Agricultural tool boom leveling controller and method

Abstract

A system and method for controlled leveling of a multi-wing spray boom assembly is described. A multi-loop cascading controller arrangement includes a master controller that presents actuation control commands to an actuator controller for causing positioning of actuators of the multi-wing boom assembly to reduce a determined current position error of the assembly relative to a determined profile. The master controller operates according to a master controller repetitive cycle period to generate positioning commands to be executed by the actuator controller, wherein the actuation commands are presented for implementation by the actuator controller within the master controller repetitive cycle period. The actuator controller executes an actuator controller loop independently of the master controller and according to an actuator controller repetitive cycle period having a duration independent of the master controller repetitive cycle period to generate control signals for the actuators of the multi-wing boom assembly.

Description

[0001] Cross Reference to Related Applications

[0002] This patent application is a non-provisional application of U.S. Provisional Application 63 / 133,621, filed January 4, 2021, entitled “AGRICULTURAL IMPLEMENT BOOM LEVELING CONTROLLER AND METHOD,” and claims priority thereto, which is expressly incorporated by reference in its entirety, including any references contained therein. TECHNICAL FIELD

[0003] The present disclosure relates generally to systems and methods for controlling boom leveling and / or orientation in agricultural spraying implements. More specifically, the present disclosure relates to a control system in which, during operation of an agricultural spraying implement, a plurality of sensor readings are processed via real-time actuator operation to control boom level and / or orientation. BACKGROUND

[0004] In agriculture, implements including booms extending laterally from a hitch point to a tractor (see Figures 1A-1C ) have been installed with sprayer assemblies including a plurality of nozzles for performing various operations such as fertilizing crops, killing insects and weeds, applying plant protection products, etc.

[0005] One challenge faced by farmers when using agricultural spraying implements is that they need to pull the implement through a field containing emerged plants using a tractor. Driving a heavy tractor through these fields can damage the emerged plant structure as well as the root structure below the surface of the field. This damage can be reduced by maximizing the lateral reach of the spraying boom carrying the plurality of nozzles to minimize the number of sweeps the tractor makes through the field. As such, it is not uncommon for spraying booms to extend 30 feet or more in each lateral direction from the hitch point (60 feet total). It is further noted that wider booms also reduce the amount of operating time required to complete a spraying mission in a field.

[0006] The relatively large span of the sprayer assembly boom presents a problem when operating the agricultural spraying implement on a field having uneven terrain. During operation, it is generally desirable for the boom to extend in a plane parallel to the field / plants. This relative positioning ensures even distribution of the sprayed material on the field / plants. The assurance of even distribution, in turn, helps to use less material (if an excess supply of material would otherwise be required to ensure that all plants are applied with sufficient material) and ensures that over-application of material is prevented (if, due to uneven terrain, the boom / nozzles are too close to the crops / plants / land).

[0007] Importantly, the dynamic boom leveling coupled with the distance sensor helps to maintain a tighter relationship between the boom / nozzle and the field / plants. As a result, material can be applied even in conditions of moderate wind without undue risk that the sprayed material will be carried away by the wind (i.e., achieving the desired reduction in material drift). To achieve this sprayer performance / characteristic, the boom height is sometimes required to be close to 25-35 cm above the ground. When spraying, it is also desirable to move as fast as possible.

[0008] Additionally, different boom orientations can be required depending on crop pattern changes caused by field shape or slope. One such example is during a headland turn. In this case, nothing is being sprayed. For maneuverability, it can be beneficial to slightly raise the left and right wings (variable geometry) of the boom assembly. This temporary raising of the wings increases the ground clearance, thereby increasing the safe buffer distance from the land to accommodate the possible centripetal forces acting on the boom during a sharp turn that causes the boom wings to tilt.

[0009] Effective dynamic boom wing level control is practically impossible to achieve by manually raising and lowering the wings in response to terrain changes and other situations that arise during operation (requiring raising / lowering of one or both boom wings). For this reason, automatic control has been proposed. However, the implementation of automatic control has its own challenges, ensuring that the boom wings are raised and / or lowered in the appropriate amount and in a timely manner. SUMMARY

[0010] Described herein is a method for controlling a leveled multi-wing boom assembly by a multi-loop cascaded controller arrangement, the arrangement comprising a master controller giving actuation control commands to an actuator controller for causing positioning of actuators of the multi-wing boom assembly to minimize a determined current position error of components of the multi-wing boom assembly relative to a determined profile of a field through which the multi-wing boom assembly is currently passing. The method comprises: executing, by the master controller, a master controller loop according to a master controller repetitive cycle period to generate positioning commands to be executed by the actuator controller, wherein the actuation commands are given for the actuator controller to implement within the master controller repetitive cycle period. The method further comprises: executing, by the actuator controller independently of the master controller and according to an actuator controller repetitive cycle period having a duration independent of the master controller repetitive cycle period, an actuator controller loop to generate control signals for the actuators of the multi-wing boom assembly.

[0011] Additionally, a system for performing controlled leveling of a multi-wing boom assembly using a multi-loop cascade controller arrangement is described herein. The system includes a master controller that issues actuation control commands. The system also includes an actuator controller that causes positioning of actuators of the multi-wing boom assembly to minimize a determined current position error of components of the multi-wing boom assembly relative to a determined contour of a field that the multi-wing boom assembly is currently traversing. The master controller and the actuator controller are configured to perform a method comprising: executing, by the master controller, a master controller loop according to a master controller repetitive loop cycle to generate positioning commands to be executed by the actuator controller, wherein the actuation commands are issued for implementation by the actuator controller during the master controller repetitive loop cycle. The implemented method also includes: executing, by the actuator controller, an actuator controller loop independent of the master controller and according to an actuator controller repetitive loop cycle having a duration independent of the main controller repetitive loop cycle to generate control signals for the actuators of the multi-wing boom assembly. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] While the appended claims particularly set forth the features of the present invention, the invention and its advantages will be best understood from the following detailed description taken in conjunction with the accompanying drawings, wherein:

[0013] Figure 1A is a schematic diagram of an exemplary tractor and an underlying multi-wing spray boom including a boom wing level control arrangement according to an illustrative example;

[0014] Figure 1B is a schematic diagram of an exemplary tractor and subsequent multi-wing spray boom according to an illustrative example, and depicting various controlled / actuated hinge points and corresponding boom / wing motions therein;

[0015] Figure 1C is a schematic diagram of an exemplary tractor and subsequent multi-wing spray boom, wherein the left wing is raised by an actuator positioned at a hinge point between the left wing and the center portion of the spray boom, according to an illustrative example;

[0016] Figure 2 is a schematic diagram summarizing a control arrangement implemented by a multi-controller, multi-sensor, multi-data source / stream spray boom leveling control arrangement according to an illustrative example;

[0017] Figure 3 is a flow chart summarizing a simple / simplified primary control loop according to an illustrative example;

[0018] Figure 4 is a flow chart summarizing details of a main control loop determination of a profile error signal for generating a cylinder actuation control signal to a cylinder controller according to an illustrative example;

[0019] Figure 5 is a flowchart summarizing a simple / simplified boom actuator control loop according to an illustrative example;

[0020] Figure 6 is a flowchart summarizing an example set of operations performed by the control arrangement to obtain measurements from sensor and state observer inputs according to an illustrative example;

[0021] Figure 7 depicts a pendulum model of a two-wing (and single mid-section) spray boom assembly used by the main controller to perform dynamic control calculations according to an illustrative example;

[0022] Figure 8 depicts a trapezoidal model of a two-wing (and single mid-section) spray boom assembly used by the main controller to perform dynamic control calculations according to an illustrative example;

[0023] Figure 9 is a flowchart summarizing operations for obtaining raw current crop height according to an illustrative example;

[0024] Figure 10 is a flowchart summarizing operations for obtaining corrected current crop height according to an illustrative example;

[0025] Figure 11 is a flowchart summarizing operations for updating historical crop height estimates according to an illustrative example; and

[0026] Figure 12 is a chart providing an illustrative example of ultrasonic sensor signals versus time for a multi-sensing arrangement (crop and land) according to an illustrative example. DETAILED DESCRIPTION

[0027] In this disclosure, a robust system and method is described to actively control a multi-winged boom assembly as Figure 1A Horizontal and / or directional of a multi-winged boom assembly is shown by way of example. According to an illustrative example, a multi-winged spray boom assembly 100 is attached to and is being towed behind a tractor 101. While illustrative examples of spray boom physical geometry control arrangements are provided in the context of a towed spray boom assembly, the present disclosure is also applicable to a front-mounted spray boom, a self-propelled sprayer, a slurry tanker, a vineyard spray assembly, a mounted sprayer, etc.

[0028] In the illustrative example, the multi-wing boom assembly 100 includes a left wing 102a, a right wing 102b, and a mid-section 102c. In the illustrative example, only the left wing 102a and the right wing 102b rotate up / down relative to the mid-section 102c that is mounted on a carriage (or other suitable support structure) that supports the multi-wing boom assembly 100. In the illustrative example, the left wing 102a and the right wing 102b are coupled to the mid-section 102c at hinges 104a and 104b, respectively.

[0029] With continued reference to Figure 1A , a plurality of individually controllable actuators are provided to facilitate repositioning and / or reorienting the multi-wing spray boom assembly 100, and individually, each of the left wing 102a, the right wing 102b, and the mid-section 102c. A lift actuator 110 facilitates raising / lowering the spray boom assembly 100 as a whole relative to a mounting carriage 118. By way of illustrative example, the lift 110 actuator raises / lowers the boom assembly 100 along a path defined by vertical / parallel rails of the mounting carriage 118. It is emphasized that the physical orientation and geometry of the mounting carriage 118, including the particular arrangement of the rails, varies according to various implementations. Wing lift actuators 112a and 112b rotate the left wing 102a and the right wing 102b up / down at the hinges 104a and 104b, respectively, that connect the mid-section 102c to each of the left wing 102a and the right wing 102b bases. Wing tip lift actuators 114a and 114b rotate outer sections of the left wing 102a and the right wing 102b up / down at hinge points 104c and 104d (between sections within the wings 102a and 102b), respectively, that are similar to the hinges 104a and 104b. A boom tilt actuator 116 is provided to rotate the spray boom assembly 100 as a whole relative to the mounting carriage 118 (e.g., the sprayer chassis) - e.g., to cause the left side of the spray assembly to rotate above horizontal and the right side to rotate below horizontal (i.e., to raise one side / wing above horizontal while lowering the other side by an equal amount below horizontal).

[0030] The example multi-wing spray boom assembly 100 also includes nozzles 106, e.g., a linear array of nozzles arranged to be downwardly directed, for providing downwardly sprayed material onto the crop. Additionally, in accordance with the illustrative example, profile proximity sensors 108a, 108b, 108c, 108d, 108e, and 108f are arranged on the left wing 102a, the mid-section 102c, and the right wing 102b of the multi-wing spray boom assembly 100 to provide feedback ground / crop proximity (distance) measurement signals as feedback parameters to aid in actively controlling the distance between the nozzles 106 arranged on the left wing 102a, the mid-section 102c, and the right wing 102b and the underlying ground / crop. A variety of sensor types are contemplated for the profile proximity sensors 108a, 108b, 108c, 108d, 108e, and 108f. In addition to known distance measurement systems (laser, ultrasonic, LIDAR, radar, etc.), mounted video cameras can also provide optical proximity sensor information.

[0031] Figure 1A The depicted multi-wing spray boom assembly 100 is exemplary in nature. In view of the disclosure herein, those skilled in the art will appreciate that the boom assembly height / orientation control arrangements described herein are applicable to a wide variety of physical spray boom assemblies in accordance with the various embodiments of the control arrangements described herein.

[0032] Turning to Figure 1B , a schematic diagram of an example tractor and subsequent multi-wing spray boom is provided in which various controlled / actuated hinge points and corresponding boom / fin movements are depicted. Reference lines 122a and 122b correspond to hinge / actuator combinations that in operation cause upward / downward rotation of the entire left wing 102a and right wing 102b. Reference lines 124a and 124b correspond to hinge / actuator combinations that in operation cause upward / downward rotation of the left and right winglets. Reference line 126 corresponds to the position at which the entire boom assembly 100 is rotated (on line 126).

[0033] Turning simply to Figure 1C , the left wing 102a is shown in an upwardly rotated position. It is further noted that with respect to Figure 1C , the sudden raising of the left wing 102a can / will cause a pendulum (up / down) motion of the right wing 102b.

[0034] Turning to Figure 2 , a schematic diagram summarizes the multi-wing boom level control arrangement in accordance with the illustrative example. In particular, it is noted that Figure 2The control arrangements depicted in the middle are not intended to constrain the physical component arrangement of the sensor and processing elements in any way. For example, the master control and actuator control can be configured and executed on a single physical processor. Conversely, the control operations of the master control and actuator control described herein can be executed on distributed intelligent sensor / actuator components (e.g., distance sensors and / or boom actuators operating in a coordinated manner based on shared dates).

[0035] According to the illustrative example, closed loop control is performed independently by the master controller 202 and the actuator controller 204. The actuator controller 204 performs a control loop driven by actuator commands (e.g., actuator set points) issued by the master controller 202 and actuator position sensor data indicating the current state of the boom vane position actuators. The actuator controller 204 derives an actuator position error based on the actuator command (set point) and the feedback actual position data. The actuator controller 204 thereafter issues actuator control signals based on the derived actuator position error.

[0036] The master controller 202, based on a variety of input data types (described below), executes an automatic multi-wing spray boom positioning control scheme (enhanced by human operator override input) that derives and issues actuator repositioning commands to be submitted to the actuator controller 204 to correct for positioning errors (i.e., the difference between the current distance between the spray nozzles 106 and the underlying field profile and the target distance). This control can be very complex, including consideration of mechanical / physical interactions between actuation of the left and right wings 102a and 102b (as well as resulting tilt of the overall boom assembly 100 resulting from asymmetric raising / lowering of the individual wings 102a and 102b of the spray boom assembly 100).

[0037] As Figure 2 shown, the master controller 202 receives input from a variety of sources. The target distance between the spray nozzles mounted on the wings 102a and 102b and the current field surface is provided by profile reference data 214. The profile reference data 214 provides the generally desired target distance between the spray nozzles 106 on the wings 102a and 102b of the multi-wing spray boom assembly 100 relative to the field surface (e.g., the ground, the top of the crop, etc.) as a function of the current position of the spray boom assembly 100 within the field. The current (measured) distance to the field / crop profile beneath the wings 102a, 102b is provided to the master controller 202 and the actuator controller 204 by profile sensor data 208. The profile sensor data 208 is typically a set of distances measured by the sensors 108a, 108b, 108c, and 108f in the illustrative example. Figures 1A-1C

[0038] ​Additionally, the current state of configuration (shape / orientation) of the multi-wing spray boom assembly 100 is provided to the master controller 202 by boom sensor data 210.

[0039] Finally, the master controller 202 is configured to additionally receive input from a state observer 212. The state observer 212 receives input from the boom sensors 210 (indicative of the current physical configuration of the wings 102a and 102b). The output of the state observer to the master controller 202 includes, for example, the absolute boom rotation angle (compared to gravity) and velocity. Additionally, the state observer 212 can also provide the master controller 202 with boom external measurements, such as the current tilt of the trailer to which the boom is mounted (relative to gravity), etc. Such tilt information is provided, for example, by a gyroscope mounted to the trailer. In this case, a three-dimensional "disturbance" vector is provided from the output of the gyroscope.

[0040] Based on the combined input from the above-identified exemplary sources, the master controller 202 derives the positioning error of each wing component of the multi-wing spray boom assembly 100 and issues actuator control commands (which correspond to the required actuator adjustments in view of the positioning error) to the actuator controller 204. The actuator controller 204 thereafter executes an actuator control loop with respect to each actuator (cylinder) in accordance with the actuator control commands received from the master controller 202 to effectuate the re-orientation (i.e., rotation up / down at the hinge point) of the left wing 102a and right wing 102b relative to the mid-section 102c of the multi-wing spray boom assembly 100. The operation of the actuator controller 204 to provide control output to the actuators 206 is additionally based on feedback provided, for example, by the boom sensor data 210 (indicative of the current state / position of the boom actuators) and the profile sensor data 208 (providing short-latency / high-speed "safety" override actuation to prevent damage to boom system 100 components resulting from impact between the boom and an encountered obstacle).

[0041] According to various illustrative examples, the master controller 202 incorporates any one or more of various closed-loop control arrangements. Such control arrangements include PID (proportional integral derivative) control in multiple-input / multiple-output format, multiple single-input / single-output format, etc.; state space feedback; H ∞ control; and / or nonlinear control (e.g., sliding mode control). In an illustrative example incorporating a two-stage cascaded control arrangement, the master controller 202 executes an upstream control loop on the relatively slow-responding system (i.e., the boom assembly 100) to render actuator (repositioning) device setpoints. The actuator controller 204 executes a downstream control loop on the relatively fast-responding actuator components to execute the actuator repositioning commands (specifying actuator setpoints / increments) specified by the master controller 202.

[0042] In illustrative examples, the main controller 202 and the actuator controller 204 independently implement any wide range of control output dampening. Depending on the conditions under which the boom assembly 100 operates (e.g., rapidly changing terrain), the physical characteristics of the boom assembly (wing length, mass, flexibility, spring, damper, etc.), and the actuator device performance parameters (e.g., power, capacity, etc.), certain control output dampening characteristics can be adjusted / configured / set (prior to and during field operation). In illustrative examples, an under-damped control scheme is used to ensure sufficient responsiveness of the system to quickly eliminate positioning errors (of the wings) caused by relatively sudden changes (e.g., sudden changes in the field profile under one or both of the left wing 102a and the right wing 102b) caused by disturbances.

[0043] The degree of dampening exhibited in the control output of the main controller 202 and the actuator controller 204 is adjustable according to various illustrative embodiments and configurable control modes. Thus, according to illustrative examples, for systems and / or operating conditions in which disturbances are generally small and / or the responsiveness of the boom wing positioning system is relatively slow, the dampening is increased and over-damped control is required to be output by the main controller 202 to avoid over-correction / compensation, which would result in oscillations as the spray boom system 100 seeks to designate target boom wing positions. However, in illustrative examples, the relatively slow responsiveness of the boom system 100 as a whole does not impede the actuator controller 204, including under-damped control output, to quickly / precisely control the actuators of the boom wings to transition the boom wings from a current position to a target position commanded by the main controller 202 or required as a result of encountering an obstacle - requiring an emergency / safety response to be autonomously made by the actuator controller 204.

[0044] Importantly, the cascaded control arrangement performed by the main controller 202 and the actuator controller 204 facilitates configuring the degree of dampening applied in the actuator control scheme implemented by the actuator controller 204 (which is according to the performance characteristics of the controlled actuators) independently of the degree of dampening applied in the main controller 202 control configuration (which can be established with respect to the operating and performance conditions of the spray boom system 100 as a whole). Thus, two different degrees of dampening can be specified in the cascaded control arrangement performed by the main controller 202 (which performs the boom wing positioning control scheme and provides repositioning commands based on the error between current / target wing positions) and the actuator controller 204 (which performs the actuator control arrangement to cause repositioning of the actuators of the left wing 102a and the right wing 102b according to the repositioning commands issued by the main controller 102).

[0045] By Figure 2The above two-part / cascaded spray boom control arrangement is advantageously implemented to account for dynamic wing control (discussed in detail below) that takes into account the cross-interaction between the left wing 102a and the right wing 102b of the multi-wing spray boom assembly 100. For example: lifting the left wing 102a upward causes a tilt of the boom assembly 100 that results in a tilt of the right wing 102b (see Figure 1C ). This tilt and the resulting effect on the opposing wing is a result of the relatively high mass and moment of inertia of the initially actuated / lifted wing. The control arrangement described herein (including the two different control loops implemented separately by the master controller 204 and the actuator controller 202) is able to accommodate and compensate for the cross-interaction between the boom wings, thereby facilitating a fast, but smooth and accurate transition between the current and target boom position, without causing / experiencing system oscillations or instability.

[0046] The master controller 202 includes various configuration / adaptation features. By way of example, the master controller 202 includes well-known sliding mode control. Sliding mode control is a non-linear control that, in a particular illustrative example, is included into the master controller 202 for use in conjunction with a physical model of the spray boom system (e.g., the boom assembly 100) to be controlled in a predetermined manner. An advantage of augmenting the sliding mode control with the model-based control of the spray boom system 100 described herein is its robustness (its ability to accommodate various changes in the terrain beneath a potentially very large spray boom system having a very wide wingspan, and to avoid oscillations / shaking of one or more boom wings during operation). Moreover, in the illustrative implementation, the operation of the master controller 202 is augmented by using state feedback provided by the state observer 212 in the calculations made by the master controller 202 as part of the sliding mode control. However, acceptable operation of the master controller 202 is achievable without using state feedback and only by using state feedback provided by the state observer 212.

[0047] According to the illustrative example provided in Figure 2 The actuator controller 204 generally controls the operation of the controlled system elements (i.e., the actuators), and thus can operate the relatively fast downstream control loop (of the cascaded control loop arrangement) to precisely and accurately control each of the above-described actuators according to the actuator repositioning commands received from the master controller 202, in accordance with the illustrative example provided in Figures 1A-1C) of the actuators (e.g., hydraulically actuated piston / cylinder systems). For example, the actuator controller 204 is capable of executing a downstream control loop that results in precise repositioning of the actuators (e.g., hydraulic cylinders / pistons) of the multi-wing spray boom assembly 100 with a relatively fast response compared to the relatively slow response of the multi-wing spray boom assembly 100 controlled by the master controller 202 executing an upstream control loop to render the actuator repositioning commands provided to the actuator controller 204, which executes the repositioning commands on the various actuators of the boom assembly 100 according to the cascade control arrangement.

[0048] Thus, according to the cascade control arrangement, actuator device control is performed as downstream control (of the illustrative cascade control arrangement) by the actuator controller 204 in accordance with actuator repositioning (setpoint, increment, etc.) commands / instructions determined by the master controller 202 and communicated downstream. Thus, the actuator controller 204 operates a downstream control loop (of the cascaded upstream / downstream cascaded control loop pair) on a controlled component (i.e., an actuator) that responds significantly faster to forces acting on the controlled component than the boom assembly 100, which has a significantly greater controlled mass / inertia. Thus, the cascade control loop arrangement facilitates the decoupling of: (1) determining an actuator position setpoint (or a difference / increment between a current and a target setpoint), and (2) causing the actuator to transition to the new setpoint (thereby causing the position of the boom flaps (e.g., boom flaps 102a and 102b) to transition to the corresponding desired position setpoint). In the cascade control arrangement provided herein, this decoupling of the two control loops avoids the introduction of nonlinearities—for example, caused by the actuation of multiple hydraulic valves—that would otherwise occur if direct actuator control were performed by the master controller 204 control loop issuing actuator set points.

[0049] Furthermore, separating the overall control of the spray boom assembly positioning into a cascade of control loops, including an upstream control loop (actuator set point determination) and a downstream control loop (controlling the physical actuator to achieve a specified actuator set point over a period of time based on the physical limitations of the specific actuator system and boom geometry and configuration) reduces the complexity of the upstream loop executed by the main controller 202 while significantly increasing the customizability, configurability, robustness, and performance of the complete controller arrangement. For example, the actuator controller 204 may implement a simple / known proportional gain algorithm enhanced by incorporating a Smith predictor and valve profile compensation. Reference Figures 1A-1C, the actuator controller 204 issues control signals to the actuators 112a, 112b, 114a, and 114b of the left and right wings 102a, 102b, and the lift actuator 110 and the boom tilt actuator 116 (for the boom assembly 100 as a whole) in accordance with downstream control actions in the overall cascade control arrangement to affect repositioning of the actuators in accordance with received actuator repositioning commands to reduce / eliminate boom flap positioning errors determined by the master controller 202.

[0050] exist Figure 2 In the control architecture diagram provided in FIG, the spray boom vane actuator 206 component corresponds to a plurality of actuators associated with the spray boom assembly (e.g., Figures 1A-1C 112a, 112b, 110, 116, 114a and 114b, and more specifically, a hydraulically actuated piston system), such as Figures 1A-1C For example, the output from the actuator controller 204 causes the opening and closing of the hydraulic cylinder valve of the piston system to Figures 1A-1C The illustrative arrangement depicted affects the lifting / lowering of a single spray boom flap / wingtip as well as lifting and rotating the entire spray boom assembly 100 .

[0051] Although Figures 1A-1C The illustrative example in provides a spray boom assembly 100 that includes a plurality of actuators in the form of hydraulic pistons operating on articulated joints, but the control arrangement can be implemented in association with a variety of physically reconfigurable spray boom assemblies having any of a variety of suspensions (including: pendulum, trapezoidal member, cable-roller, etc.).

[0052] Furthermore, while hydraulic cylinder-based actuator systems are specifically mentioned above, other actuator types are also contemplated, such as: pneumatic, electrical (eg, stepper motors and linear actuators), mechanical, etc.

[0053] Similarly, the type of signal received by the boom component actuator can be any of a variety of signals, including: analog (level), digital, pulse width (modulated), switch (eg, on / off signal), etc.

[0054] In the illustrative example, boom sensor data 210 provides actual positions of individual boom wings affected by actuation signals / commands issued by actuator controller 204 to spray boom wing actuator 206. By way of example, boom sensor data 210 provides one or more of the following examples of boom wing position data types: boom tilt, boom tilt speed (i.e. rate of rotation of the overall boom assembly about the central pivot point); left wing rotation; right wing rotation; left wing tip angle (relative to the hinge point between the wing and the central portion of the spray boom assembly 100); and right wing tip angle.

[0055] Boom sensor data 210 corresponds to any sensor (signal) data type that provides output values that are not relevant to detecting the field (crop / land). Boom sensor data 210 provides information related to the current physical configuration of each wing of the multi-wing boom assembly 100 (e.g. boom shape and behavior, and possible perturbations). Depending on the sensor type, the measurements can be used directly by the main controller 202. Examples of boom sensor data 210 types include: actuator position sensor (length measurement) for tilt cylinder, actuator position sensor (length measurement) for left wing cylinder, actuator position sensor (length measurement) for right wing cylinder, and gyroscope on central carriage. Additionally and / or alternatively, boom sensor data 210 can include the following types: distance sensor, angle sensor (e.g. inclinometer), speed sensor, rotational speed sensor (e.g. gyroscope), acceleration sensor (e.g. accelerometer), rotational acceleration sensor, barometric pressure sensor - (provides height measurement), and magnetometer.

[0056] Contour sensor data 208 provides data indicative of the current distance between sensors mounted at known locations on the boom assembly (e.g. at known lateral positions along one of the boom wings). By way of example, contour sensor data 208 includes data provided by ultrasonic sensors mounted, for example, at known distances along one of the left wing 102a and right wing 102b. Any of a variety of (and quantity of) sensor types are contemplated in accordance with the various embodiments of the boom control arrangement described herein. Other sensor types that provide contour sensor data 208 include: radar, lidar, video camera, barometric pressure, etc. Other multi-point measurements can be obtained, for example, by: 2d radar, 2d LIDAR, 3d LIDAR, and video camera. The spray boom assembly 100 can be used to sense and adjust the boom geometry in any of a variety of field types including, for example, bare land, small crops, tall crops, sparse / dense vegetation, fields with highly non-uniform shape (e.g. potato fields), vineyards.

[0057] Additionally, the control implementation of the main controller 202 and / or actuator controller 204 is enhanced by providing any of a variety of data related to the controlled components, including: hydraulic circuit component measurements (e.g., pressure, flow, valve status); non-boom components measurements (e.g., sprayer chassis, intermediate frame, connected spring and / or damping elements / structures, tractor equipment, rotational speed sensors (e.g., gyroscope); and tractor data (e.g., oil flow / pressure, enabled hydraulic functions, geospatial position, vehicle speed. Such measurements are helpful in determining disturbance variables acting on the system, which can be used to improve control performance.

[0058] The state observer 212 (based on the type of sensor used) provides additional processing of the sensor output signals to give an accurate representation of the current state of the multi-winged sprayer boom assembly 100 for use by the controller 202 in executing the primary (slow) control loop of the dual fast / slow control loop arrangement executed by the main controller 202 and actuator controller 204. Thus, in addition to pre-processing the raw data stream using filtering algorithms, an observer can be used to give a more robust determination of state, for example, the state observer 212 is a Luenberger observer. However, the state observer 212 can implement a Kalman filter or any of a variety of more advanced processing of the raw sensor data to render a summary / representation of the current state of the multi-winged sprayer boom assembly 100 (or individual components thereof).

[0059] Importantly, the control arrangements described herein are intended to be interpreted broadly and without significant limitation with respect to implementing the present disclosure in a variety of sprayer boom assembly application environments. Moreover, while inclusion of lift / lower control can indeed be included in the described control schemes, the disclosed control arrangements can be implemented without such control. Rather, the control is intended to primarily focus on raising / lowering particular portions of the sprayer boom assembly 100 in accordance with sensed changes in operating conditions of the sprayer boom assembly 100.

[0060] Having described control apparatus for performing boom wing leveling, reference is made hereinafter to Figures 3-6 An exemplary set of operations performed by the main controller 202 and actuator controller 204 are described.

[0061] Turning to Figure 3 , the flowchart summarizes the operations according to the foregoing reference Figure 2The described control arrangement, the master controller 202 performs operations for an exemplary method of acquiring and processing a set of raw data points. According to the illustrative example, during 300, the master controller 202 determines the relative positions of the left wing 102a and the right wing 102b of the multi-wing spray boom assembly 100 relative to (e.g., from) the field contour (e.g., the ground of the field). However, the field contour can be established by the crop canopy or any other reference contour of interest. In the illustrative example, the boom positions relative to the field contour are determined based on analysis of profile (e.g., land to sensor distance, crop canopy to sensor distance, etc.) input data provided by the profile sensor 208.

[0062] Thereafter, during 310, the master controller 202 determines the status of the midsection 102c, the rotational speed of the midsection 102c, the status of the left wing 102a, and the status of the right wing 102b of the multi-wing spray boom assembly 100 based on analysis of input boom midsection winglet position data provided by the boom sensor 210.

[0063] During 320, the master controller 202 determines target positions for the left wing 102a, the midsection 102c, and the right wing 102b - more specifically, the spray nozzles 206 mounted thereon - of the multi-wing spray boom assembly 100 relative to the sensed profile below the left wing 102a and the right wing 102b. For example, the target boom positions are provided by the profile reference data 214 based on the current sensed position of the boom assembly 100 in the (crop) field. Additionally, according to the illustrative example, the selection of the many potentially available target boom positions provided by the profile reference data 214 is based on the status of the left wing 102a and the right wing 102b established during 310.

[0064] Based on the combination of inputs obtained during operations 300, 310, and 320, at 330 the host controller determines boom wing and midsection positioning errors. According to the current illustrative example, the boom position error is the difference between: (1) the current positions of the left and right wings 102a and 102b of the multi-winged spray boom assembly 100 established during 300; and (2) the target boom positions established during 320 (as informed by the boom state obtained during 310). Thus, at 330 the host controller 202 determines repositioning of the actuators to move the boom wings 102a and 102b to new positions such that the current boom positions will converge to the target boom positions relative to the profile of interest (e.g., the ground or crop top surface). The profile error can be presented in any of a variety of forms. By way of example, the boom wing positioning error relates to the multiple degrees of rotation of the wing 102a and 102b and / or the boom assembly 100 itself (at a single point of rotation) that reconfigure the geometry and / or orientation of the boom assembly 100. By way of further example, the boom positioning error can also specify an amount to raise / lower the boom assembly 100 using the lift actuator 110.

[0065] Thereafter, at 340 the host controller 202 applies the known relationship between actuation (e.g., extension / retraction) of the boom assembly wing lifters / actuators (e.g., hydraulic actuation pistons) and resulting absolute positioning of the left and right wings 102a and 102b of the multi-winged spray boom assembly 100 to minimize / eliminate the boom wing positioning errors determined during 330.

[0066] During 350, in accordance with the two-loop cascaded control arrangement described above, the host controller 202 sends appropriate actuator repositioning (e.g., extension / retraction) commands to the actuator controller 204 to cause the actuators of the left wing 102a, right wing 102b, and boom assembly 100 to modify the positions of the wing / boom components to perform adjustments in accordance with the repositioning commands generated by the calculations performed by the host controller 202 during 340.

[0067] This set of operations is generally performed as a control loop (although the order of operations is not necessarily adhered to, and some operations can be repeated several times for each executed control loop / iteration completed by the host controller 202).

[0068] Turning to Figure 4 The flowchart summarizes the details of the host controller 202 upstream control loop operations performed during 340 to apply the dynamic model of the boom assembly 100 described in detail below by way of illustrative example (see Figure 7) generate actuator repositioning. During 400, the master controller 202 determines an initial set of uncompensated actuator repositioning values for changing the state of the vane actuators (e.g., actuators 112a, 112b, 114a, and 114b) to reduce / eliminate the determined boom position error established during 330.

[0069] Thereafter, during 410, the master controller 202 determines an actuator repositioning correction for each actuator based on the dynamic model of the boom assembly 100 and including the current spray boom orientation (and its vanes).

[0070] Thereafter, during 420, the master controller 202 sums the initial repositioning and correction repositioning values for each actuator to give a full set of dynamically model-compensated actuator repositioning values for each of the actuators 112a, 112b, 114a, and 114b in the set of actuators (e.g., the set of actuators 112a, 112b, 114a, and 114b) on the boom assembly 100. Figures 1A-1C Figure 4

[0071] Turning to Figure 5 , the flowchart summarizes a simple / simplified boom actuator control loop executed by the actuator controller 204 (in a downstream cascading controller role relative to the master controller 202 providing actuator repositioning commands in an upstream controller mode of a cascading control loop pair) according to the illustrative example. During 500, the actuator controller 204 obtains the actuator current state / position by analyzing the boom sensor data 210 for each of the actuators 112a, 112b, 110, 116, 114a, and 114b.

[0072] During 510, the controller 204 obtains / retrieves target state / position data for each of the actuators 112a, 112b, 110, 116, 114a, and 114b. This target state / position data corresponds to the target actuator positioning instructions provided within the repositioning commands / signals provided by the master controller 202 to the actuator controller 204 during 350. During 520, the actuator controller 204 determines actuator control commands / signals based at least on the difference between the current and target state / position determined for each of the actuators 112a, 112b, 110, 116, 114a, and 114b. Optionally, the actuator control commands / signals are further established based on a history of previously issued actuator commands (the control loop executed by the actuator controller 204 being a PID control loop in the illustrative example). Figure 5 ​​the degree of damping / filtering / smoothing provided during each iteration of the depicted) to determine the actuator command issued to a particular one of the actuators 112a, 112b, 110, 116, 114a, and 114b. During 530, the actuator controller 204 issues an actuator command / signal to each of the actuators 112a, 112b, 114a, and 114b.

[0073] Turning to Figure 6 The flowchart summarizes an exemplary set of operations performed by the control arrangement to obtain measurements from sensor and state observer inputs, according to an illustrative example. During 600, the main controller 202 obtains boom-on sensor measurements (corresponding to the boom sensor data 210) indicative of the current physical configuration of various physical subcomponents of the boom assembly 100 (e.g., the wings 102a and 102b).

[0074] During 610, the main controller 610 obtains off-boom sensor measurements. For example, during 610, the main controller 202 obtains measurements of various parameters related to the geospatial (three-dimensional) shape of the field / crop (e.g., variation in hill slope along the length of the boom assembly (orthogonal to the direction of travel)).

[0075] During 620, the main controller 202 uses the measurements obtained in 600 to compute known boom states and disturbances. Depending on the sensor arrangement, this step can provide anything from 0 states to all states of the system.

[0076] Thereafter, during 630, the main controller 202 obtains missing states. By way of example, the main controller 202 obtains missing state information from the state observer 212, which computes such information using a Luenberger observer to estimate missing states.

[0077] During 640, the main controller combines the state and sensor measurement information obtained during 620 and 630 to provide a complete state of the machine as well as known disturbances (e.g., sensed rotation caused by tilting of the boom assembly 100). This information is used in decision making by the main controller 202 during, for example, operations 320 and 340.

[0078] Turning to Figure 7 A pendulum-based model of a two-wing (and single midsection) spray boom assembly is depicted that is used by the main controller 202 to perform dynamic control calculations, according to an illustrative example. Figure 7 A pendulum-based model of a two-wing (and single midsection) spray boom assembly is depicted that is used by the main controller 202 to perform dynamic control calculations, according to an illustrative example. Figure 1AA simplified schematic overview of the spray boom assembly 100 depicted in FIG. The pendulum-based model of the depicted boom assembly 100 includes four components, identified in the figure using four different subscripts "i" taken from a group consisting of four different labels (l, r, 1 and 2) corresponding to each of the four different components of the model. According to the pendulum model of the spray boom assembly 100, the movement of the boom 100 relative to the trailer frame of the tractor (carrying the boom assembly 100) or the sprayer chassis can be expressed by four angles θ i To describe. At those angles θ i On the surface, there exists a corresponding moment τ i The component itself has a mass m i and inertia I i . To obtain a dynamic model (i.e. when the part is moving), the angles and their time derivatives (velocity and acceleration) are considered. Where applicable, these angles can be related to the actuator displacement (in this case, through trigonometry). The torque consists of several components: actuator actuation force, friction, damping, springs, gravity, etc. Constraints can also be considered, such as the range of motion of the actuator, motion stops, etc. Using this information, a system of equations can be established that describes the dynamic behavior of the boom in this simplified form (in this case, using the Euler-Lagrange equations).

[0079] Steering Figure 8 , depicts an alternative, trapezoidal-based model of a two-wing (and single mid-section) spray boom assembly. The trapezoidal-based model is similar to Figure 7 The pendulum-based model differs in that the mid-section 102c is modeled as a trapezoid suspended from a fixed rotation point – rather than a horizontal rod and a vertical rod connected to the horizontal rod at one end.

[0080] Dynamic model-based compensation repositioning commands of the master controller 102

[0081] According to the illustrative example, the main controller 202 includes a dynamic model for controlling the actuation of the flaps 102a and 102b and the boom assembly 100 as a whole. For example, the main controller 202 uses a model based on the Euler-Lagrange equations (or virtual work) and knowledge of the components (mass, dimensions, etc.) of the boom assembly 100. For example, the equations take into account: component mass, component inertia, component dimensions (2D), forces acting on / accelerating the components, gravity, spring constants, buckling / bending, damping, friction, etc.

[0082] The mechanical model applied to and contained within the mathematical operation and its tunable parameters are established and refined from experimentation on various models of the boom assembly 100. One such example of a tunable parameter value is friction. By way of example, the mechanical model includes a state space equation with eight (8) state variables: trailer / attachment point to subframe angle, trailer / attachment point to subframe angular velocity, subframe to center frame angle, subframe to center angular velocity, left boom vane angle, left boom vane angular velocity (pivoting up / down under force of actuator), right boom vane angle, and right boom vane angular velocity.

[0083] The model also includes three (3) manipulated inputs: tilt cylinder force, left boom vane cylinder force, and right boom vane cylinder force.

[0084] Other inputs to the dynamic model implemented by the host controller 202, by way of example, include: disturbance inputs and absolute trailer / mount assembly rotation.

[0085] The need for precise knowledge of the cylinder forces can require additional measurements and precise knowledge of the hydraulic circuit. On the other hand, by using a downstream control-loop executed by the actuator controller 204 arranged in cascade with the host controller 202, the position of the actuators 112a, 112b, 110, 116, 114a, and 114b (e.g., hydraulically actuated cylinder / piston systems) can be precisely controlled, and the inputs can be converted from force inputs to position inputs (provided by the host controller 202 to the actuator controller 204). By way of example, the state space model can be appropriately converted, resulting in a dynamic model including:

[0086] A. two (2) state variables of interest: (1) trailer / attachment point to subframe angle, and (2) trailer / attachment point to subframe angular velocity. And

[0087] B. four (4) manipulated input variables: (1) subframe to center frame angle, (2) left boom vane angle, (3) right boom vane angle, and (4) boom height relative to mount point (at midsection 102c).

[0088] These two state variables can be transformed to represent different angles and angular velocities, such as total boom tilt and total boom angular velocity.

[0089] The state space equation can be extended by a representative model of the downstream cascade control loop, which is greatly simplified compared to the full modeling of the hydraulic behavior in the current illustrative example.

[0090] The resulting state space system includes:

[0091] A. Six (6) state variables of interest: (1) total boom angle, (2) total boom angular velocity, (3) tilt cylinder position, (4) left boom vane cylinder position, (5) right boom vane cylinder position, and (6) height cylinder position, and

[0092] B. Four (4) manipulated input variables (which will be sent to the downstream loops of the cascade): (1) tilt cylinder position reference, (2) left boom vane cylinder position reference, (3) right boom vane cylinder position reference, and (4) height cylinder position reference.

[0093] The main controller 202 operates

[0094] To facilitate the main controller 202 implementing dynamic model-reference sliding mode control, a model representation of the multi-wing spray boom assembly 100 system (see the model discussed above Figure 7 and Figure 8 ) is provided, for example, in state space format. In addition, an ideal / desired model behavior, sliding mode parameters, and an estimate of the noise and disturbances to the system are also specified.

[0095] The model itself is discussed above. With respect to the ideal / desired model behavior, the system can be controlled through state space feedback, which is a linearized control technique that is susceptible to un-modeled dynamics or non-linearities, among other things. The system of current interest (the multi-wing spray boom assembly 100) is subject to a variety of disturbances and non-linearities. Therefore, a state space feedback controller is used to determine the ideal / desired model behavior, and the sliding mode control will ensure that the actual machine approaches the ideal / desired model behavior.

[0096] With respect to state space feedback with direct eigens structure assignment, the state space feedback matrix F is used in the input equation u = Fx + Gr, where x is the state of the machine. F is determined through direct eigens structure assignment. This involves a choice of either the eigenvalues of the closed loop system, or the eigenvalues and eigenvectors of the closed loop system.

[0097] The eigenvalues are used to provide the state space feedback. However, the addition of the eigenvectors as design parameters creates new possibilities. This adaptation of the operation of the control loop upstream of the main controller 102 should address the dynamic behavior of the boom. One such property of the system is the cross-interaction. Through the eigens structure assignment, the closed loop behavior of the boom assembly 100 horizontal control system can be tuned to minimize these effects.

[0098] The other term of the input, Gr, is determined by the reference input r, which will be the profile (or profile error) that the boom should follow. This is shaped through the matrix G to align the applied reference with the corresponding output value.

[0099] Sliding mode parameters

[0100] To match the spray boom to the ideal model, sliding mode control is used. Several design possibilities are possible: robust eigenstructure assignment, quadratic minimization (used here) and direct eigenstructure assignment (a method similar to the state space feedback case) - just to name a few of the many possibilities.

[0101] A typical approach in sliding mode control is to separate the generated input signal coming out of the controller into a linear and a nonlinear component. The nonlinear component is scaled to overcome known and unknown uncertainties in the system. This sliding mode control approach is sufficient to make the state space feedback component and the reference feedforward Gr not necessary. However, it is wise to separate the control effort in this way, as they consist of known signals.

[0102] Turning to Figure 9 , the flowchart summarizes, by way of example, the operations performed by the main controller 202 for obtaining the raw current crop height based on received sensor signals (e.g., ultrasonic) indicative of the currently sensed distance between sensors (e.g., sensor 108a) mounted on the boom assembly 100. By way of example, the provided sensor signals are presented by a sensor that provides a single distance measurement for each reading period - more specifically, the sensor is an ultrasonic sensor that provides a distance corresponding to the first received echo of a transmitted pulse. However, in an alternative arrangement illustratively depicted in Figure 12 , the provided sensor signals are presented by a sensor that provides multiple (e.g., two) distance measurements based on sensed peaks of ultrasonic echo signals for each transmitted ultrasonic burst corresponding to the "crop top" and "ground" distances. With continued reference to Figure 12 , the echo signals generated by a single burst (excitation) signal can be fed to a dual Kalman filter to present: (1) a distance to the crop top - corresponding to the first detected peak (satisfying the signal amplitude, duration, prominence, etc. of the crop top profile); and (2) a distance to the ground - corresponding to the second detected peak (satisfying the signal amplitude, duration, prominence, etc. of the ground profile). The above-described multi-distance sensing arrangement is merely one particular example of an alternative arrangement for obtaining / generating input distances for performing boom level control in accordance with the illustrative examples provided herein.

[0103] In the illustrative examples, the distance sensor is an ultrasonic sensor that provides a single distance measurement corresponding to the shortest sensed distance between the sensor and an object within the sensing volume. By way of example, Figure 9The operations summarized in the middle execute approximately 100 times per second. However, other sampling rates are contemplated. In addition, a queue (FIFO buffer) is maintained of the last 64 raw measurements (stored during operation 950 discussed below). The history of distance sensor raw measurements, in conjunction with processing of such signals to provide historical crop height estimates (described herein), aids in making informed decisions as to whether a current distance sensor measurement corresponds to the crop top or to the ground for purposes of determining a raw current crop height. In addition, according to illustrative examples, the host controller 202 can also maintain a history of past values of measured distance to crop, measured distance to ground, average / fi ltered distance, etc. There is no intended limitation on the type, frequency, duration, etc. of saved historical data (maintained in any of a variety of data structure types including queues, stacks, etc.).

[0104] During 900, the host controller 202 obtains a current distance measurement from the ultrasonic sensor. The ultrasonic sensor measurement indicates the distance between the sensor and a sensed physical feature below the boom assembly 100. The distance is presented, for example, by presenting the distance to crop or ground - or in the case of a dual Kalman filter, the distance to both crop and ground - by providing the ultrasonic signal to a Kalman filter.

[0105] During 910, the host controller 202 stores the current ultrasonic sensor distance measurement at the head of a current distance measurement queue (e.g., a 64-element FIFO buffer). The ultrasonic distance measurements stored in the current distance measurement queue can represent the crop top, the ground, anything in between, or even be invalid (e.g., no echo received). Figure 9 The remaining operations summarized in the middle involve computing a raw estimate of the current crop height based on applying decision logic to the history of (64) previous distance measurements stored in the current distance measurement queue.

[0106] During 920, the host controller 202 determines a raw current distance to crop based on the contents of the current distance measurement queue. By way of example, the raw current distance to crop is selected to be the shortest distance measurement of the 64 measurements stored in the current distance measurement queue. Similarly, during 930, a raw current distance to ground is selected to be the longest distance measurement of the 64 measurements stored in the current distance measurement queue. The above examples for determining the current distance are illustrative in nature and are not intended to limit the disclosure in its broadest aspects.

[0107] Thereafter, during 940, the host controller 202 computes a raw (uncorrected) current crop height by subtracting the current distance to crop from the current distance to land, based on the raw current distance to crop and raw current distance to land values established during 920 and 930. During 950, the raw current crop height is stored for further processing (during the operations of Figure 10

[0108] Turning to Figure 10 , an exemplary set of operations for presenting a current distance to crop measurement value based on a comparison of the raw current crop height stored during 950 with a historical crop height maintained by the host controller (see, for example, the Figure 11 discussed below) is summarized. During 1000, the host controller 202 obtains the raw current crop height (stored during 950) and a (long-term) historical crop height estimate (see, for example, the Figure 11 discussed below). Thereafter, during 1010, the host controller 202 compares the values of the raw current crop height and the historical crop height estimate to make a decision as to whether to retain or reject the raw current crop height for use in providing a current distance to crop value for the particular one of the distance sensors driving the distance to crop control performed by the host controller 202 with respect to the boom assembly 100 and its individual wings. Thus, Figure 9 , Figure 10 and Figure 11 the flowchart represents operations performed with respect to each of the ultrasonic sensors - each providing a separate distance measurement value for use by the host controller 202 in the overall boom leveling control arrangement. For example, the current filtered distance to crop measurement value for each physical sensor location on the boom assembly 100 is applied by the host controller 202 to boom geometry configuration determiner logic to present a target boom geometry. For example, such logic can specify a particular range of target distance to crop values for each distance measurement location on the boom assembly 100. The current distance measurement values for each sensor location on the boom can also be compared to identify anomalous distance measurement values indicative of potential abnormal conditions (e.g., sensor failure, terrain changes, field edge, etc.).

[0109] By way of a particular example, during 1010, the host controller 202 determines whether the raw current crop height exceeds a minimum distance computed as a fraction / percentage of the historical crop height estimate. If the raw current crop height exceeds the minimum, a current distance to crop is established based on the raw current distance to crop established during 920. Otherwise, the historical crop height estimate is used to establish the current distance to crop. More specifically, the current distance to crop is computed using the raw current distance to land and the historical crop height estimate. During 1020, the host controller stores the current distance to crop established during 1010. ​

[0110] Turning to Figure 11 , a summary of exemplary operations for maintaining a historical crop height estimate (used during operation of the Figure 10 is provided. In particular, during 1100, the host controller 202 obtains an original current crop height and a historical crop height estimate. During 1110, the host controller 202 presents an estimate of the current crop height based on currently available sensor readings and the historical crop height estimate. During 1110, the host controller actually determines whether the current set of sensor measurements held by the sensor measurement queue (FIFO buffer) corresponds to measurements taken of the crop. Based on that determination (for a few sensor readings, the sensor is measuring land distance rather than crop), the current distance measurements can be replaced by a current crop height estimate derived from the (filtered) historical crop height estimate. Using this scheme, the measurements presented to the leveling algorithm are a more intuitive representation for the host controller 202 to determine the appropriate target position for boom leveling control.

[0111] During 1110, the host controller 202 compares the value of the original current crop height and the current historical crop height estimate to make a decision about whether to use the original current crop height to update the historical crop height estimate (or use another value in place of it). By way of a particular example, during 1110, the host controller 202 determines whether the original current crop height exceeds a minimum distance calculated as a fraction / percentage of the historical crop height estimate. If the original current crop height exceeds the minimum, the current crop height is established based on the original current crop height established during 940 (for the purpose of updating the historical crop height estimate). Otherwise, the current crop height (for the purpose of calculating the current historical crop height estimate) is established using the historical crop height estimate - more specifically, by multiplying the historical crop height estimate by a fraction (e.g., 0.95).

[0112] During 1120, the host controller 202 updates the historical crop height estimate. By way of a particular example, the updated value of the historical crop height estimate is presented by a filtering operation based on the current historical crop height estimate, the best estimate of the current crop height (presented during 1110), and a filter time constant selected for long term stability over a few seconds of machine operation. Thus, a historical estimate of the crop height is established over time. Based on the comparison mentioned above, the historical estimate can be updated with a new original crop height measurement or by a previous crop height estimate recalculated. In doing so, a correctly updated crop height estimate is provided upon detection of an actual crop and land. Using the alternative update value, based on the historical estimate, it is ensured that the crop height estimate converges to the actual crop height over time even in the absence of proper crop or land measurements. By Figure 11Upon initialization of the operations summarized, the operator can provide an initial crop height estimate. Alternatively, a minimum crop height can be provided (e.g., for potato fields, the berm height is provided as a minimum).

[0113] For example, the historical crop height (see description below of Figure 11 ) is generated (during 1120 described below) by a filtering operation with a relatively high time constant parameter (i.e., providing a relatively slowly changing output).

[0114] For example, when presenting the current crop height estimate (during 1120), the historical crop height estimate is weighted 0.95, while the current crop height measurement is weighted 0.05.

[0115] In exemplary embodiments, a user interface is provided through which the user interacts (even overrides) the operation of the automatic multi-wing / segment boom control system described herein with reference to Figure 2 From a design / feature perspective, the control configuration and operation of the main controller 202 is separate from manual control and configuration performed by the user / operator via manual controls via a user interface operating in a complementary manner to the main controller 202. Further, in illustrative examples, the main controller 202 implements any of a variety of implement / tractor electronic interfaces to facilitate more complete observation of the operating environment of the spray boom assembly 100, including operating parameters of the associated tractor / implement carrying / towing apparatus.

[0116] A function of the exemplary user interface is to enable the user / operator to tune the configuration of the control algorithm executed by the main controller 202. In addition, the exemplary control arrangement executed by the main controller 202 supports a broad and expandable set of configurable control operation settings. User selection of configurable sensor modes (streaming, filtering, noise suppression, etc.) and controller responsiveness / aggressiveness (e.g., coarse adjustment to profile changes: actuator response time constant and signal change magnitude limit) enables the user to make substantial changes to the performance and / or stability of the main controller 202 - without reliance on the “self-learning / repair” capabilities of the main controller 202, which would unnecessarily delay the adaptation of the boom leveling system to rapidly changing operating environments. In the case where the boom includes foldable wingtips, the user can be provided with an optional configuration interface that allows the user to specify such major changes (including operation of the boom system 100 in a semi-folded physical configuration) to the main controller 202 to ensure proper modeling of the boom system by the controller when implementing the dynamic model.

[0117] Other potential configuration inputs that affect the operation of the control operations performed by the main controller 202 include: providing geospatial mapping information and control / response configuration based on contour patterns (e.g. landmasses) from previous runs through the same field using the same spray boom assembly 100 to help configure the controller 202.

[0118] From the user’s perspective, the configuration and utilization of the spray boom leveling control performed by the main controller 202 can be quite simple. By way of example, the user customizes the acquired control program and configuration by providing a basic set of configuration parameters including identification of the manufacturer / model / version of the spray boom assembly to establish a rough configuration independent of usage. Following the rough configuration of the main controller 202 is a self-learning operation based on a few test scans / turns at the intended operating speed to fine tune the signal noise rejection and filtering of the main controller 202 input and output signals. Thereafter, during field operations, the system continues to receive and process sensor and user inputs to continually adapt to various changing operating conditions. Over time, the system can learn its own behavior and desires from the operator and use this to automatically tune (self-tune) its performance. Thus, the control configuration performed by the main controller 202 adapts / changes over time. Self-tuning / automatic parameter optimization, deep learning, machine learning, and artificial intelligence can also be important parts of enabling the main controller 202 to seamlessly adapt to the mechanical system in question.

[0119] While example instances have been depicted and described with reference to the towed spray boom assembly 100 as described above, the present disclosure is not limited to such systems. It will be readily appreciated that the advantages of the present disclosure also apply to various spray boom assembly geometries and applications in view of the present disclosure. Thus, the present disclosure is intended to apply to a wide variety of spray boom assemblies and associated electronic control systems - with appropriate adjustments to the determinations described above to accommodate changes in requirements and response characteristics due to the presence of different types of boom geometries, sensors, and actuators.

[0120] All references, including publications, patent applications, and patents, cited herein are hereby incorporated by reference to the same extent as if each reference were individually and specifically indicated to be incorporated by reference and were specifically set forth in its entirety herein.

[0121] In the context of describing the application (especially in the context of the following claims), the use of the terms “one” and “a” and “the” and “at least one” and similar referents in reference to an element should be interpreted to cover both the singular and plural unless otherwise indicated by context and / or the specification. The use of the term “at least one” followed by a list of one or more items (for example, “at least one of A and B”) should be interpreted to mean one item from the list of items (A or B) or any combination of two or more of the items in the list (A and B). The terms “comprising,” “having,” “including,” and “containing” are to be construed as open-ended terms (i.e., meaning “including, but not limited to,”) unless otherwise noted. Recitation of ranges of values herein are merely intended to serve as a shorthand method of referring individually to each separate value falling within the range, unless otherwise indicated herein, and each separate value is incorporated in the specification as if it were individually recited herein. All methods described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The use of any and all examples, or exemplary language (e.g., “such as”) provided herein, is intended merely to better illuminate the application and does not pose a limitation on the scope of the application unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the application.

[0122] Preferred embodiments of this application are described herein, including the best mode known to the inventors for practicing the application. Variations of those preferred embodiments can become apparent to those of ordinary skill in the art upon reading the foregoing description. The inventors expect skilled artisans to employ such variations as appropriate, and the inventors intend for the application to embrace all such modifications and equivalents. Accordingly, the application includes all modifications and equivalents that can come within the scope of the claims appended hereto as permitted by applicable law. Moreover, any combination of the above-described elements in all possible variations thereof is encompassed by the application unless otherwise indicated herein or otherwise evident to one of ordinary skill in the art.

Claims

1. A method for controlling the leveling of a multi-wing boom assembly by a multi-loop controller arrangement, the arrangement comprising a master controller giving actuation control commands to actuator controllers for causing positioning of actuators of the multi-wing boom assembly to reduce a determined current position error of components of the multi-wing boom assembly relative to a determined contour of a field currently being traversed by the multi-wing boom assembly, the method comprising: executing a master controller loop by the master controller according to a master controller repeating cycle period to generate positioning commands to be executed by the actuator controller, wherein actuation commands are given to the actuator controller for implementation within the master controller repeating cycle period; and An actuator controller loop is executed by the actuator controller independently of the master controller and according to an actuator controller repetitive cycle having a duration independent of the master controller repetitive cycle period to generate control signals for the actuators of the multi-wing boom assembly.

2. The method according to claim 1, wherein The main controller loop includes determining a position error.

3. The method according to claim 1, wherein The positioning commands are provided based on a dynamic model of the multi-wing boom assembly.

4. The method according to claim 3, wherein: The dynamic model includes a pendulum physical model of the multi-wing boom assembly.

5. The method according to claim 3, wherein The dynamic model includes a trapezoidal physical model of the multi-wing boom assembly.

6. The method according to claim 3, wherein: The positioning command is calculated based on a boom state based in part on boom state sensor data.

7. The method according to claim 6, wherein: The boom state is enhanced by state information based on off-boom measurements provided by a state observer.

8. The method according to claim 2, wherein: The main controller presents the position error based on distance measurement values ​​imparted from ultrasonic distance measurements using an ultrasonic sensor.

9. The method according to claim 8, wherein The distance measurements are presented by an ultrasonic distance measurement arrangement of single distance per transmitted pulse.

10. The method according to claim 8, wherein The distance measurements are presented by an ultrasonic distance measuring arrangement at multiple distances per transmitted pulse.

11. The method according to claim 8, wherein The ultrasonic distance measurement is generated using a peak detection based filter.

12. The method according to claim 8, wherein The raw current crop height is based on a multi-sample measurement sequence, and wherein the current to-crop distance is determined based on a comparison of: the original current crop height, and The historical crop height estimate is given by a filtering operation performed on the historical crop height estimates and the current estimated crop height.

13. The method according to claim 1, wherein The positioning commands include positioning commands for a boom raised position.

14. The method according to claim 1, wherein The main controller loop is executed based on the current boom angle measurement to perform a boom flap balancing operation.

15. The method according to claim 1, wherein The actuator controller is further configured to perform an obstacle avoidance control operation based on the profile sensor distance measurements.

16. The method according to claim 3, wherein The dynamic model is enhanced by modeling the mechanical configuration of the actuator related components.

17. A system for performing controlled leveling of a multi-wing boom assembly by a multi-loop controller arrangement, comprising: a main controller that gives an actuation control command; and an actuator controller for causing positioning of actuators of the multi-wing boom assembly to reduce a determined current position error of components of the multi-wing boom assembly relative to a determined contour of a field over which the multi-wing boom assembly is currently traversing, Wherein, the main controller and the actuator controller are configured to perform a method, the method comprising: executing a master controller loop by the master controller according to a master controller repeating cycle period to generate positioning commands to be executed by the actuator controller, wherein actuation commands are given to the actuator controller for implementation within the master controller repeating cycle period; and An actuator controller loop is executed by the actuator controller independently of the master controller and according to an actuator controller repetitive cycle having a duration independent of the master controller repetitive cycle period to generate control signals for the actuators of the multi-wing boom assembly.

18. The system according to claim 17, wherein: The main controller loop includes determining a position error.

19. The system according to claim 17, wherein: The positioning commands are provided based on a dynamic model of the multi-wing boom assembly.

20. The system of claim 19, wherein: The dynamic model includes a pendulum physical model of the multi-wing boom assembly.

21. The system of claim 19, wherein: The dynamic model includes a trapezoidal physical model of the multi-wing boom assembly.

22. The system of claim 19, wherein: The positioning command is calculated based on a boom state based in part on boom state sensor data.

23. The system of claim 22, wherein: The boom state is enhanced by state information based on off-boom measurements provided by a state observer.

24. The system of claim 18, wherein: The main controller presents the position error based on distance measurement values ​​imparted from ultrasonic distance measurements using an ultrasonic sensor.

25. The system of claim 24, wherein: The distance measurements are presented by an ultrasonic distance measurement arrangement of single distance per transmitted pulse.

26. The system of claim 24, wherein: The distance measurements are presented by an ultrasonic distance measuring arrangement at multiple distances per transmitted pulse.

27. The system of claim 24, wherein: The ultrasonic distance measurement is generated using a peak detection based filter.

28. The system of claim 24, wherein: The raw current crop height is based on a multi-sample measurement sequence, and wherein the current to-crop distance is determined based on a comparison of: the original current crop height, and The historical crop height estimate is given by a filtering operation performed on the historical crop height estimates and the current estimated crop height.

29. The system of claim 17, wherein: The positioning commands include positioning commands for a boom raised position.

30. The system of claim 17, wherein: The main controller loop is executed based on the current boom angle measurement to perform a boom flap balancing operation.

31. The system of claim 17, wherein: The actuator controller is further configured to perform an obstacle avoidance control operation based on the profile sensor distance measurements.

32. The system of claim 19, wherein: The dynamic model is enhanced by modeling the mechanical configuration of the actuator related components.

Images