SYSTEMS FOR DETECTING NON-CROPER OBJECTS

Abstract

Example devices, systems, and methods are provided herein. In some examples, a system or method can be provided to determine the presence of one or more objects located in a portion of an agricultural field situated near a harvesting head of the working machine, to determine one or more characteristics of the one or more objects, to determine whether the one or more objects are desired objects based at least in part on one or more characteristics of the one or more objects, to determine one or more object identifiers of the one or more objects, and to transmit instructions to one or more systems to adjust the performance of the agricultural machine, at least in part, based on one or more object identifiers.

Description

AREA OF INVENTION

[0001] This application relates to systems for determining the operating conditions of a grain harvester. BACKGROUND

[0002] Various agricultural vehicles perform a wide range of agricultural tasks, such as combine harvesters and rakes, which harvest a variety of crops. Depending on the plant or other factors, the cutting units used for harvesting can have significantly different geometries, plant engagements, and separation devices. Examples of cutting unit platforms include rotary mower conditioners and belt headers. SUMMARY

[0003] Example devices, systems, and methods are provided herein. In some examples, a system for monitoring a portion of an agricultural field during the operation of an agricultural machine may be provided. The system may include one or more processors. The system may include one or more sensors configured to transmit sensor data to one or more processors. The system may include a storage device coupled to one or more processors, the storage device containing instructions which, when executed by the one or more processors, cause the one or more processors to determine the presence of one or more objects located in a portion of an agricultural field situated near a harvesting head of the machine.to determine one or more characteristics of one or more objects, to determine whether the one or more objects are desired objects that are based at least partially on one or more characteristics of the one or more objects, to determine one or more object identifiers of the one or more objects, and to transmit instructions to one or more systems to control the performance of the agricultural machine at least partially on the basis of one or more object identifiers. BRIEF DESCRIPTION OF THE DRAWINGS Fig. Figure 1 is a schematic view of a grain harvester from the left side. Fig. Figure 2 is a schematic representation of a control system of the grain harvester from Figure 2. Fig. Figure 3 is a right-side view of a working machine including optical sensors. Fig. Figure 4 is a flowchart showing exemplary steps for monitoring part of an agricultural field. Fig. Figure 5 is a flowchart that shows exemplary steps for monitoring a part of a working machine. DETAILED DESCRIPTION

[0004] When operating a machine, such as a grain harvester, it is desirable to determine which objects are permitted to pass through and / or under the machine. It is also desirable to determine the condition of one or more parts of the machine. Accurate detection of objects, such as non-harvest items, and machine conditions, such as wear, can be advantageous for precisely and promptly adjusting the machine's operation. This disclosure relates to detection systems for identifying non-harvest items and machine conditions, such as wear or damaged components.

[0005] One problem encountered with previous image-based systems for detecting non-crop objects and wear is the detection of objects in the immediate vicinity of the machine (e.g., directly in front of the machine, on, in, or behind a combine harvester header). Some objects may appear or be detectable just before interaction with the combine harvester header and / or the combine itself. Another problem encountered with previous image-based systems is that the condition of certain parts of the machine may be missed using conventional monitoring methods.

[0006] The present disclosure provides exemplary detection systems that may include image-based detection aspects for recognizing non-crop objects and working machine conditions. By providing one or more sensors, such as optical sensors, in conjunction with other aspects of the detection systems, the working machine can monitor and detect objects and conditions that might otherwise go undetected.

[0007] With reference to Fig. Figure 1 now depicts a working machine, specifically a grain harvester 102 in the form of a combine harvester. The grain harvester 102 includes a control unit 104, which controls various aspects of the grain harvester 102 and / or facilitates its operation.

[0008] As shown, the example grain harvester 102 comprises a chassis 106 with ground-penetrating wheels 108 or tracks. The wheels 108 are rotatably mounted on the chassis 106 and engage the ground to drive the grain harvester 102 in a direction of travel T. An operator's cab 110, also mounted on the chassis 106, houses an operator as well as various devices for controlling the grain harvester 102, such as one or more operator input devices 112 and / or display devices 114, which are described below.

[0009] The wheels 108 and other components of the grain harvester 102 are powered by an internal combustion engine 116 or another energy source. The internal combustion engine 116 can be operated based on inputs from the operator and / or the control unit 104.

[0010] A harvesting header 118 is mounted on the front of the chassis 106 of the grain harvester 102 for cutting and collecting crops from a field. The harvesting header 118 is supported by an inclined conveyor 120, which is pivotally mounted on the chassis 106. The harvesting header 118 comprises a frame 122 that supports a cutter bar 124, which extends substantially along the length of the harvesting header 118 and serves to cut crops close to the ground. The harvesting header 118 may also include a mechanism for collecting the cut material from the cutter bar 124. In this example, the harvesting header 118 includes an auger 130 to convey the cut crop toward the center of the harvesting header 118. Other examples may include one or more conveyor belts.The harvesting header 118 can include an actuator 132, which serves to reposition the harvesting header 118 relative to the ground and / or in the forward and rearward directions. The inclined conveyor 120 can, for example, include an inclined conveyor belt (not shown) to transport cut crop from the harvesting header 118 into the body of the grain harvester 102.

[0011] After passing through a guide drum or feed accelerator 134, the crop material from the inclined conveyor 120 reaches a generally rearward-facing threshing device or separator 136. Other embodiments may include laterally oriented or other threshing devices (not shown). In the illustrated embodiment, the separator 136 comprises a rotor 138 on which various threshing elements are mounted. The rotor 138 rotates over one or more threshing concaves 140 or separating concaves equipped with grates or sieves, so that the crop material passing between the rotor 138 and the threshing concaves 140 is at least partially separated into grain and chaff (or other “material other than grain” (MOG)). The threshing concaves 140 can be opened and / or closed by one or more actuators 142 (shown schematically).The actuators 142, as well as other actuators assigned to the threshing concaves 140, can be operated by commands from the operator and / or the control unit 104. The MOG (mound of grain) is carried to the rear and released between the rotor 138 and the threshing concaves 140. Most of the grain (and some of the MOG) separated in the separator 136 falls through openings in the threshing concaves 140.

[0012] Agricultural material flowing through the threshing concaves 140 falls (or is actively fed) into a cleaning subsystem (or cleaning shoe) 144 for further cleaning. The cleaning subsystem 144 comprises a blower 146 driven by a motor 148, which generally generates a rearward airflow, as well as a sieve 150 and a pre-sieve 152. The sieve 150 and the pre-sieve 152 are suspended relative to the chassis 106 by an actuating arrangement 154, which may include pivot arms and rocker arms mounted on discs (or other devices). While the blower 146 blows air over and through the sieve 150 and the pre-sieve 152, the actuating arrangement 154 may cause a reciprocating motion of the sieve 150 and the pre-sieve 152 (e.g., by moving the rocker arms).The movement of the sieve 150 and the pre-sieve 152 in the combine harvester with the airflow of the blower 146 generally causes the lighter chaff inside the grain harvester 102 to be blown upwards and backwards, while the heavier grain falls through the sieve 150 and the pre-sieve 152 and collects in a trough 156 for clean grain near the bottom of the grain harvester 102.

[0013] A screw conveyor 158 for clean grain, located in the trough 156, conveys the material to one side of the grain harvester 102 and deposits the grain at the lower end of an elevator 160. The clean grain, lifted by the elevator 160, is conveyed upwards until it reaches the upper outlet of the elevator 160. The clean grain is then released from the elevator 160 and falls into or is deposited in a grain tank 162.

[0014] Most of the grain entering the cleaning subsystem 144 is not conveyed to the rear, but instead flows downwards through the pre-sieve 152 and then through the sieve 150. Smaller MOG particles are blown out of the rear of the grain harvester 102 by air from the blower 146 to the rear of the sieve 150 and the pre-sieve 152. Larger MOG particles and grain are not blown out of the rear of the grain harvester 102, but fall off the cleaning subsystem 144.

[0015] Heavier material conveyed to the rear end of the pre-screen 152 exits the grain harvester 102. Heavier material conveyed to the rear of the screen 150 falls into a pan and is then carried downwards by gravity into a trough 164 in the form of residues, typically a mixture of grain and MOG (moisture-based grain). A return screw conveyor 166, located in the trough 164, conveys the grain residues to one side of the grain harvester 102 and into a return conveyor 168. The return conveyor 168 communicates with the return screw conveyor 166 at an inlet opening of the return conveyor 168, where grain residues are received for transport to further processing. An outlet opening (or other discharge point) 170 is provided at an upper end of the return conveyor 168 (e.g., for return to the threshing stage).

[0016] In a passive return implementation, the return conveyor 168 transports the return material upwards and deposits it at a front end of the rotor 138 for re-threshing and separation. A discharge drum 172 is provided for removing material from the rotor 138. The now separated residue is released behind the grain harvester 102 to fall to the ground in a windrow, or it is fed to a residue subsystem 174, which may include a chopper 176 and a spreader 178, for chopping by the chopper 176 and spreading on the field by the spreader 178. Alternatively, in an active residue implementation, the return conveyor 168 can convey the grain residues upwards to an additional threshing unit (not shown) which is separate from the separator 136 and where the grain residues are further threshed before being fed to the main crop flow at the front of the cleaning subsystem 144.

[0017] The grain harvester 102 can include one or more image acquisition sensors 180 arranged on one or more image acquisition areas 182 within or around the grain harvester 102. In addition to the aspects described below, each image acquisition sensor 180 is positioned to capture images of a flow of harvested material within its respective image acquisition area of ​​the grain harvester 102. These images can be processed by the controller 104 as described below to measure objects to be determined in an agricultural field or to monitor the condition of the working machine. Each image acquisition sensor 180 can be any suitable sensor type, including radar sensors, camera sensors, LiDAR sensors, infrared sensors, near-infrared sensors, and any other sensor suitable for providing images for spectral analysis.

[0018] There are several suitable locations within the grain harvester 102 for the arrangement of the image acquisition sensors 180. Some of these are in Fig. The diagram shows the locations schematically, with different specific locations marked by a suffix "a", "b", etc. It should be noted that the locations are shown in general terms only. Within a given area of ​​the machine, the image acquisition sensor should be positioned and oriented to best capture a moving airflow of air, entrained grain, and MOG (moisture, ground, and debris). Some locations may be used to assess grain quality at the site. Other locations may be used to assess grain loss. Some locations may be relevant for both grain quality and grain loss.

[0019] In a first example, an image acquisition sensor 180a can be arranged in the area of ​​the threshing device or the separator 136. The data from sensor 180a can be representative of the grain quality in the area of ​​sensor 180a, especially if they are located in the upstream sections of the threshing device or the separator 136. The data from sensor 180a can be representative of the grain loss into the residue system if the sensor 180a is arranged at the downstream end of the threshing device or the separator 136.

[0020] In another example, an image acquisition sensor 180b can be arranged in the area of ​​the cleaning shoe 144. The data from sensor 180b in the area of ​​the cleaning shoe can be representative of the grain quality of the finished separated grain product.

[0021] In another example, an image acquisition sensor 180c can be arranged in the area of ​​the residue processing subsystem 174. The data from sensor 180c can be representative of the grain loss through the residue processing subsystem 174.

[0022] In yet another example, an image acquisition sensor 180d can be positioned in the area of ​​the residue processing system 164, 166, or at point 180e further along the return conveyor 168. The data from sensors 180d or 180e can be representative of the grain quality at these locations.

[0023] Additional image acquisition sensors can be arranged at various points sequentially downstream at locations 180f, 180g, and 180h in the area above the sieve 150 and the pre-sieve 152. The one located further downstream can provide data representative of the grain quality at these locations. The other location located further downstream can provide data representative of the potential grain loss from the rear of the machine 10, where the chaff stream is blown out of the machine.

[0024] As in Fig. As shown schematically in Figure 2, the grain harvester 102 comprises a control system 200, including the control unit 104. The control unit 104 can be part of the machine control system of the grain harvester 102, or it can be a separate control module. The control unit 104 can, for example, be mounted in a control panel located at the operator station 110. The control unit 104 is configured to receive input signals from the various sensors. The signals transmitted from the various sensors to the control unit 104 are described in Fig. 2 schematically represented by lines connecting the sensors to the controller, with an arrowhead indicating the flow of the signal from the sensor to the controller 104.

[0025] For example, image signals 180aS-180hS from the image acquisition sensors 180a-180h are received by the controller 104. The controller 104 can also receive a fan speed signal 202S from the fan speed sensor 202, which is assigned to the fan 146. The controller 104 can also receive an air velocity signal 204S from the air velocity sensor 204, which may be located in the grain harvester 102 next to one or more image acquisition areas 182. There may be several air velocity sensors 204 present in the grain harvester 102.

[0026] Similarly, the controller 104 generates control signals to control the operation of various actuators of the grain harvester 102. These actuators can be connected, for example, to various subsystems of the grain harvester that influence grain loss within the machine. These actuators can include, for example, the threshing concave actuators 142, the blower motor 148, and the actuator assembly 154 associated with the sieve 150 and the chaffer 152, to name just a few.

[0027] The controller 104 comprises a processor 206, a computer-readable medium 208, a database 210, and an input / output module or control panel 212 with, or connected to, the aforementioned display 114. The aforementioned input / output device 112, such as a keyboard, joystick, or other user interface, is designed to allow the human operator to input instructions to the controller. It is understood that the controller 104 described herein may be a single controller with all the described functions, or it may comprise multiple controllers, with the described functionality distributed among the multiple controllers.

[0028] Various operations, steps, or algorithms, as described in connection with the controller 104, can be embodied directly in hardware, in a computer program product 218, such as a software module executed by the processor 206, or in a combine harvester consisting of both. The computer program product 218 can be located in RAM, flash memory, ROM, EPROM, EEPROM, registers, a hard disk, a removable disk, or any other form of computer-readable medium 208 known in the field. An exemplary computer-readable medium 208 can be coupled to the processor 206 so that the processor can read information from and write information to the memory / storage medium. Alternatively, the medium can be integrated into the processor.The processor and the medium can be contained within an application-specific integrated circuit (ASIC). The ASIC can be located in a user terminal. Alternatively, the processor and the medium can be separate components within a user terminal.

[0029] The term “processor,” as used herein, may refer to at least general or specific processing devices and / or logic as they might be understood by a person skilled in the art, including, but not limited to, a microprocessor, a microcontroller, a state machine, and the like. A processor may also be implemented as a combination of computer devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.

[0030] Data storage on a computer-readable medium 208 and / or a database 210 may, in certain embodiments, include a database service, cloud databases, or the like. In various embodiments, the computer network may include a cloud server and, in some implementations, be part of a cloud application, with various functions, as disclosed herein, being distributed by their nature between the computer network and other distributed computing devices. One or all of the distributed computing devices may be implemented as at least one vehicle control unit, server device, desktop computer, laptop, smartphone, or other electronic device capable of executing instructions. A processor (e.g., a microprocessor) of the devices may be a generic hardware processor, a specialized hardware processor, or a combination thereof.

[0031] Fig. Figure 3 shows an example of a working machine 300 (e.g., a grain harvester 102) that includes a variety of optical sensors 302 (which may include image acquisition sensors 180). In some examples, the working machine 300 may include one or more optical sensors 302, such as one or more cameras (e.g., optical or visual radiation cameras or red, green, blue (RGB) cameras), LiDAR sensors, radar sensors (e.g., long-range terahertz radar, millimeter radar, ultra-wideband radar, frequency-modulated continuous wave radar (FMCW), ground-penetrating radar), ultrasonic sensors, thermal sensors (e.g.,Thermal imaging cameras), stereo cameras, laser vibrometers, infrared nuclear magnetic resonance (NMR) cameras, infrared short-wave infrared (SWIR) cameras, infrared terahertz sensors, three-dimensional sensors and three-dimensional cameras, as well as combine harvesters thereof, among other types of sensors suitable for capturing or generating one or more images or data corresponding to one or more images of the field of view.

[0032] As discussed below, in some examples, one or more processors, such as the controller 104 and / or the processor 206 and the optical sensor 302, can be configured to evaluate at least partially acquired information from the optical sensor 302, for example, at the pixel level or based on a collection or range of pixels, among other things, as a basis for evaluation. Such an evaluation can, for example, be based at least partially on a color or light level that is or is not present in one or more areas or pixels in the acquired information, as well as on associated depth information.

[0033] The examples shown herein describe methods and systems for monitoring a portion of an agricultural field to determine whether non-crop objects could interfere with the operation of a working machine (e.g., a grain harvester 102). As described above, monitoring the field condition can be advantageous in determining whether there are any non-plant objects that could impede the operation of a working machine. In some examples, the working machine can employ monitoring techniques and methods to determine whether there are any non-crop objects located directly in front of a header and / or between a header and a combine harvester that could impede the operation of the working machine 300.

[0034] Fig. Figure 3 shows an exemplary machine 300, which includes an optical sensor 302, such as the image acquisition sensor 180. The optical sensor 302 can be positioned to monitor parts of a cutting unit, such as the cutting unit 118, and at least a portion of the crop in front of the cutting unit 118. The optical sensor can also be positioned to detect and locate non-plant objects on the ground between the cutting unit 118 and the harvesting machine 102. For example, the optical sensor can be arranged on a front part of the machine 300 and positioned to detect an area between the header 118 and the rear sections of the machine 300.In other examples, the optical sensor can be arranged separately from the working machine 300, for example on an aerial drone, and communicatively coupled to the working machine 300, and positioned to detect an area between the harvesting head 118 and the rear sections of the working machine 300. Although the examples presented herein involve the detection of non-harvested objects, in some examples one or more processors can be configured to monitor goods such as grain lying on the ground under / behind the harvesting head 118 at position 302, which can also indicate potential grain loss, as described in relation to the image acquisition sensors 180 described above.

[0035] The examples shown herein describe methods and systems for monitoring a part of the machine 300 to determine the machine's condition. Condition monitoring of machines can be advantageous in determining whether machine conditions such as wear necessitate measures like repairs or operational adjustments.

[0036] As described above, sensor data can be received from one or more sensors to determine the working machine condition. For example, the one or more sensors can be optical sensors 302, vibration sensors, or thermal sensors. In some examples, an optical sensor 302 can be positioned to detect and locate working machine conditions, such as combine harvester header conditions.

[0037] In some examples, the one or more sensors may comprise a variety of sensors positioned to determine the condition of one or more working machine components. For example, the working machine components may be one or more reel and harvester header components, such as row unit divider tips, augers, conveyor belts, dividers / points, separating devices (e.g., blades, knife headers, sections, knives, or guards), fluid delivery hoses and reservoirs, wiring harnesses, lights, and indicators. As such, in some examples, data is received to determine the working machine operating conditions of a combine harvester header, based at least partially on one or more of these components.In such examples, the sensor data can be data that, either alone or in combination with one or more other data sets, is suitable for determining the characteristics of one or more header components in a combine harvester. This data could include, for example, visual data, thermal data, vibration data, or GPS data. In such examples, the data can be used to determine whether components of a combine harvester header are in various states (e.g., operational, compromised, not operational, etc.).

[0038] In some examples, operating conditions may include conditions involving slight wear of components that do not impair operation. Compromised conditions may include conditions that could impair the operation of the working component over time. Non-operational conditions may include conditions that impair the operational quality of the working machine at a given time or that could further impair the working machine. Each of the conditions can trigger a unique response from one or more processors of the working computer. In some examples, one or more processors may indicate one or more conditions by means of a warning, such as a display and / or an audible warning. The conditions described above are examples, and one or more other conditions may be detected by the processor.

[0039] In some examples, data such as temperature, vibration, and rotational speed can be used to determine specific component conditions (e.g., component wear or failure). For instance, measured heat data can indicate friction, friction, component failure (e.g., bearings, stalling or slippage, material bonding, deposits, or accumulation on or under the belt). In some examples, heat dissipation in the form of thermal radiation and / or waves can be detected and can be determined to indicate particle shedding, oil, lubricant, or liquid spraying / drip / accumulation. In some examples, heat dissipation from various components can be measured (e.g., bearings, stalling or slippage, material bonding, deposits, or accumulation on or under the belt).

[0040] In some examples, sensor data can be used and / or combined to increase the accuracy of condition detection. For instance, certain combine harvester conditions can cause unique vibration patterns for a working machine when the machine is in operation. In such cases, optical and vibration data can be combined to identify and verify the presence and characteristics of the combine harvester's and / or other machine components' conditions.

[0041] Fig. Figure 4 is a flowchart showing example steps 400 for monitoring a portion of an agricultural field. The procedure can be implemented by a system for monitoring a portion of an agricultural field, which may include one or more processors, such as the Controller 104 and the Processor 206.

[0042] At 402 in the Fig. In the example process shown in Figure 4, one or more processors receive sensor data from one or more sensors configured to transmit this data to one or more processors. As described above, the sensor data can be received from one or more sensors to determine the operating conditions of the machine. For example, one or more sensors can be optical sensors (e.g., camera, infrared camera, etc.), vibration sensors, or heat sensors. In some examples, where the data is received to determine the location of objects within a desired area of ​​a combine harvester header, the sensor data can be data that, either alone or in combination with one or more other data points, is suitable for determining the properties of one or more objects.In such examples, the data can be processed to determine whether objects scattered across a combine harvester header are specific types of objects, such as desired objects (e.g., crop objects) or non-crop objects. Non-crop objects might include objects that should not be picked up by a combine, such as stones, debris (e.g., sticks / branches, fencing material, and dirt), or tools. In some examples, the sensor data can be combined to increase the accuracy of object detection. For instance, certain types of debris can cause unique vibration patterns for a working machine as the machine interacts with the debris by picking it up or rolling over it. In such examples, optical and vibration data can be combined to determine the presence and properties of objects, such as...to identify and verify non-crop objects.

[0043] In a 404 error, one or more processors are assigned to the processor in the Fig. The example procedures shown in section 4 indicate the presence of one or more objects located within a portion of an agricultural field, defined by a specific area, such as a predetermined distance in the forward direction of a cutterbar of the agricultural machine (e.g., 50 m, 30 m, 5 m). In some examples, the detected area may be a lateral region relative to the direction of the machine's movement (e.g., 50 m, 30 m, 5 m). In example processes where objects are detected within the specified area, one or more sensors, such as optical sensors, can transmit data indicating that one or more objects are located within a desired environment of the agricultural machine. The desired environment may, for example, be a radius of approximately 0–100 m.In some examples, as described above, the radius may include a space in the forward direction of a combine harvester header and a space between a combine harvester header and a working machine.

[0044] At 406, determine in the Fig. In the example process shown in Figure 4, one or more processors analyze one or more features of one or more objects. For example, one or more processors can analyze optical data showing specific outlines of non-crop objects and compare the data to a record in a database. In some examples, the database may contain sample objects of both crop and non-crop types for comparison. In some examples, the objects may be analyzed by a machine learning process, such as a computer vision model, which can determine which features the object contains. Sample features may include shape size, color, temperature, or other relevant features for determining an object type (e.g., texture, reflectance, orientation, position, arrangement, etc.).

[0045] At 408, determine in the Fig. In the example process shown in Figure 4, one or more processors determine whether one or more objects are desired or undesired, at least partially based on one or more features of the one or more header components. For example, the one or more processors may use an algorithm or machine learning process to determine that the objects could be grain, raw material, or milled material, thus classifying the object as a desired object that can pass through the combine harvester under normal operating conditions. In some examples, the processor may determine that the objects are undesired, such as stones, debris, tools, or other non-harvestable objects.

[0046] At 410 in the Fig. In the example process shown in Figure 4, one or more processors determine one or more object identifiers (or object identifiers) of one or more objects. For example, one or more processors can determine, through machine learning or comparison with one or more objects in an object library, that an object contains an identifier. Identifiers can include a shape, color, size, texture, temperature, or other attributes associated with the object. In such examples, one or more processors can determine that the object has an identifier that can indicate whether the object is a desired object, a crop object, or a non-crop object.

[0047] In some examples, one or more processors can be configured to classify one or more objects. For example, one or more processors can classify the object as a non-harvest object, a harvest object, an unwanted object, a desired object, or other appropriate classifications. The classification can be based on one or more object identifiers, other features, or combinations of features. For example, an object can be classified according to sensory aspects such as an optical identifier, a thermal signature, and a vibration signature.

[0048] At 412 in the Fig. In the example process shown in Figure 4, one or more processors transmit instructions to one or more systems to adjust the performance of the agricultural machine based, at least in part, on the one or more object identifiers and the classification of the one or more objects. For example, one or more processors may determine that one or more objects are non-crop objects that are undesirable for inclusion in the combine harvester. In such examples, the one or more processors may send instructions to an actuator to stop a pickup movement of the machine. The instructions may also be sent to systems such as electronic actuators or motor controls of the machine to slow down or stop the machine during a harvesting operation.

[0049] In some examples, one or more processors can determine a predicted interaction of one or more objects, based at least partially on classification information. The classification information might indicate, for example, that objects are likely to damage the machine if swallowed. In other examples, one or more processors might predict that the objects could unbalance the combine harvester or block its path. Therefore, one or more processors can account for operational deviations based on the detected objects and their associated predictions.In some examples, information can be transferred from one or more processors to a central processor, such as a central server, or a remote server that can process data such as captured data and for the operation and prediction of one or more working machines.

[0050] In some examples, one or more processors can be configured to receive a field position of one or more objects. In such examples, one or more processors can use proximity data, such as relative proximity, mapping, or GPS data, to determine a location where one or more objects are detected. The location data can be used to determine one or more future predicted object locations, based at least partially on the sensor data. In some examples, one or more processors can send the data to a central server or analyze the data on an internal processor. In some examples, one or more processors can use the data to determine where objects are located and to correlate the objects with specific field positions for future operations.In some examples, the data can also be correlated with locations where objects are more likely to appear, such as at a field boundary or a road intersection. In such examples, one or more processors can use logic-based analysis and / or machine learning to determine a course of action and / or predict future object encounters.

[0051] Fig. Figure 5 is a flowchart showing example steps 500 for monitoring a part of the working machine 300. The process can be implemented by a system for monitoring a part of the working machine 300, which may include one or more processors, such as the controller 104 and the processor 206.

[0052] At 502 in the Fig. In the example process shown in Figure 5, one or more processors receive sensor data from one or more sensors configured to transmit sensor data to one or more processors. The sensor data can be from a sensor such as an optical sensor like a camera, an infrared camera, or another optical sensor. In some examples where the data is used to determine a condition of one or more features of a combine harvester header, the sensor data can be data suitable for determining various hardware features, either alone or in combination with one or more other data sets.

[0053] In some examples, the data may be thermal, vibration, or GPS data used to determine if components exhibit anomalies such as optical or vibration defects. Anomalies may include attributes indicating that the driven machine is in a condition outside its original operating parameters, such as an attachment. This could include damage, misalignment, missing components, displacement, or other characteristics that deviate from a standard operating condition. In some examples, non-standard operating conditions may include situations where two or more components of the driven machine are in contact with each other that would not be in contact during the machine's assembly (e.g., a drum with a hose or structure, or auger with a structure or wear / underride guard).For example, a harvester header may have a curved component that differs visually from a non-bent component and causes different vibration modes than a non-bent component.

[0054] In some examples, sensor data can be combined to determine a condition more precisely. For instance, certain components can cause specific vibration patterns when they are distorted or bent to collide with other parts of the machine. In such cases, optical and vibration data can be combined to identify and verify the presence of machine anomalies.

[0055] At 504 in the Fig. In the example process shown in Figure 5, one or more processors receive an indication of the presence of one or more anomalies on a machine. In examples where anomalies are detected on the machine, one or more sensors, such as optical sensors, can transmit data indicating the presence of one or more anomalies. Anomalies can be any conditions that deviate from a specific typical operating condition. For example, a surface of the machine might be bent, resulting in visual indications, vibrations, or other conditions that differ from standard operating conditions. In such examples, the sensors can transmit visual, vibration, or other condition data to a processor to determine the physical nature of the anomalies.

[0056] At 506 in the Fig. In the example process shown in Figure 5, one or more processors determine one or more features of one or more components of a machine attachment. For example, the one or more processors can analyze optical data that determines certain visual properties of machine components, at least partially, based on reference images. In other examples, the machine components can be analyzed, at least partially, using machine learning to determine visual features. In some examples, these features can include shape size, color, orientation, geometry, profile, temperature, heat signature, or other relevant features related to determining the state of a machine part.

[0057] In some examples, one or more processors use machine state data correlated with the machine's operating conditions. This machine state data may include examples of machine components in various states and repair conditions for comparison with components in operation. Component state data might include categories such as blade state, surface, and hose state. In some examples, component state data might encompass the cutting device's condition, component orientation during crop collection, geometry position, surface profile, fluid supply system integrity, and so on.

[0058] At 508 in the Fig. In the example process shown in Figure 5, one or more processors determine whether the working machine properties are desired working machine properties (e.g., working machine state features) that are at least partially based on one or more features of the one or more header components. For example, the one or more processors can determine, through an algorithm or machine learning procedure, that the components may have one or more abnormal properties, so that the one or more abnormal properties can be classified or determined to be a desired or undesired property.

[0059] At 510 in the Fig.In the example process shown in Figure 5, one or more processors transmit instructions to one or more systems to adjust the performance of the working machine, at least partially, based on one or more characteristic features. For example, one or more processors may determine that one or more features are undesirable anomalies (e.g., machine condition features indicating anomalies). In such cases, the one or more processors may send instructions to a combine harvester drive to stop the drive to prevent ingestion and / or to release a collection system, such as a header or an attachment carried by the working vehicle, to prevent the working machine's threshing or processing system from being confronted with unwanted material.

[0060] The instructions can also be sent to systems such as the motors or control units of the working machine to slow down or stop the working machine during a harvesting operation in order to prevent undesirable operating conditions and to allow the conditions to be rectified.

[0061] In some examples, one or more processors can determine the future integrity of a working machine, based at least in part on feature classification information. This classification information might include basic component features such as blade shape, hose temperature, and vibration modes of the working machine. In some examples, one or more processors can predict certain further component anomalies based on previous component anomalies. For example, one or more processors might predict that a bent blade could deflect debris in such a way that the resulting deposits could further bend other components of the working machine. Therefore, one or more processors can account for deviations in operation based on these predictions.In some examples, information can be transferred from one or more processors to a central processor, such as a central server, or a remote server that can process data such as captured data, and for the operation and prediction of one or more working machines.

Claims

[1] System for monitoring a part of an agricultural field during the operation of an agricultural machine, the system comprising: one or more processors; one or more sensors configured to transmit sensor data to one or more processors; a storage device coupled to one or more processors, wherein the storage device contains instructions which, when executed by at least one or more processors, cause the one or more processors to: to determine the presence of one or more objects located in a part of an agricultural field situated near a harvesting head of the working machine; Determining one or more characteristics of one or more of the objects; Determine whether one or more objects are desired objects that are at least partially based on one or more features of the one or more objects. Determine one or more object identifiers of one or more objects; and Transmitting instructions to one or more systems to adjust the performance of the agricultural machine, at least partially, based on one or more object identifiers. [2] System according to claim 1, wherein one or more sensors comprise an optical sensor. [3] System according to claim 1, wherein the one or more processors are further configured to classify the one or more objects. [4] System according to claim 3, wherein the one or more processors are further configured to determine a predicted interaction of the one or more objects, based at least partially on classification information. [5] System according to claim 1, wherein one comprises several sensors, a vibration sensor and an optical sensor, and wherein the processor is configured to determine one or more objects based at least partially on optical and vibration data from the vibration sensor and the optical sensor. [6] System according to claim 1, wherein the environment comprises a section of soil in the agricultural field between the header and the working machine. [7] System according to claim 1, wherein the environment comprises a soil section of the agricultural field in front of the attachment in respect of a direction of travel of the working machine. [8] System according to claim 1, wherein the processor is configured to receive a field position of one or more objects. [9] System according to claim 1, wherein the processor is configured to transmit sensor data to a central processor, the central processor being configured to determine one or more future predicted object locations based at least partially on the sensor data. [10] System according to claim 9, wherein the processor is configured to determine a predicted object at least partially on the basis of a received field position of the working machine. [11] System according to claim 10, wherein the received field position is a field boundary. [12] System according to claim 9, wherein the central processor is configured to transmit one or more future health states to one or more processors. [13] System according to claim 12, wherein one or more processors are configured to provide operating instructions to one or more systems of the working machine.

Images