volume; generate, through the processing device (602), a three-dimensional (3D) profile of the elevator area, at least in part, analyzing the images using triangulation, the 3D profile (806) including a grid of squares, each square in the grid of squares having a predefined area and corresponding height in relation to an elevator plane; filtering, through the processing device (602) and the 3D profile, data points corresponding to a paddle (502) in the elevator to produce a filtered 3D profile; and determine, by the processing device (602), a volume of a material (102) in the elevator based on the filtered 3D profile.
[0014]
Method according to claim 13, characterized by the fact that it still comprises: calibrate the 3D profile (806) based on the calibration data, where the calibration data is determined using other images from the stereo camera system (302a, 302b) generated before the material was positioned in the elevator (104), and where the calibration data considers (i) a position of a rear plate of the elevator, (ii) a deformation in the rear plate of the elevator and (iii) a characteristic of a blade (502) in the elevator.
[0015]
Method according to claim 13, characterized by the fact that it further comprises: determining the speed of the elevator (104); and multiply the speed of the elevator by the volume to determine the total volume of material (102) in the elevator.
[0016]
Method according to claim 13, characterized by the fact that it still comprises: determining the composition of the volume of material (102) in the elevator (104); per divide the images into segmented regions based on a visual characteristic of the material in the elevator; generate corrected targeted regions by correcting the targeted regions for at least one lighting, scale or perspective; perform the extraction of characteristics in the corrected segmented regions to determine one or more characteristics in the corrected segmented regions; and classifying one or more characteristics as garbage or a target material to determine the composition of the material in the elevator; and modify the volume of material in the elevator based on the composition.
[0017]
Method according to claim 16, characterized in that it further comprises changing a fan speed associated with a primary extractor (110) or a secondary extractor based on the composition of the volume of the material (102).
[0018]
Method according to claim 13, characterized by the fact that it still comprises: receiving, from a sensor, a signal associated with a combine parameter (100); and generate a yield map based on the volume of the material (102) and the parameter.
[0019]
Method according to claim 18, characterized by the fact that generating the yield map comprises: generate a yield map based on the volume of the material (102); generate a corrected GPS location by modifying the parameter by a distance; spatially filter the yield to generate a filtered yield value; and generate a yield map based on the filtered yield value and the corrected GPS location.
[0020]
Combine harvester (100), characterized by the fact that it comprises: an elevator (104) having an elevator rear plate with an irregular pattern of openings configured to reduce errors in determining volume; and a stereo camera system (302a, 302b) coupled to a combine (100) and configured to detect an area associated with the elevator and transmit images associated with the area; where the combine (100) is configured to: generate a three-dimensional (3D) profile (806) of the elevator area, at least in part, analyzing the images using triangulation; adjust a coordinate plane of the 3D profile to align with a plane of the elevator so that the coordinate plane of the 3D profile includes (i) a first axis that is oriented along a direction of the movement of the blade (502) in the elevator, (ii) a second axis that is oriented from the elevator towards the stereo camera system (302a, 302b) and (iii) a third axis that is orthogonal to the first axis and the second axis; determining a material composition (102) by analyzing the 3D profile of the elevator area; and determine a volume of a crop included in the material based on the composition of the material.
[0021]
Combine harvester (100) according to claim 20, characterized by the fact that it still comprises a blade sensor (206) configured to detect the position of a blade in the elevator (104) and transmit a usable blade sensor (206) signal for: determine the speed of the elevator; and operate the stereo camera system (302a, 302b).
[0022]
Combine harvester (100) according to claim 20, characterized by the fact that it still comprises a lighting system (204) that includes: a sensor configured to detect an amount of ambient light; and a plurality of lights positioned to illuminate the elevator area (104) and separate from the stereo camera system (302a, 302b), wherein the harvester (100) is configured to operate the plurality of lights to illuminate the elevator area based on the amount of ambient light detected using the sensor.
[0023]
Combine harvester (100) according to claim 20, characterized by the fact that it still comprises a sensor configured to detect a harvester parameter (100) and transmit a sensor signal associated with the parameter that is used to: determine a cultural yield based on the volume of the material (102) and the second signal.
[0024]
Combine harvester (100) according to claim 23, characterized in that the yield is useful for generating a yield map.
[0025]
Combine harvester (100) according to claim 20, characterized in that the material composition (102) is useful for changing the speed of a fan associated with a first extractor (110) or a second extractor.
[0026]
System, characterized by the fact that it comprises: a combine (100) comprising: base cutter; and a stereo camera system (302a, 302b) positioned under the combine (100), the stereo camera system (302a, 302b) being configured to capture an area under or behind the combine (100) and to transmit images the area; a processing device (602) in communication with the stereo camera system (302a, 302b); and a memory device (606) in which instructions executable by the processing device are stored to cause the processing device to: generate a three-dimensional (3D) profile (806) of the area by analyzing the images using a stereo vision algorithm; and modify a height of the base cutter based on the 3D profile.
[0027]
System according to claim 26, characterized in that the memory device (606) still comprises instructions executable by the processing device (602) to make the processing device: determine a lateral position of a stubble line or a stubble height based on the 3D profile of the area; and modify a height of the base cutter based on the lateral position of the stubble line or the first height of the stubble.
[0028]
System according to claim 26, characterized by the fact that it still comprises a debris barrier (1104) configured to protect the stereo camera system (302a, 302b) from debris or damage.
[0029]
System according to claim 26, characterized by the fact that it still comprises a lighting system (204) that includes: a sensor configured to detect an amount of ambient light; and at least one light positioned under the combine (100) to illuminate the area.
[0030]
System according to claim 26, characterized in that the memory device (606) comprises instructions executable by the processing device (602) to make the processing device: receive images from the stereo camera system (302a, 302b); determine a calibration factor when analyzing the images; and determine the height of the base cutter based on the calibration factor.
[0031]
System according to claim 26, characterized in that it comprises a sensor configured to detect a harvester location (100) and transmit a second signal associated with the location and in which the memory device (606) further comprises executable instructions by the processing device (602) to cause the processing device to: analyze the images to determine information about a material (102) represented in the images, the information includes the height of the material, a number of pieces of material fallen out of the length of a piece of material fallen, or a number of roots of the material; and generate a geographic map based on the combine's location (100) and information.
[0032]
Method, characterized by the fact that it comprises: receiving, by a processing device (602), images from a stereo camera system (302a, 302b) positioned under a combine, comprising a base cutter, images showing an area that is under or behind the combine (100 ); generate, through the processing device (602), a three-dimensional (3D) profile of the area analyzing the images using a stereo vision algorithm; and modify, by the processing device, a height of the base cutter based on the 3D profile.
[0033]
Method according to claim 32, characterized by the fact that it still comprises: filtering images from the stereo camera system (302a, 302b) to generate filtered images; determining a lateral position of a material line (102) depicted in the images or a height of the material line depicted in the images using the filtered images.
[0034]
Method, according to claim 33, characterized by the fact that it still comprises: calibrate the height of the material line (102) based on the calibration data, where the calibration data is determined using (i) other images from the stereo camera system (302a, 302b) and (ii) and a user input .
[0035]
Method according to claim 33, characterized by the fact that it still comprises: receiving, from a sensor, a signal indicating the location of the combine (100); and generate a geographic map based on the location of the combine (100) and the height of the material line (102).
[0036]
Method according to claim 35, characterized by the fact that it still comprises: analyzing the images to determine information about a material (102) represented in the images, the information including a number of pieces of fallen material, a length of a piece of fallen material, or a number of roots of the material; and generate a geographic map based on the combine's location (100) and information.
[0037]
Method according to claim 33, characterized in that filtering the images comprises filtering the images to reduce an effect of debris in the air in the 3D profile (806).
类似技术:
公开号 | 公开日 | 专利标题
BR102014027364B1|2021-01-12|cutting height measurement and control system for a basic harvester, method, and harvester
US10151839B2|2018-12-11|Systems and methods for determining crop yields with high resolution geo-referenced sensors
Yu et al.2004|Automatic detection of harvested trees and determination of forest growth using airborne laser scanning
US8340402B2|2012-12-25|Device and method for detecting a plant
US20180084708A1|2018-03-29|Agricultural work machine for avoiding anomalies
CN104714547A|2015-06-17|Autonomous gardening vehicle with camera
US10534086B2|2020-01-14|Systems and methods for determining crop yields with high resolution geo-referenced sensors
US11175170B2|2021-11-16|Estimating yield of agricultural crops
Reiser et al.2018|Iterative individual plant clustering in maize with assembled 2D LiDAR data
Kim et al.2021|Stereo-vision-based crop height estimation for agricultural robots
EP3732941A1|2020-11-04|Residue monitoring and residue-based control
Chen et al.2021|Navigation path extraction for greenhouse cucumber-picking robots using the prediction-point Hough transform
US10959377B2|2021-03-30|Crop scanner
Cho et al.2014|Vision-based uncut crop edge detection for automated guidance of head-feeding combine
Swanson et al.2010|A multi-modal system for yield prediction in citrus trees
WO2015193822A1|2015-12-23|Method and device for measuring vegetation cover on farmland
Wang et al.2012|Design of crop yield estimation system for apple orchards using computer vision
KAIZU et al.2012|Grass height and yield estimation using a three-dimensional laser scanner
CN107656287B|2019-08-16|A kind of Boundary Extraction device and method of the crudefiber crop row based on laser radar
CN207992810U|2018-10-19|A kind of field automatic travelling device for detecting plant growth condition
Jiang et al.2016|An imaging-based system for high-throughput phenotyping of cotton plants under field conditions
Sun2019|3D Imaging and Automated Point Cloud Analysis for In-Field Plant Mapping
강길송2018|On-the-go Embedded System for Spatial Mapping of Lettuce Fresh Weight in Plant Factory
Griepentrog2019|Crop plant reconstruction and feature extraction based on 3-D vi-sion
Tsoulias et al.2020|In-situ detection of apple fruit using a 2D LiDAR laser scanner
同族专利:
公开号 | 公开日
AU2014256411B2|2018-09-27|
BR102014027364A2|2016-08-09|
CN104718874B|2020-03-03|
CN104718874A|2015-06-24|
IN2014DE03130A|2015-07-03|
AU2014256411A1|2015-05-21|
US20150124054A1|2015-05-07|
US10371561B2|2019-08-06|
引用文献:
公开号 | 申请日 | 公开日 | 申请人 | 专利标题

DE19726917A1|1997-06-25|1999-01-07|Claas Selbstfahr Erntemasch|Device on agricultural machinery for contactless scanning of contours extending over the ground|
US20020014116A1|1999-02-26|2002-02-07|Campbell Ronald H.|Methods and systems for measuring crop density|
US6690413B1|1999-04-21|2004-02-10|Michael S. Moore|Tractor-trailer viewing system|
WO2001060718A2|2000-02-17|2001-08-23|Bintech. Lllp|Bulk materials management apparatus and method|
DE10214648A1|2002-04-02|2003-10-16|Claas Selbstfahr Erntemasch|Measuring device on an agricultural machine|
CA2430737C|2003-06-02|2011-12-20|Centre De Recherce Industrielle Du Quebec|Method and apparatus for estimating surface moisture content of wood chips|
DE102004038404B4|2004-08-07|2021-01-07|Deere & Company|Device for the automatic adjustment of the cutting height of a header for harvesting stem-like plants|
US7707003B2|2005-02-23|2010-04-27|Bin Tech L.L.L.P.|Method and apparatus for tracing and blending commingled non-liquid bulk materials|
CN1695419A|2005-06-23|2005-11-16|上海交通大学|Intelligent system for measuring yield based on remote data transmission|
EP2004339B1|2006-03-31|2012-01-25|Thomas Valerio|Method and apparatus for sorting fine nonferrous metals and insulated wire pieces|
GB0714942D0|2007-08-01|2007-09-12|Cnh Belgium Nv|A biomass cleaner improvements in corp harvesting machine and related methods|
US20100269470A1|2009-04-24|2010-10-28|Price Randy R|Optical Yield Monitor for a Sugar Cane Harvester|
US8306750B2|2009-08-18|2012-11-06|Plant Health Care, Inc.|Computer-automated spatial analysis|
DE102010017688A1|2010-07-01|2012-01-05|Claas Selbstfahrende Erntemaschinen Gmbh|Device for detecting and determining the composition of bulk material|
US9348111B2|2010-08-24|2016-05-24|Apple Inc.|Automatic detection of lens deviations|
CN102379189B|2010-08-31|2013-06-26|中国农业机械化科学研究院|Grain yield distribution information measurement method and device|
CN102741882B|2010-11-29|2015-11-25|松下电器(美国)知识产权公司|Image classification device, image classification method, integrated circuit, modeling apparatus|
US8649940B2|2012-02-10|2014-02-11|Deere & Company|Method and stereo vision system for managing the unloading of an agricultural material from a vehicle|
US8988426B2|2012-06-05|2015-03-24|Apple Inc.|Methods and apparatus for rendering labels based on occlusion testing for label visibility|
JP6270361B2|2013-07-16|2018-01-31|株式会社キーエンス|3D image processing apparatus, 3D image processing method, 3D image processing program, computer-readable recording medium, and recorded apparatus|US11212962B2|2013-02-20|2022-01-04|Deere & Company|Field condition determination|
US9961832B2|2014-06-13|2018-05-08|Deere & Company|Control of base cutter orientation in sugarcane harvesters|
US10234837B2|2015-09-28|2019-03-19|Deere & Company|Adaptive performance targets for controlling a mobile machine|
US10303677B2|2015-10-14|2019-05-28|The Climate Corporation|Computer-generated accurate yield map data using expert filters and spatial outlier detection|
US9788485B2|2015-10-23|2017-10-17|Deere & Company|Variable speed elevator control system|
US10645876B2|2016-03-01|2020-05-12|Deere & Company|System and method for controlled cleaning in a cane harvester|
WO2018035082A1|2016-08-15|2018-02-22|Raptor Maps, Inc.|Systems, devices, and methods for monitoring and assessing characteristics of harvested specialty crops|
CN108200804B|2016-12-19|2021-05-04|上海交通大学|Visual sensing grain flow online detection device and detection method|
GB201621879D0|2016-12-21|2017-02-01|Branston Ltd|A crop monitoring system and method|
SE543160C2|2017-01-16|2020-10-13|Tracy Of Sweden Ab|A method for determining and managing origin identifation of logs|
US10317272B2|2017-05-17|2019-06-11|Deere & Company|Automatic wireless wagon detection apparatus and method|
US10827676B2|2018-01-29|2020-11-10|Deere & Company|Monitor and control system for a harvester|
US11006577B2|2018-02-26|2021-05-18|Cnh Industrial America Llc|System and method for adjusting operating parameters of an agricultural harvester based on estimated crop volume|
US11178818B2|2018-10-26|2021-11-23|Deere & Company|Harvesting machine control system with fill level processing based on yield data|
US11240961B2|2018-10-26|2022-02-08|Deere & Company|Controlling a harvesting machine based on a geo-spatial representation indicating where the harvesting machine is likely to reach capacity|
US11234366B2|2019-04-10|2022-02-01|Deere & Company|Image selection for machine control|
US11079725B2|2019-04-10|2021-08-03|Deere & Company|Machine control using real-time model|
CN110089260A|2019-04-15|2019-08-06|江苏大学|A kind of the cereal flow monitoring method and monitoring system of the defeated grain of scraper-type|
US10829033B1|2019-08-20|2020-11-10|Cnh Industrial America Llc|Automatically controlled header work lights|
US11231317B2|2019-08-20|2022-01-25|Cnh Industrial America Llc|Systems and methods for filtering sensor signal interference deriving from powered components of a header|
WO2021051150A1|2019-09-20|2021-03-25|Seeing Systems Pty Ltd|Systems and methods for gathering data relating to crops and for managing crop growing operations|
法律状态:
2016-08-09| B03A| Publication of an application: publication of a patent application or of a certificate of addition of invention|
2018-11-06| B06F| Objections, documents and/or translations needed after an examination request according art. 34 industrial property law|
2020-06-23| B06U| Preliminary requirement: requests with searches performed by other patent offices: suspension of the patent application procedure|
2020-11-10| B09A| Decision: intention to grant|
2021-01-12| B16A| Patent or certificate of addition of invention granted|Free format text: PRAZO DE VALIDADE: 20 (VINTE) ANOS CONTADOS A PARTIR DE 31/10/2014, OBSERVADAS AS CONDICOES LEGAIS. |
优先权:
申请号 | 申请日 | 专利标题
US201361898600P| true| 2013-11-01|2013-11-01|
US61/898600|2013-11-01|
US14/527,152|US10371561B2|2013-11-01|2014-10-29|Yield measurement and base cutter height control systems for a harvester|
US14/527152|2014-10-29|
[返回顶部]