Work vehicles

Abstract

To provide a work vehicle that can prevent a decrease in work accuracy, etc., even if a malfunction occurs in the positioning device during autonomous driving. [Solution] The above problem is solved by a work vehicle 1 configured to be able to drive automatically in a field, comprising: a shooting device that acquires field image data by shooting the area in front of the work vehicle; an uncultivated area extraction unit that extracts uncultivated areas from the field image data; a cultivated area extraction unit that extracts cultivated areas; a boundary line calculation unit that calculates the boundary line between uncultivated and cultivated areas of the field based on the extracted uncultivated and cultivated areas; a boundary line analysis unit 57 that analyzes the boundary line calculated by the boundary line calculation unit; and a boundary line information recording unit that records the analysis information regarding the boundary line from the boundary line analysis unit as boundary line recording information; and a driving means that automatically steers the work vehicle during automatic driving based on the field image data and boundary line recording information acquired by the shooting device.

Description

Technical Field

[0005] ,

[0001] The present invention relates to a work vehicle that performs farming operations while automatically driving in a field.

Background Art

[0002] For example, as shown in Patent Document 1 below, there is known a work vehicle that can switch between manual driving in which an operator operates a steering wheel (so-called, a handle) and automatic driving in which the steering wheel is automatically steered by an in-vehicle computer. This conventional work vehicle calculates its own position using position information obtained from a satellite positioning system, and automatically controls the steering angle of the steering wheel so as to eliminate the deviation between the calculated own position and the target travel route, thereby automatically traveling along the target travel route in the field while performing farming operations (hereinafter, sometimes simply referred to as operations).

[0003] In addition, in the field of agricultural machinery, various efforts have been made to analyze and utilize image information captured by a camera during operation. For example, as shown in Patent Document 2 below, a control device processes an image captured by a photographing device (CCD camera), detects the boundary between unplowed land and plowed land in a field, and controls the traveling direction of the vehicle body by acquiring position information. When it is determined that the vehicle body is deviating from the boundary between unplowed land and plowed land and is traveling, a technique for ensuring safety by immediately stopping the vehicle body is known.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] In conventional work vehicles, during autonomous driving, the vehicle automatically steers itself to follow a target route by acquiring positional information. Therefore, if a malfunction occurs in the positioning device, making it impossible to acquire positional information, or if errors occur in positional information due to factors such as solar flares, autonomous driving may not be able to continue, or the vehicle may deviate from the target route. This would reduce work accuracy and safety, and stopping the vehicle would significantly decrease work efficiency. Thus, with conventional work vehicles, a malfunction in the positioning device during autonomous driving raised concerns about a decline in work accuracy, safety, and work efficiency.

[0006] Therefore, the present invention aims to solve these problems and provide a work vehicle that can prevent a decrease in work accuracy, safety, and work efficiency even if a malfunction occurs in the positioning device during autonomous driving. [Means for solving the problem]

[0007] To achieve the above objective, the first invention is: A work vehicle comprising a vehicle body that travels on a field, a work implement mounted on the vehicle body, a positioning device that measures the vehicle's own position, and a steering device that steers the steering wheel, wherein the steering device is controlled based on the vehicle's own position information acquired by the positioning device, thereby automatically steering the steering wheel and enabling automatic operation on a field, A camera that acquires field image data taken from the front of the aforementioned work vehicle, An uncultivated land area extraction unit for extracting uncultivated land areas from the image data of the field, and a cultivated land area extraction unit for extracting cultivated land areas from the image data of the captured image, The system comprises: a boundary line calculation unit that calculates the boundary line between uncultivated and cultivated land in a field based on the uncultivated and cultivated land areas extracted by the uncultivated land area extraction unit and the cultivated land area extraction unit; a boundary line analysis unit that analyzes the boundary line calculated by the boundary line calculation unit; and a boundary line information recording unit that records the analysis information regarding the boundary line from the boundary line analysis unit as boundary line record information. The present invention provides a work vehicle equipped with a driving means that automatically steers the vehicle during autonomous driving based on the field image data and boundary line recording information acquired by the imaging device.

[0008] According to the first invention described above, by providing a driving means that automatically steers based on field image data and boundary line recording information acquired by a photographing device, it is possible to provide a work vehicle that can prevent a decrease in work accuracy, safety, and work efficiency even if a malfunction occurs in the positioning device during automatic driving.

[0009] The second invention is, in the first invention described above, The boundary line recording information includes information regarding the angle and distance of the boundary line relative to the reference point, The driving means is configured to automatically steer the vehicle in a manner that, during automatic driving, calculates the angle and distance of the boundary line relative to a reference point from the field image data acquired by the imaging device using the boundary line analysis unit, and compares the calculated angle and distance with the angle and distance of the boundary line relative to a reference point recorded in the boundary line recording information, thereby eliminating any deviation in the angle and distance of the boundary line.

[0010] According to the second invention described above, in addition to the effects of the first invention described above, Based on the boundary line between uncultivated and cultivated land, the current direction of travel of the machine can be corrected (automatically steered) to reproduce the direction of travel at the time the boundary line recording information was recorded. As a result, automatic driving can be optimally performed to follow the trajectory of the target driving path without relying on the acquisition of position information. Consequently, even if a malfunction occurs in the positioning device during automatic driving, a decrease in work accuracy, safety, and work efficiency can be more effectively prevented. [Effects of the Invention]

[0011] According to the present invention, it is possible to provide a work vehicle that can prevent a decrease in work accuracy, safety, and work efficiency even if a malfunction occurs in the positioning device during autonomous driving. We can provide work vehicles. [Brief explanation of the drawing]

[0012] [Figure 1] Figure 1 is a left side view of a work vehicle according to an embodiment of the present invention. [Figure 2] Figure 2 is a schematic diagram of the front of the aircraft cabin. [Figure 3] Figure 3 is a block diagram showing the configuration of the control system including the control device C of the work vehicle 1. [Figure 4] Figures 4A and 4B are explanatory diagrams showing an example of a target travel route L designed by the target travel route design unit 53. [Figure 5] Figures 5A to 5E are explanatory diagrams showing how the work of the work vehicle progresses based on the designed target travel path. [Figure 6] Figure 6 is a flowchart showing the flow of boundary line recording processing. [Figure 7] Figure 7 is a flowchart showing the flow of the boundary line calculation and analysis process. [Figure 8] Figures 8A to 8C show image representations of field data acquired at the starting point, intermediate point, and ending point of Figure 5B. [Figure 9] Figures 9A to 9C show an analysis image of the field image data shown in Figures 9A to 9C. [Figure 10] Figures 10A to 10C show image representations of field data acquired at the starting point Le, the intermediate point Lm, and the ending point Lm, respectively, in Figure 5C. [Figure 11] Figures 11A to 11C show the analysis image of the field image data shown in Figures 10A to 10C. [Figure 12] FIG. 12 is a flowchart showing the flow of the boundary line automatic driving process related to boundary line automatic driving. [Figure 13] FIG. 13 is a flowchart showing the flow of the automatic driving switching process. [Figure 14] FIG. 14 is a diagram showing an analysis image of field image data according to another embodiment. [Figure 15] FIG. 15 is a rear side perspective view of a work vehicle according to another embodiment. MODE FOR CARRYING OUT THE INVENTION

[0013] Hereinafter, with reference to the accompanying drawings, preferred embodiments of the present invention will be described while referring to the drawings. In the following description, unless otherwise specified, the forward direction of the work vehicle 1 (the direction from the driver's seat 8 described later toward the steering wheel 9) is taken as the front, the reverse direction as the rear, the right side when facing forward as the right, and the left side as the left. Also, the main body of the work vehicle 1 may simply be referred to as the vehicle body.

[0014] <1. Basic Configuration of Work Vehicle> First, the basic configuration of the work vehicle 1 according to the embodiment will be described with reference to FIG. 1. FIG. 1 is a left side view of the work vehicle 1 according to the embodiment, and FIG. 2 is an external view inside the cab 7r viewed from the front of the driver's seat 8 in FIG. 1. Hereinafter, as an example of the work vehicle 1, a tractor will be described. Therefore, in the following description, the work vehicle 1 will mainly be described as the tractor 1.

[0015] The tractor 1, which is a work vehicle, is an agricultural tractor that performs work in a field while self-propelling. Further, the tractor 1 is configured to be able to execute a predetermined work while manually or automatically driving in the field by a control system centered on a control device C (see FIG. 3) disposed at an appropriate position on the vehicle body, in addition to a driver (also referred to as an operator) boarding and traveling in the field while executing a predetermined work.

[0016] As shown in Figure 1, the tractor 1 comprises a drivable vehicle body 2 and an implement W. The vehicle body 2 comprises a vehicle frame 3, front wheels 4, rear wheels 5, a bonnet 6, an engine E, a control unit 7, and a transmission case 10. The vehicle frame 3 and the transmission case 10 form the machine's skeleton and function as the main frame of the vehicle body 2.

[0017] The front wheels 4 are a pair, left and right, and primarily serve as steering wheels (i.e., steering wheels). The rear wheels 5 are a pair, left and right, and primarily serve as drive wheels (i.e., drive wheels). The tractor 1 may be configured to switch between two-wheel drive (2WD), where the rear wheels 5 are driven, and four-wheel drive (4WD), where both the front wheels 4 and rear wheels 5 are driven. In this case, both the front wheels 4 and rear wheels 5 are drive wheels. The vehicle body 2 may be equipped with crawler tracks instead of wheels (front wheels 4 and rear wheels 5). In this case, the crawler tracks function as drive wheels.

[0018] The bonnet 6 is provided at the front of the vehicle body 2 so as to be able to be opened and closed. The bonnet 6 can rotate (open and close) vertically with its rear as the pivot point. When closed, the bonnet 6 covers the engine E mounted on the vehicle frame 3. The engine E is the power source for the tractor 1 and is a heat engine such as a diesel engine or a gasoline engine.

[0019] The control unit 7 functions to operate the work vehicle 1 by receiving input from the operator. A control room 7r is provided within a cabin box 7a that covers the upper part of the vehicle body 2, forming a cabin. A driver's seat 8 is located inside the control room 7r, and various operating members that receive input from the operator, such as a steering wheel 9, are located in front of the driver's seat 8. An air conditioning unit 7e is located behind the driver's seat 8. The steering wheel 9 is a component that steers the front wheels 4, which are the steering wheels. During manual operation, the operator steers the wheel manually, and during automatic operation, the wheel is automatically steered by a steering device 31 which includes steering actuators (not shown) and the like.

[0020] A steering angle sensor 25 is provided at the rotation base of the steering wheel 9, which can detect the direction of rotation of the steering wheel 9 and the steering angle corresponding to the amount of operation (see Figure 3). This steering angle sensor 25 is composed of, for example, a rotary encoder, and can detect the rotation angle (steering angle) around the rotation axis of the input shaft (not shown) of the steering wheel 9. Since the direction of travel of the work vehicle A is determined according to this steering angle, the control device C, which will be described later, acquires information regarding the detected value of the steering angle sensor 25 and controls the steering angle of the steering wheel 9 with the steering device 31, thereby enabling control of the direction of travel of the work vehicle A.

[0021] The transmission case 10 houses the transmission (gearbox 32). The power (rotational power) output from engine E is appropriately reduced (shifted) by the transmission and transmitted to the front wheels 4 and rear wheels 5 via the front axle 4j and rear axle 5j, and also transmitted (supplied) to the PTO shaft 16. In addition, a PTO clutch and an auto brake device (braking device) 33 (not shown) are housed in the transmission case 10 (see Figure 3), and the transmission of power to the PTO shaft (on / off, shifting) can be controlled. As a result, the work vehicle 1 is able to control the drive of the work machine W.

[0022] A work implement W for use in the field is connected to the rear of the vehicle body 2, and a lifting device 12 for raising and lowering the work implement W is provided. Furthermore, a PTO shaft 16 that transmits power to drive the work implement W is positioned to protrude rearward from the transmission case 10. The PTO shaft 16 transmits rotational power, which has been appropriately reduced by the transmission, to the work implement W connected to the rear of the vehicle body 2.

[0023] The lifting device 12 comprises a hydraulic lifting cylinder 121, a lift arm 122, a lift rod 123, a lower link 124, and a top link 125. The lifting device 12 can move the implement W to a non-working position by raising it. The non-working position is the position of the implement W when it is raised, for example, when the vehicle body 2 is reversing or turning, and the implement W is supported by the vehicle body in mid-air, away from the field. The lifting device 12 can also move the implement W to a ground-level working position by lowering it. The ground-level working position is the position of the implement W when it has been lowered for work, and the implement W is supported by the vehicle body while in contact with the field.

[0024] When hydraulic fluid is supplied to the lifting cylinder 121, the lift arm 122 rotates around the pivot point axis AX to raise the work implement W, and when the hydraulic fluid is discharged from the lifting cylinder 121, it rotates around axis AX to lower the work implement W. A lift arm sensor 26 is provided at the base of the lift arm 122 (near axis AX) to detect the rotation angle of the lift arm 122. The height of the work implement W is calculated by the control device C based on the detection result of the lift arm sensor 26.

[0025] Furthermore, the lift arm 122 is connected to the lower link 124 via the lift rod 123. In this way, the lifting device 12 connects the work implement W to the vehicle body 2 so that it can be raised and lowered by the lower link 124 and the top link 125. The lower link 124 is attached to the rear of the transmission case 10.

[0026] Although not described in detail here, the tractor 1 is equipped with, in addition to the lift arm sensor 26 mentioned above, an engine speed sensor (not shown) for detecting the rotational speed of the engine E, and a mileage sensor 24 for detecting the distance traveled by the machine (see Figure 3). This mileage sensor 24 detects the distance traveled by the machine based, for example, on the rotational speed of the rear axle 5j. Also, although not shown, a steering angle sensor for detecting the steering angle of the front wheels 4, which are the steering wheels, and lever sensors for detecting the operating position of various control levers such as the auxiliary transmission lever 14 are each placed at appropriate positions. The control device C is configured to acquire detection information from these sensors via wired or wireless connection.

[0027] The positioning device 30 is a device that determines the position of the aircraft. For example, it is composed of a GNSS (Global Navigation Satellite System) antenna and can perform positioning and timing by receiving radio waves from navigation satellites S orbiting in the sky. It can also calculate the speed of movement from the history of positioning results and the Doppler effect of radio waves. The control device C, which will be described later, calculates the position of the aircraft by acquiring positioning information (in other words, the position information of the aircraft) from the positioning device 30 during automatic driving, and controls the steering device 31 to eliminate the deviation from a pre-set target driving path (hereinafter referred to as the target driving path), thereby enabling automatic driving. The positioning device 30 is also equipped with an IMU (Inertial Measurement Unit), which can simultaneously measure the inclination angle of the vehicle body 2 (i.e., the inclination of the field surface).

[0028] The implement W is a machine that performs work in the field. In the example shown in Figure 1, implement W is a rotary tiller that performs tilling work in the field, but the type of implement W is not limited to this. The rotary tiller tills the field surface (soil) by rotating the tilling tines 61 with power transmitted from the PTO shaft 16.

[0029] Furthermore, the tractor 1 is equipped with a control device C (see Figure 3). The control device C can, for example, control the engine E, control the travel speed of the vehicle body 2, and perform actions such as raising and lowering the implement W.

[0030] Furthermore, the tractor 1 is configured to be able to communicate wirelessly with an information processing terminal (a portable information terminal such as a smartphone or tablet) 100, allowing the operator to perform various settings and check information on the tractor 1 by operating the portable information terminal 100. The portable information terminal 100 includes, for example, a storage unit consisting of a hard disk, ROM (Read Only Memory), RAM (Random Access Memory), etc., and a display unit and operation unit consisting of a touch-type liquid crystal display 110. In addition, various keys and buttons may be provided separately as part of the operation unit.

[0031] Furthermore, the tractor 1 may be equipped with an obstacle detection sensor as an obstacle detection means, which detects obstacles (people, animals, objects, etc. that hinder movement in a field). This obstacle detection sensor 20 has the function of detecting obstacles around the machine and is composed of a 3D sensor, a so-called Lider (Light Detection and Ranging), which measures scattered light in response to laser irradiation. This Lider is an object detection method that obtains the direction and distance to an object by irradiating the object with near-infrared light, visible light, or ultraviolet light and detecting the reflected light with a light sensor, and can acquire 3D point cloud data around the machine. In addition, the obstacle detection sensor may be attached to the rear of the vehicle body 2 in order to detect obstacles located behind the machine, as well as in front of the machine.

[0032] Next, referring to Figure 2, the configuration of the cockpit 7r will be explained. As mentioned above, a steering wheel 9 is provided in front of the cockpit 8. A clutch pedal 18 is provided on the lower left side of the steering post 350 to which the steering wheel 9 is attached, and an accelerator pedal 19 and a brake pedal 15 are provided on the lower right side of the steering post 350. The brake pedals 15 are a pair, 15L and 15R, to allow for independent braking on the left and right sides.

[0033] A forward / reverse lever 201 is located on the upper left side of the handle post 350. An accelerator lever 351 for adjusting the engine E's rotation speed, a turn signal lever 352, and other components are located on the upper right side of the handle post 350. An engine key switch 353 and a PTO shift lever 354 for operating the engine E's drive (on / off) are located on the driver's seat 8 side of the handle post 350.

[0034] Furthermore, as shown in Figure 3, a dashboard cover 355 is provided in front of the steering wheel 9. The dashboard cover 355 is also equipped with an instrument panel 11 that is visible to the driver in the cockpit 8. The instrument panel 11 is equipped with a touch panel 356 and an engine tachometer 357, among other things. The touch panel 356 functions as an input unit 356a for inputting various information to the control device C, which will be described later, and a display unit 356b for displaying various information (see Figure 3).

[0035] Furthermore, although this configuration is well known and will not be described in detail, the left and right sides of the cockpit 8 are equipped with a main transmission lever, a sub-transmission lever, an accelerator lever, a position lever, a lifting position setting means (lifting height dial), a public road driving button, and a control panel storage compartment. Of these, the position lever is operated when raising or lowering the lift arm 122. In addition, various operation switches are provided, such as an automatic / manual PTO switch, a PTO on / off switch, an engine speed indicator, a speed increase adjustment switch, and a speed decrease adjustment switch. The control panel storage compartment houses a control panel equipped with other operation switches not mentioned above. Furthermore, voice output devices (so-called speaker devices) capable of outputting sound are placed in appropriate locations within the cockpit 7r, enabling various announcements by voice.

[0036] Next, we will describe the camera device 13 of the tractor 1. Inside the cockpit 7r, a camera 13 is installed on the ceiling approximately above the cockpit 8 to photograph the field in front of the vehicle. This camera 13 consists of a ceiling-mounted camera and photographs the foreground (forward scenery) of the work vehicle 1 at predetermined time intervals, thereby generating image data of the field in front of the work vehicle 1 (hereinafter referred to as field image data). More specifically, in this embodiment, the foreground camera 13a is configured to photograph the area in front from inside the cockpit 7r over a predetermined range, and is configured to include at least the windshield 14f in front of the cabin box 7a as part of the shooting range. The field image data also includes information about the time the image was taken. In this embodiment, the foreground camera 13a is shown to be located on the ceiling approximately above the cockpit 8, but the installation location is not limited to this, and may be located, for example, on the dashboard cover 355 or at an appropriate location on the cabin box 7a. Furthermore, the imaging device 13 may be configured as a so-called AI camera device, and by incorporating a trained model that has learned the correspondence between cultivated and uncultivated areas of a field and the cultivated and uncultivated areas of the field, it may substitute for and perform the functions of the uncultivated area extraction unit 54 and cultivated area extraction unit 55 described later.

[0037] <2. Regarding the configuration of the control device> Figure 3 is a block diagram showing the configuration of the control system of the work vehicle 1, including the control device C. The control device C is an information processing device composed of multiple ECUs (Electronic Control Units). Each of these ECUs is equipped with a CPU for performing calculations and a memory capable of reading and writing information necessary for calculations. The configuration shown as a functional block in Figure 3 is realized when the CPU operates according to various control programs stored in the memory.

[0038] As shown in Figure 3, the control device C has a positioning device 30, a camera 13, a distance sensor 24, a steering angle sensor 25, a lift arm sensor 26, an engine key switch 353, and an input unit 356a connected to its input side via an input / output signal processing unit (including a communication unit) not shown. Through this, it acquires positioning information (location information of the machine) from the positioning device 30, distance data indicating the machine's distance traveled from the distance sensor 24, field image data from the camera 13, and various input information from the input unit 356a. It also acquires operation information from the operator's actions and transmits display information to the portable information terminal 100, which is connected to send and receive information bidirectionally, and transmits display information instructing the image to be displayed on the portable information terminal 100.

[0039] As described above, the steering device 31 is configured to include a steering actuator and the like for rotating the steering wheel 9, and is a device that automatically steers the steering wheel 9 during autonomous driving. The transmission 32 is a transmission housed in the transmission case 10, and is a device that changes the speed of the rotational power output from the engine E. The auto brake device 33 is a device that can brake (restrict rotation of) the rear wheels 5 by controlling the extension and retraction of an electric cylinder (not shown), without being operated by the brake pedal 15.

[0040] Furthermore, the control device C includes a communication unit 59, which is a communication mechanism that connects to external devices physically separated from the control device C via a network NW and exchanges information through communication. The communication unit 59 is, for example, a wireless communication module that can connect to a wireless router outside the device. In this embodiment, the communication unit 59 is connected to the portable information terminal 100 via the network NW, enabling the transmission and reception of information through bidirectional communication.

[0041] The portable information terminal 100 is a small information processing terminal such as a portable telephone (smartphone), tablet, notebook computer, or wearable device in the form of glasses or a wristwatch, and is equipped with a touch panel display 101 (see Figure 1).

[0042] The control device C includes a manual driving control unit 51 that controls driving in manual driving mode and an automatic driving control unit 52 that controls driving in automatic driving mode. Here, controlling driving means, more specifically, acquiring necessary detection information from various sensors for each mode, sending necessary control commands to the driving system ECU that governs the operation during driving when each mode is selected, and causing the vehicle body 2 to move. The manual driving control unit 51 can also execute manual driving mode. When this manual driving mode is executed, the operator can steer the machine by operating the steering wheel 9 and drive it in the field. The automatic driving control unit 52 can also execute automatic driving mode. When this automatic driving mode is executed, a target driving path is designed in advance before the work machine W starts, and information regarding the designed target driving path is stored in the work information database 60. Based on the position information acquired from the positioning device 30, the control device C controls various parts such as the engine E, steering device 31, transmission 32, auto brake device 33, and lifting device 12 so that the machine drives along the pre-designed target driving path.

[0043] Furthermore, the control device C includes a target travel path design unit 53 for designing a target travel path, an uncultivated area extraction unit 54 for extracting uncultivated areas from field image data, a cultivated area extraction unit 55 for extracting uncultivated areas from field image data, a boundary line calculation unit 56 for calculating boundary lines based on the uncultivated and cultivated areas of the field image data extracted by the uncultivated area extraction unit 54 and the cultivated area extraction unit 55, a boundary line analysis unit 57 for analyzing the boundary lines calculated by the boundary line calculation unit 56, and a boundary line information recording unit 58 for recording analysis information regarding the boundary lines obtained by the analysis of the boundary line analysis unit 57.

[0044] The uncultivated land area extraction unit 54 performs image analysis of the field image data acquired by the imaging device 13 and extracts the uncultivated land area within the image. For example, the uncultivated land area extraction unit 54 can extract the uncultivated land area by inputting the acquired field image data into a trained model that has learned the correspondence between image data containing uncultivated land in the field and the uncultivated land area, and obtaining an estimated area of ​​the uncultivated land. Alternatively, a known method for extracting uncultivated land may be used, which determines a threshold by comparing and identifying the changes in brightness in the field image data.

[0045] The cultivated land area extraction unit 55 performs image analysis of the field image data acquired by the imaging device 13 and extracts the cultivated land area within the image. For example, the cultivated land area extraction unit 55 can extract the uncultivated land area by inputting the acquired field image data into a trained model that has learned the correspondence between image data including uncultivated land and the cultivated land area, and obtaining an estimated area of ​​the cultivated land. Alternatively, a known method for extracting cultivated land may be used, which determines a threshold by comparing and identifying the changes in brightness in the field image data.

[0046] The boundary line calculation unit 56 performs the function of calculating boundary lines based on the uncultivated and cultivated areas of the field image data extracted by the uncultivated area extraction unit 54 and the cultivated area extraction unit 55. A boundary line is a virtual line that indicates the boundary between cultivated and uncultivated land. For example, the boundary line calculation unit 56 calculates multiple points of contact between the uncultivated and cultivated areas in the field image data, and then performs regression analysis on the calculated points of contact to calculate the boundary line in the acquired field image data.

[0047] The boundary line analysis unit 57 performs the function of analyzing the boundary line calculated by the boundary line calculation unit 56. The boundary line analysis unit 57 calculates the angle and distance of the calculated boundary line with respect to the reference point. Details will be described later.

[0048] The boundary line information recording unit 58 performs the function of recording analysis information regarding boundary lines obtained by the analysis of boundary line analysis unit 57. In the following, the analysis information regarding boundary lines recorded by the boundary line information recording unit 58 will be referred to as boundary line recording information. The boundary line information recording unit 58 records the boundary line recording information in association with information regarding the distance traveled acquired by the distance traveled sensor 24. Furthermore, it may be configured to record it in association with position information obtained by the positioning device 30. This allows the control device C to determine which point in the field the recorded boundary line recording information represents. In addition, the boundary line information recording unit 58 creates and records right boundary line recording information, which records the analysis information of the boundary line when the calculated boundary line is to the right of the reference point, and left boundary line recording information, which records the analysis information of the boundary line when the calculated boundary line is to the left of the reference point. In other words, the boundary line recording information consists of right boundary line recording information and left boundary line recording information. The boundary line recording information is recorded, for example, in the work information database 60 described later.

[0049] The control device C is connected to a work information database 60, which is a storage device for storing work information. The work information includes various types of information necessary for the work, such as field information, target travel route, and working width of the implement W. The field information also includes field information necessary for the work, such as map information, shape, location, size, range, latitude, longitude, altitude, and entrance information for each field to be worked on. The work information database 60 may be an external server configured to send and receive information from the control device C via a network NW.

[0050] <3. Design of the target travel route> Figures 4A and 4B are explanatory diagrams showing an example of a target travel route L designed by the target travel route design unit 53. As shown in Figure 4A, the operator first manually drives the work vehicle 1 to set a reference driving line L0, which serves as the reference for the direction of travel during automatic driving (teaching drive). At the start point Ls and end point Le of the reference driving line L0, the operator performs predetermined operations to instruct the input unit 356a to start and end the setting, thereby completing the setting of the reference driving line L0. Once the setting of the reference driving line L0 is complete, as shown in Figure 4B, a target driving path L is designed that alternates between a straight path Lj and a turning path Lt in order to efficiently travel within the field H. At this time, multiple straight paths Lj are designed to be parallel to the reference driving line L0, and their length is designed to be approximately the same as the reference driving line L0.

[0051] Furthermore, the spacing between straight paths Lj is determined by the working width of the implement W, which is set in advance, so that the working areas do not overlap. In addition, the number of straight paths Lj is determined based on information such as the shape and size of the field H contained in the work information stored in the work information database 60, and as a result, the work area Hr, which is the area to be worked on, is defined. In the illustrated example, six straight paths Lj, indicated by symbols L1 to L6, are designed. In addition, the turning path Lt is a path for the work vehicle 1 to move around by turning, and multiple turning paths Lt are designed to connect the starting point Ls and ending point Le of the straight paths Lj.

[0052] Figures 5A to 5E are explanatory diagrams showing how the work of the work vehicle 1 progresses based on the designed target travel path L. Specifically, Figure 5A shows the vehicle traveling along the first straight path L1 (also called the first outbound path), Figure 5B shows the vehicle traveling along the second straight path L2 (also called the first return path), Figure 5C shows the vehicle traveling along the third straight path L3 (also called the second outbound path), Figure 5D shows the vehicle traveling along the fourth straight path L4 (also called the second return path), and Figure 5E shows the vehicle traveling along the fifth straight path L5 (also called the third return path). The figures also show the cultivated land area Hk, which represents the area of ​​cultivated land, and the uncultivated land area Hm, which represents the area of ​​uncultivated land, both of which are included in the work target area Hr. The configuration of the control device C according to the present invention will be further explained below, using the travel of the work vehicle 1 shown in Figures 5A to 5E as an example.

[0053] <4. Regarding boundary line recording processing> The control device C is configured to record information regarding boundary lines, which are virtual lines indicating the boundary between cultivated and uncultivated land, while the work vehicle 1 is in motion. The control device C is configured to execute boundary line recording processing for creating boundary line recording information. The boundary line recording processing is performed by a series of processes carried out by the boundary line information recording unit 58.

[0054] Figure 6 is a flowchart showing the boundary line recording process. Figure 7 is a flowchart showing the boundary line calculation process. The boundary line recording process is configured to be automatically executed when the work vehicle 1 is traveling along the second straight path L2 (also called the first return path) and the third straight path L3 (also called the second outbound path), for example, by the control device C referring to information about the target travel path L and the position information of the machine. That is, the process starts at the starting point Ls of the second straight path L2 (also called the first return path) and the third straight path L3 (also called the second outbound path), and ends at the ending point Le.

[0055] When the boundary line recording process begins, the control device C calculates the distance traveled from the starting point Ls, as shown in Figure 6 (step #101). Next, it executes the boundary line calculation analysis process, which is a subroutine (step #102). This boundary line calculation analysis process will be described later. Based on the results of the boundary line calculation analysis process, it is determined whether the boundary line calculation was successful or not (Y in step #102). If it is determined that the boundary line calculation was successful, it is determined whether the boundary line is to the right of the reference point based on the boundary line analysis information (step #104). On the other hand, if it is determined that the boundary line calculation failed (N in step #103), the process returns to step #101.

[0056] Next, if the boundary line is to the right of the reference point (Y in step #104), the analysis information of the boundary line (angle and distance relative to the reference point) calculated by the boundary line analysis unit 57 is recorded as right boundary record information (step #105). On the other hand, if the boundary line is to the left of the reference point (N in step #104), the analysis information of the boundary line (angle and distance relative to the reference point) calculated by the boundary line analysis unit 57 is recorded as left boundary record information (step #106). Following the above procedure, the boundary line information recording unit 58 classifies and records the analysis information regarding the boundary line obtained by the analysis of the boundary line analysis unit 57 into right boundary record information and left boundary record information, depending on whether the boundary line is located to the left or right of the reference point, or in other words, whether the boundary line is located to the left or right of the machine. When one run (in other words, run along one straight path Lj) is completed, the boundary line recording process is completed (step #107).

[0057] <5. Regarding the boundary line calculation and analysis process> Figure 7 is a flowchart showing the flow of the boundary line calculation and analysis process. The boundary line calculation and analysis process is a process that analyzes the boundary lines calculated by the boundary line calculation unit 56, and a series of processes are executed by the uncultivated land area extraction unit 54, the cultivated land area extraction unit 55, and the boundary line analysis unit 57. When the boundary line calculation and analysis process is started, the control device C acquires field image data from the imaging device 13 (step #201). Subsequently, the uncultivated land area extraction unit 54 extracts the uncultivated land areas from the acquired field image data (step #202), and the cultivated land area extraction unit 55 extracts the cultivated land areas from the acquired field image data (step #203).

[0058] If the uncultivated land area extraction unit 54 and the cultivated land area extraction unit 55 successfully extract the uncultivated and cultivated land areas from the field image data (Y in step #204), the boundary line analysis unit 57 calculates boundary lines based on the extracted uncultivated and cultivated land areas of the field image data (step #205), draws a horizontal line for analysis (described later) from a reference point to the calculated boundary lines (step #206), calculates the angle of the calculated boundary lines (step #207), calculates the distance of the calculated boundary lines (step #208), outputs a message indicating successful boundary line calculation as a result of the processing, and returns to the main routine (step #209). On the other hand, if the extraction of uncultivated and cultivated land areas from the field image data fails, it outputs a message indicating failure to calculate boundary lines as a result of the processing, and returns to the main routine (step #210). For example, if only uncultivated or cultivated areas can be extracted from the field image data, the boundary line cannot be calculated, resulting in a failure to calculate the boundary line. Next, we will explain in detail the content of the boundary line calculation and analysis related to steps #206 to #208.

[0059] <6. Details of Boundary Line Calculation and Analysis> Figures 8A to 8C show image data of the field image data acquired at the starting point Le, the intermediate point Lm, and the ending point Lm, respectively, in Figure 5B. Figures 9A to 9C show an analysis image of the field image data shown in Figures 8A to 8C. As shown in Figures 8A to 8C, a reference point P is defined in the image data of the field image data. This reference point P may be defined by placing a marker (such as a fluorescent sticker) on the windshield 14f or hood 6 and photographing it with the imaging device 13, or by performing a predetermined image processing so that a predetermined position in the field image data becomes the reference point P. It is preferable, but not limited to, that the reference point P be set approximately in the center in the left-right direction in the field image data, as shown in the figures.

[0060] Figures 8A to 8C show that a cultivated land area exists to the right of the machine. As shown in Figures 9A to 9C, the uncultivated land area extraction unit 54 and the cultivated land area extraction unit 55 extract the uncultivated land area G1 and the cultivated land area G2, and calculate the boundary line Lx. Once the boundary line Lx is calculated, the boundary line analysis unit 57 draws a horizontal line Ly for analysis from the reference point P to the boundary line Lx, and calculates the angle (interior angle) between the horizontal line Ly and the boundary line Lx as the angle α of the boundary line Lx. It also calculates the length of the horizontal line Ly from the reference point P to the boundary line Lx (for example, the unit can be the number of pixels) as the distance of the boundary line Lx. In the examples shown in Figures 9A and 9C, the boundary line Lx is located to the right of the reference point P (the horizontal line Ly is drawn to the right). Therefore, the angle α and distance of the boundary line Lx, calculated by the boundary line analysis unit 57 as analysis information for the boundary line Lx, are recorded as right boundary line recording information.

[0061] Furthermore, Figures 10A to 10C show image representations of field image data acquired at the starting point Le, the intermediate point Lm, and the ending point Lm, respectively, in Figure 5C. Figures 11A to 11C show analysis images of the field image data shown in Figures 10A to 10C. In Figures 10A to 10C, a cultivated land area exists to the left of the machine, and in the same manner as in the case of the right side, the uncultivated land area extraction unit 54 and the cultivated land area extraction unit 55 extract the uncultivated land area G1 and the cultivated land area G2, and the boundary line Lx is calculated. In the example of Figures 11A to 11C, the boundary line Lx is located to the left of the reference point P (the horizontal line Ly is drawn toward the left), so the angle α and distance of the boundary line Lx calculated as analysis information of the boundary line Lx by the boundary line analysis unit 57 are recorded as left boundary line recording information.

[0062] <7. Regarding automatic driving along boundary lines> The control device C is configured to allow selection and execution of two driving methods during automatic driving (mode): a driving method that uses position information from the positioning device 30 for automatic driving (automatic steering) (hereinafter referred to as "position information automatic driving") and a driving method that uses field image data and boundary line recording information from the imaging device 13 for automatic driving (automatic steering) (hereinafter referred to as "boundary line automatic driving"). The operator can switch between position information automatic driving and boundary line automatic driving at appropriate timings by performing predetermined operations. Since known technology can be used for position information automatic driving, a detailed explanation will be omitted. The following will mainly describe boundary line automatic driving.

[0063] Figure 12 is a flowchart showing the flow of the boundary line automatic driving process related to boundary line automatic driving. When the boundary line automatic driving process is executed, the control device C calculates the distance traveled from the starting point Ls on the currently traveling straight path Lj using the travel distance sensor 24 (step #301). Next, it executes the boundary line calculation analysis process (see Figure 7) (step #302). Based on the results of the boundary line calculation analysis process, it is determined whether the boundary line calculation was successful or not (step #303). If it is determined that the boundary line calculation was successful (Y in step #303), it is determined whether the boundary line is to the right of the reference point based on the boundary line analysis information (step #304). On the other hand, if it is determined that the boundary line calculation failed (N in step #303), it returns to step #301. If the boundary line is to the right of the reference point (Y in step #304), the right boundary line recording information is referred to (step #305). If it is to the left (N in step #304), the left boundary line recording information is referred to (step #310).

[0064] Next, when the boundary line is to the right of the reference point, the control device C compares the angle α' of the boundary line calculated and analyzed in step #302 with the angle α of the boundary line in the right boundary line recording information. Furthermore, the control device C compares the distance D' of the boundary line calculated and analyzed in step #302 with the distance D of the boundary line in the right boundary line recording information (step #306). At this time, it refers to the travel distance calculated in step #301 and the analysis information of the boundary line associated with the same travel distance. As a result, when angle α' > α or distance D' > D, it can be determined that the current direction of travel of the aircraft is shifted to the left (a deviation occurs) compared to the direction of travel of the aircraft at the time the right boundary line recording information was recorded. In other words, it can be determined that the position of the aircraft is deviating to the left with respect to the target travel path L. Therefore, the control device C corrects the direction of travel to the right in order to eliminate the deviation. In other words, similar to when a deviation to the left occurs during autonomous driving based on position information, the steering angle is corrected by automatic steering, and the aircraft's direction of travel is corrected to the right (step #307).

[0065] On the other hand, when the angle α´ < α or the distance D´ < D (Y in step #308), it can be determined that the current traveling direction of the aircraft is shifted to the right (a deviation has occurred) with respect to the traveling direction of the aircraft at the time of recording the right boundary line recording information. That is, it can be determined that the position of the aircraft is deviated to the right with respect to the target traveling path L. Therefore, in order to eliminate the deviation, the control device C corrects the traveling direction to the left. That is, similar to the case where a left deviation occurs during automatic driving based on position information, the steering angle is corrected by automatic steering, and the traveling direction of the aircraft is corrected to the left (step #309).

[0066] Also, when the boundary line is to the left of the reference point (N in step #304), the control device C refers to the left boundary line recording information in the same manner, and compares the angle α´ of the boundary line calculated and analyzed in step #302 with the angle α of the boundary line in the left boundary line recording information, and compares the distance D´ of the boundary line calculated and analyzed in step #302 with the distance D of the boundary line in the right boundary line recording information, and corrects the traveling direction of the aircraft (steps #310 to step #314). The above procedure is repeated until the boundary line automatic driving ends (step #315). As a result, based on the boundary line between the unplowed land and the plowed land, the current traveling direction of the aircraft can be corrected (automatically steered) so as to reproduce the traveling direction at the time of recording the boundary line recording information. Therefore, even when a malfunction occurs in the positioning device 30 during automatic driving, it is possible to suitably prevent a decrease in work accuracy, safety, and work efficiency.

[0067] <8. Regarding the Automatic Switching Process of the Traveling Method> Hereinafter, an automatic driving switching process for automatically switching between position information automatic driving and boundary line automatic driving will be described. The automatic driving switching process is executed, for example, when automatic driving (mode) is started by a predetermined operation of an operator.

[0068] FIG. 13 is a flowchart showing the flow of the automatic driving switching process. When the automatic driving switching process is initiated, the control device C acquires position information from the positioning device 30 (step #401). If the acquisition of position information is successful (Y in step #401), automatic driving based on the position information (i.e., position-based automatic driving) is performed. On the other hand, if the acquisition of position information fails (N in step #401), there is a possibility that there is some malfunction in the positioning device 30, so the control device C performs boundary line calculation analysis processing (see Figure 7) in order to perform boundary line automatic driving (step #404). Based on the results of the boundary line calculation analysis processing, it is determined whether or not the boundary line calculation was successful (step #05), and if it is determined that the boundary line calculation was successful (Y in step #405), it is determined whether or not the boundary line is to the right of the reference point based on the boundary line analysis information (step #406). Next, if the boundary line is to the right of the reference point, it is determined whether there is right boundary line recording information (step #407). If the boundary line is to the left of the reference point, it is determined whether there is left boundary line recording information (step #408). If boundary line recording information exists in the direction corresponding to the left and right of the boundary line, automatic driving based on the boundary line recording information (i.e., boundary line automatic driving) is performed (step #409). With this automatic driving switching process, if a malfunction occurs in the positioning device 30, it is possible to automatically switch to boundary line automatic driving under predetermined conditions, thereby preventing a decrease in work accuracy, safety, and work efficiency more quickly and effectively.

[0069] The embodiments of the present invention have been described above. The present invention is not limited to the embodiments described above. It goes without saying that modifications can be made as appropriate within the scope of the technical idea. Another embodiment will be described below.

[0070] <9. Regarding alternative embodiments> Figure 14 shows an image of field image data analysis according to another embodiment. As shown in Figure 14, the boundary line analysis unit 57 may be configured to, once the boundary line Lx is calculated, draw an analysis horizontal line Ly from the reference point P to the boundary line Lx, and further draw a second horizontal line Ly2 toward the cultivated land area G2. Furthermore, the unit may be configured to stretch or shrink the size of the field image data before analysis so that the relationship between the length of the horizontal line Ly2 and the working width of the implement W is a predetermined constant ratio.

[0071] Figure 15 is a rear perspective view of a work vehicle according to another embodiment. In a tractor equipped with an AI camera for monitoring the rotary tiller, the camera may monitor the vertical movement of the rotary tiller, and if hunting (vertical movement exceeding a certain number of times and amount of movement within a certain period of time) is detected, the sensor sensitivity for tillage depth control may be automatically reduced. The AI ​​camera for monitoring the rotary tiller is installed inside the tractor's rear fender, below the windshield washer fluid tank. As shown in Figure 15, the AI ​​camera is mounted on the washer tank stay. The camera monitors the vertical movement of the rotary tiller, and if hunting (vertical movement exceeding a certain number of times and amount of movement within a certain period of time) is detected, the sensor sensitivity for tillage depth control is automatically reduced. Subsequently, if no hunting occurs for a certain period of time, the sensitivity can be returned to its original level. In a tractor equipped with an AI camera for monitoring the front and an AI camera for monitoring the rear, the tillage width is estimated from the size of the rotary tiller using the rear camera image, and a hypothetical area of ​​several meters in width and length is set in front, and that area in the direction of travel is monitored. The AI ​​camera can be configured to issue an alarm if it detects a stone of a certain size or larger within the area.

[0072] The tractor has an AI camera that monitors the front and another that monitors the rear. The rear camera image is used to estimate the tilling width from the size of the rotary tiller, and a hypothetical area of ​​several meters in width and length is set up in front of the tractor, which is then monitored in the direction of travel. The AI ​​camera monitors the amount and condition of straw and rice stalks in the area, and the rear camera performs similar monitoring. The results of the front and rear monitoring are compared to determine how much the visible straw and rice stalks have decreased after tilling. If the decrease is not below a certain value, the tilling depth is automatically increased, or the PTO rotation or engine speed is increased. In a tractor equipped with an AI camera that monitors the front and an AI camera that monitors the rear, the tilling width is estimated from the size of the rotary using the rear camera imagery, and a hypothetical area of ​​several meters in length and width is assumed to exist in front of the tractor. In this configuration, the estimated tilling width is corrected from the actual tilled track (at the start of tilling, the width remains as estimated. Once tilling begins, the width is corrected from the tilled track).

[0073] In a tractor equipped with an AI camera that monitors the front and an AI camera that monitors the rear, the tilling width is estimated from the size of the rotary tiller using the rear camera image. A hypothetical area of ​​several meters in width and length is set up in front of the tractor, and this area in the direction of travel is monitored. The estimated tilling width is corrected from the actual tilled track (at the start of tilling, the width remains as estimated. Once tilling begins, the width is corrected from the tilled track). At this time, the area for monitoring impurities such as stones is set to match the wider side, and the area for monitoring plowing is set to match the narrower side. In a tractor equipped with an AI camera that monitors the rear, the tilled track is monitored by the AI ​​camera during rotary tilling, and the finished state is determined from the size and distribution of soil clumps. The operator is notified of the finished level through a meter panel display, etc. In a tractor equipped with an AI camera that monitors the front, when tilling adjacent fields, the front camera image can be used to determine the edge (boundary) of the adjacent cultivated field, and the tractor can be configured to automatically steer along that boundary. [Explanation of Symbols]

[0074] 1. Tractor (work vehicle) 2. Running vehicle 3. Vehicle frame 4 Front wheels 5 Rear wheels 6. Hood 7. Control Unit 7a Cabin Box 7r Cockpit 8. Cockpit 9 Steering wheel 10 Mission Case 11. Meter Panel 12 Lifting device 13. Imaging device 14f Windshield 14s side glass 15 (15L, 15R) Brake pedal 16 PTO shafts 18. Clutch pedal 19. Accelerator pedal 20 Obstacle detection sensors 26 Lift arm sensor 30 Positioning device 353 Engine Key Switch 100 Mobile Information Terminals 101 displays E-engine Field H Hk Cultivated land Hm uncultivated land h1 ridge G1 Uncultivated land area G2 Cultivated land area P reference point Lx boundary line Ly horizontal line W work machine

Claims

1. A work vehicle comprising a vehicle body that travels on a field, a work implement mounted on the vehicle body, a positioning device that measures the vehicle's own position, and a steering device that steers the steering wheel, wherein the steering device is controlled based on the vehicle's own position information acquired by the positioning device, thereby automatically steering the steering wheel and enabling automatic operation on a field, A camera that acquires field image data taken from the front of the aforementioned work vehicle, An uncultivated land area extraction unit for extracting uncultivated land areas from the image data of the field, and a cultivated land area extraction unit for extracting cultivated land areas from the image data of the captured image, The system comprises: a boundary line calculation unit that calculates the boundary line between uncultivated and cultivated land in a field based on the uncultivated and cultivated land areas extracted by the uncultivated land area extraction unit and the cultivated land area extraction unit; a boundary line analysis unit that analyzes the boundary line calculated by the boundary line calculation unit; and a boundary line information recording unit that records the analysis information regarding the boundary line from the boundary line analysis unit as boundary line record information. A work vehicle characterized by having a driving means that automatically steers the vehicle during autonomous driving based on the field image data and boundary line recording information acquired by the camera.

2. The boundary line recording information includes information regarding the angle and distance of the boundary line relative to the reference point, The driving means is configured to automatically steer the work vehicle so as to eliminate any deviation in the angle and distance of the boundary line by using the boundary line analysis unit to calculate the angle and distance of the boundary line relative to the reference point from the field image data acquired by the camera during automatic driving, and comparing the calculated angle and distance with the angle and distance of the boundary line relative to the reference point recorded in the boundary line recording information.

Images