Work vehicle

Abstract

To detect a distance to an obstacle over a wide range as compared with a conventional technology.SOLUTION: The control means (200) determines the breadth (L1) of the work implement (18) based on the image captured by the image capturing device (51), creates the composite image (54) in which the imaginary line (53) corresponding to the breadth (L1) of the work implement (18) is combined with the captured image, and detects the obstacle (56) present inside the breadth (L1) of the work implement (18) in the traveling direction of the work implement (18) based on the imaginary line (53); By measuring the distance to the obstacle (56), it is possible to detect the distance to the obstacle over a wide range as compared with the prior art.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a work vehicle such as a tractor or a pest control machine. [Background technology]

[0002] A technology is known in which a work vehicle equipped with a work machine such as a tractor or pest control machine is equipped with an obstacle sensor and an imaging device, and the imaging device detects an obstacle and the obstacle sensor detects the distance to the obstacle (Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-85683 Summary of the Invention [Problem to be solved by the invention]

[0004] While conventional technology can accurately measure the distance to an obstacle using an obstacle sensor, the distance measurement range is narrow and it is not possible to measure distances over long distances. Therefore, the measurement is performed only when the obstacle is close enough that it is not possible to avoid it in time, or the driver may have to steer sharply to avoid it, potentially damaging the field. Furthermore, obstacle sensors can only measure the distance in the direction in which they are facing, and in the case of a wide implement, such as a pesticide spraying boom, they may not be able to cover the entire width of the implement.

[0005] The present invention has a technical object to make it possible to detect the distance to an obstacle over a wider range than conventional techniques. [Means for solving the problem]

[0006] The above-mentioned problems of the present invention are solved by the following means. The invention described in claim 1 is a work vehicle characterized by comprising: a work implement (18) supported on a vehicle body (1a) and performing work on a field; an imaging device (51) that captures an image that includes at least a portion of the work implement (18) within an imaging range; and a control means (200) that determines the width (L1) of the work implement (18) based on the image captured by the imaging device (51), creates a composite image (54) by combining a virtual line (53) corresponding to the width (L1) of the work implement (18) with the captured image, detects an obstacle (56) that exists inside the width (L1) of the work implement (18) in the direction of travel of the work implement (18) based on the virtual line (53), and measures the distance to the obstacle (56).

[0007] The invention described in claim 2 is a work vehicle described in claim 1, characterized in that the control means (200) identifies a ridge (57) in the field based on an image captured by the imaging device (51), detects the distance between the ridge (57) and the vehicle body (1a), determines whether the vehicle body (1a) is approaching or moving away from the ridge (57) based on the temporal history of the distance between the ridge (57) and the vehicle body (1a), and corrects steering while the vehicle body (1a) is traveling based on the result of this determination.

[0008] The invention described in claim 3 is a work vehicle described in claim 1 or 2, characterized in that it is equipped with a control means (200) that varies the virtual line (53) in accordance with variations in the width (L1) of the work implement (18) and creates a composite image (54) using the varied virtual line (53). [Effects of the Invention]

[0009] According to the invention of claim 1, the control means (200) determines the width (L1) of the work machine (18) based on the image captured by the imaging device (51), creates a composite image (54) by combining a virtual line (53) corresponding to the width (L1) of the work machine (18) with the captured image, detects an obstacle (56) that exists inside the width (L1) of the work machine (18) in the direction of travel of the work machine (18) based on the virtual line (53), and measures the distance to the obstacle (56), thereby making it possible to detect the distance to an obstacle over a wider range than with conventional technology.

[0010] According to the invention of claim 2, in addition to the effects of the invention of claim 1, the control means (200) determines whether the vehicle body (1a) is approaching or moving away from the ridge (57) based on the image captured by the imaging device (51) and the time history of the distance between the ridge (57) and the vehicle body (1a), and corrects the steering while the vehicle body (1a) is traveling according to the determination result, thereby enabling the vehicle body (1a) to travel along the edge of the ridge with high accuracy.

[0011] According to the invention of claim 3, in addition to the effects of the invention of claim 1 or 2, the control means (200) varies the virtual line (53) in accordance with fluctuations in the width (L1) of the work implement (18) and creates a composite image (54) using the varied virtual line (53), thereby enabling the worker to smoothly check the width (L1) in accordance with the condition of the work implement (18). [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is an explanatory diagram of a tractor as an example of a work vehicle according to an embodiment, and is an explanatory diagram of a state in which a work implement is lowered to a height at which work can be performed. [Figure 2] FIG. 2 is an explanatory diagram of a tractor as an example of a work vehicle according to an embodiment, with the work implement in a raised position. [Figure 3] 3A and 3B are explanatory diagrams of examples of images captured by the camera of the embodiment, where FIG. 3A is an explanatory diagram of an image behind the work vehicle, and FIG. 3B is an explanatory diagram of an image in front of the work vehicle. [Figure 4]FIG. 4 is a functional block diagram of the control unit according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0013] FIG. 1 is an explanatory diagram of a tractor as an example of a work vehicle according to an embodiment, and is an explanatory diagram of a state in which a work implement is lowered to a height at which work can be performed. FIG. 2 is an explanatory diagram of a tractor as an example of a work vehicle according to an embodiment, with the work implement in a raised position. 1 and 2, a tiller tractor 1 as an example of a work vehicle of the present invention is provided with front wheels 2, 2 and rear wheels 3, 3 as an example of a traveling device at the front and rear of a traveling body (an example of a vehicle main body) 1a. An engine 4 is mounted inside a hood 6 at the front of the traveling body 1a. The rotational power of the engine 4 is appropriately reduced by a speed change device in a transmission case 5 and is transmitted to the front wheels 2, 2 and the rear wheels 3, 3. A working implement such as a tiller (an example of a working implement, a rotary tiller) 18 that tills the ground (field) behind the tractor 1 is attached to the rear of the tractor 1, and power is transmitted via a PTO shaft 9 to drive the working implement. In this specification, the left and right sides of the tractor 1 when viewed in the forward direction are referred to as the left and right sides, respectively, and the forward direction is referred to as the front side, and the backward direction is referred to as the rear side.

[0014] A cabin 7 is supported on the upper part of the traveling vehicle body 1a. Inside the cabin 7, a driver's seat 8 is disposed above the transmission case 5, and a steering wheel 10, a parking brake (not shown), and the like are disposed in front of the driver's seat 8. Also disposed in front of the driver's seat 8 are a display panel (meter panel) for a speedometer (not shown), various operating switches (not shown), and the like. Disposed below and in front of the driver's seat 8 are driving operation tools such as a brake pedal 12 and an accelerator pedal 13.

[0015] In Figure 1, a hydraulic cylinder case 14 is provided above the rear of the transmission case 5, and lift arms 15, 15 are pivotally mounted on both the left and right sides of this hydraulic cylinder case 14. Lift rods 17, 17 are interposed and connected between the lift arms 15, 15 and lower links 16, 16, and a tiller 18 is connected to the rear of the lower links 16, 16.

[0016] When hydraulic oil is supplied to the hydraulic cylinder 14a housed in the hydraulic cylinder case 14, the lift arms 15, 15 are rotated upward, and the tiller (work machine) 18 is raised via the lift rod 17, lower link 16, etc. Conversely, when the hydraulic oil in the hydraulic cylinder 14a is discharged into the transmission case 5, which also serves as a hydraulic tank, the lift arms 15, 15 are lowered. In addition, the work implements attached to the rear of the traveling body 1a, i.e., the work implements to which drive is transmitted from the PTO shaft 9, are not limited to rotary tillers for agricultural work, but also include plows, seed sowing machines, seedling transplanters, fertilizer spreaders, pesticide spreaders, etc.

[0017] 3A and 3B are explanatory diagrams of examples of images captured by the camera of the embodiment, where FIG. 3A is an explanatory diagram of an image behind the work vehicle, and FIG. 3B is an explanatory diagram of an image in front of the work vehicle. 1 and 2, a positioning device (positioning unit) 50 is installed on the upper surface of the ceiling of the cabin 7. Cameras 51 (51A, 51B) (examples of imaging devices) are installed in the upper front and upper rear of the cabin 7 to capture images of the front and rear (examples of the outside of the traveling body 1a). In FIG. 3, the cameras 51A, 51B are installed to capture images of the fields in front of and behind the traveling body 1a. The orientations (up-down and left-right orientations) of the cameras 51 are set so that at least a portion of the work implement 18 is included within the range (so-called angle of view, field of view) captured by the cameras 51. In this embodiment, the field of view of the rear (rear-facing) camera 51B is set to include a portion of the work implement 18, while the field of view of the front (forward-facing) camera 51A does not include the work implement 18. When a work implement, such as a chemical sprayer or lawnmower, is attached to the front of the traveling body 1a, the orientation of the camera 51A is set so that a portion of the work implement 18 is included within the field of view of the front camera 51A. Furthermore, in this embodiment, the camera 51 is set so that the photographing range (field of view) of the camera 51 includes part of the work machine 18 and also part of the traveling vehicle body 1a.

[0018] In addition, by attaching a lens such as a wide-angle lens or a fisheye lens to the camera 51, it is possible to adjust and change the width (horizontal and / or vertical width) of the shooting range (field of view) to any desired width. Furthermore, the number of cameras 51 can be two, one at the front and one at the back, but it is preferable to install cameras facing in each direction so that they can capture images of the front, rear, and sides, etc., thereby eliminating blind spots. In particular, when the vehicle is traveling automatically, the camera 51 capturing images of the sides is preferably installed facing the right side (outside of the turning direction) of the traveling vehicle body 1a if the turning direction during traveling is set counterclockwise, in order to detect ridges, and is preferably installed on the left side if the turning direction is clockwise. In addition, in the case of a model in which the work implement 18 is mounted in only one location on the traveling vehicle body 1a, it is possible to use only one camera 51. It is also possible to install multiple cameras facing the same direction and measure the distance to an object (such as an obstacle or a ridge) using stereo vision with the multiple cameras.

[0019] (Explanation of the control unit) FIG. 4 is a functional block diagram of the control unit according to the embodiment. In the block diagram of FIG. 4, elements that are not related to the description of the embodiment of the present invention are not shown or described. The tractor 1 of the embodiment is configured to be able to send and receive information via a communication line 160 to and from a distribution server 161 as an example of an information processing device and a tablet terminal 162 as an example of a terminal. The communication line 160 is preferably a wireless communication line such as a mobile phone line or a wireless LAN line, but wired communication can also be used.

[0020] The distribution server 161 stores information about the map of the field, information about the travel route when work is performed (work start position, work end position, swing start position, swing end position, etc.), and information about the work implement that activates / stops the work implement 18 (raised position / lowered position of the work implement 18, etc.). The distribution server 161 can distribute work information including the map information of the field to the tractor 1.

[0021] By incorporating (installing) a dedicated application program, the tablet terminal 162 can display information about the field, the tractor 1's driving route, and the tractor 1's status (work status, remaining fuel, remaining materials, etc.), as well as operate the tractor 1 and give instructions to start / stop autonomous driving. The tablet terminal 162 of the embodiment is configured to be able to input an automatic work mode in which the tractor 1 travels autonomously while working along a travel route, and a reverse tillage mode in which the tiller 18 is rotated in reverse while the tractor 1 is reversed at the edge of a field, etc., to till the field.

[0022] (Explanation of the control unit of the work vehicle) 4, the tractor 1 of the embodiment has a vehicle ECU 200 as an example of a control means (control unit) that controls each function. The vehicle ECU 200 of the embodiment is configured with a small information processing device, a so-called microcomputer. Therefore, the vehicle ECU 200 can realize various functions by executing programs stored in a ROM or the like.

[0023] The vehicle ECU 200 outputs signals to controlled elements (signal output elements) such as a communication unit 201 as an example of a communication means, an engine 4 as an example of a driving source for driving, a vehicle brake 203 as an example of a braking device, a steering motor 204 as an example of a steering device, a forward / reverse clutch 205 as an example of a forward / reverse switching device, a hydraulic cylinder 14a as an example of a lifting device, a PTO clutch 206, and a display panel in the cabin 7.

[0024] The communication unit 201 communicates (transmits and receives information) with a distribution server 161 and a tablet terminal 162 via a communication line 160. The engine 4 generates power to drive the wheels 2 and 3 and power to drive the PTO shaft 9 . The vehicle brake 203 brakes (applies the brakes) the tractor 1 while it is traveling. The steering motor 204 rotates the steering wheel 10 to change the steering angle and steer the tractor 1. The forward / reverse clutch 205 switches the tractor 1 between forward and reverse. The hydraulic cylinder 14a raises and lowers the work machine 18. The PTO clutch 206 switches between transmitting and not transmitting drive to the work machine 18 .

[0025] In addition, signals are input to the vehicle ECU 200 from signal input elements such as a communication unit 201, a steering angle sensor SN1 as an example of a steering angle detection member, a positioning device 50, a camera 51 (51A, 51B), an elevation sensor SN2, and an input switch (not shown) in the cabin 7. The steering angle sensor SN1 detects the steering angle, which is the amount of operation of a steering wheel 10, which is an example of a steering member.

[0026] The positioning device 50 has a GNSS receiver 50a and an inertial measurement unit (IMU) 50b, and measures the current position of the tractor 1. The GNSS receiver 50a receives positioning signals from artificial satellites using the GNSS (Global Navigation Satellite System) system to measure the current position of the tractor 1. The IMU (Inertial Measurement Unit) 50b measures acceleration and angular velocity to measure the attitude (left and right tilt and front and back tilt) of the tractor 1. Therefore, by correcting the measurement results of the GNSS receiver 50a using the IMU 50b, the positioning device 50 can measure the current position with higher accuracy than when measuring the current position using only the GNSS system. The camera 51 captures images in front of and behind the tractor 1. The lift sensor SN2 detects the height and elevation of the work implement 18 by measuring the tilt angle of the lift arm 15. The accuracy of the height of the work implement 18 measured by the lift sensor SN2 can be improved by correcting for the effects of the height and attitude of the tractor 1 based on the measurement results of the GNSS receiver 50a and IMU 50b.

[0027] In addition, signals from sensors installed on the work implement 18 are also input to the vehicle ECU 200 depending on the type of work implement 18 (tiller, seedling planting device, chemical sprayer, sower, plow, etc.). For example, if a tiller is installed as the work implement 18, a signal from a tilling height sensor (not shown) that detects the height of the tiller (work implement) 18 during operation, i.e., the tilling depth, is input. Furthermore, if a seedling planting device is installed as the work implement 18, a signal from a planting depth sensor (not shown) that detects the height of the seedling planting device during operation, i.e., the planting depth, is input. Furthermore, the accuracy of the planting depth can be improved by correcting it using the rotation angle of a float provided on the seedling planting device.

[0028] Furthermore, when a chemical sprayer is attached as the work implement 18, a signal is input from a work implement deformation sensor SN3 that detects the deployment / storage of the boom. The boom deformation (deployment / storage) can also be detected from a stroke sensor of the actuator that deforms the boom or a pressure sensor of the actuator. It is possible to provide multiple methods for detecting deformation, but if multiple methods are provided, it is desirable to prioritize them, and by setting the priorities, processing can be performed according to the priority even if conflicting signals are input.

[0029] The vehicle ECU 200 according to the embodiment has the following functional means (functional modules). The work information acquisition means 251 of the vehicle ECU 200 acquires work information performed by the tractor 1. The work information acquisition means 251 acquires work information including map information, route information, work implement information, etc. from the distribution server 161. When acquiring the work information, if there is any associated registered information, it is also acquired. Therefore, if information such as turning operation information, obstacle avoidance operation information, unevenness response information, steering start position and steering angle, braking amount (braking pressure), braking position, and forward / reverse switch position is registered in association with information on the travel route, etc., this information is also acquired.

[0030] The positioning means 252 measures (estimates) the current position of the tractor 1 from the measurement results of the positioning device 50. The imaging means 253 acquires the image captured by the camera 51 . The work implement width determination means 254 has a work implement lift determination means 254a and a work implement deformation determination means 254b, and determines the width L1 of the work implement 18 from the image captured by the camera 51. The work implement width determination means 254 in this embodiment determines (calculates, calculates) the width L1 of the work implement 18 by utilizing the fact that the length L0 of a portion of the traveling vehicle body 1a reflected in the image capture range of the camera 51 (the rear camera 51B in this embodiment) is known. Therefore, the width L1 of the work implement 18 can be calculated even if the type of work implement 18 attached (tiller, chemical sprayer, etc.) is different.

[0031] When the work implement 18 rises or falls, the width L1 fluctuates in the image captured by the camera 51. However, in the embodiment, when the work implement rise / fall discrimination means 254a detects rise / fall by the rise / fall sensor SN2, the work implement width discrimination means 254 does not recalculate the width L1 in the image captured by the camera 51. Furthermore, if the work implement 18 is a chemical sprayer, the arm for spraying chemicals, called a boom, changes shape when it is stored and deployed. Therefore, the width L1 of the work implement 18 varies both in the image captured by the camera 51 and in the actual work implement 18. In this embodiment, the work implement width determination means 254 recalculates the width L1 in the image captured by the camera 51 when the work implement deformation determination means 254b detects deformation using the work implement deformation sensor SN3. Note that the detection of deformation of the work implement 18 is not limited to using a sensor; it is also possible to perform image analysis on the image captured by the camera 51 and determine the deformation of the work implement 18 using AI (artificial intelligence) that has previously learned about deformations and lifting and lowering from images of various types of the work implement 18.

[0032] Furthermore, the calculated width L1 of the work implement 18 is compared with information from the distribution server 161 to estimate the type of work implement 18, and the type of work implement 18 being used can be notified to the worker or the administrator of the distribution server 161, or used to determine whether it is lifted or deformed. When multiple work implements 18 are to be estimated, it is also possible to calculate the estimation probability for each type of work implement based on, for example, whether it is attached to the front or rear of the traveling vehicle body 1a, and the shape and appearance other than the width L1, and to create a list and notify the worker or administrator. The worker or administrator can determine the correct work implement from the list, assign a correct label, and use the list as training data for AI learning.

[0033] The virtual line generating means 255 generates a virtual line 53 that corresponds to the width L1 of the work implement. That is, virtual lines 53 that extend from the right and left ends of the work implement 18 along the traveling direction are generated in accordance with the width L1 of the work implement 18. Therefore, in this embodiment, a pair of virtual lines 53 are generated on the left and right sides of the work implement 18. Note that if the work implement width determination means 254 recalculates the width L1 in conjunction with deformation of the work implement 18, the virtual line generating means 255 regenerates the virtual line 53 that corresponds to the recalculated width L1. It is preferable that when the type of work implement 18 is specified, the virtual line 53 is displayed as a solid line, for example, and when the type of work implement 18 is not specified, the virtual line 53 is displayed in a different manner, such as as a dashed line or a flashing line, to notify the worker and warn the worker not to rely too much on the display of the virtual line 53. Furthermore, when the type of work implement 18 is specified and the planar shape (plan view) of the work implement 18 can be acquired from the distribution server 161 or the like, it is preferable to additionally display a plan view (bird's-eye view) of the tractor 1 and work implement 18 in addition to the composite image 54. The plan view is preferably displayed as an icon of the travel route in the field, which can accurately display the shape of the vehicle and easily encourage safe driving.

[0034] The composite image creation means 256 generates a composite image (composite image) 54 by combining an image of the virtual line 53 with the image captured by the camera 51. The composite image creation means 256 of the embodiment generates a composite image 54' by combining the virtual line 53 with not only the image captured by the rear camera 51B (composite image 54), but also the image captured by the front camera 51A. The generated composite images 54, 54' are displayed as guide images (navigation images) on a display panel inside the cabin 7.

[0035] The obstacle detection means 257 detects obstacles 56 based on images captured by the camera 51. In this embodiment, the obstacle detection means 257 detects obstacles 56 that exist inside the width of the work implement 18 in the traveling direction of the work implement 18 (the traveling direction (forward or backward) of the traveling body 1a), i.e., in the area included between the virtual lines 53. In this embodiment, the images from the camera 51 are analyzed and identified using AI that has been trained in advance using a large number of images that include obstacles and a large number of images that do not include the obstacles 56, to detect the obstacles 56.

[0036] The obstacle distance calculation means 258 calculates the distance from the work implement 18 to the obstacle 56 detected by the obstacle detection means 257. The obstacle distance calculation means 258 in this embodiment calculates the distance to the obstacle 56 by using AI that has been trained in advance to analyze and identify images based on the image captured by the camera 51. The detected obstacle 56 and the distance to the obstacle 56 are displayed on the display panel in the cabin 7 or on the display screen of the tablet terminal 162 and notified to the worker. This enables the worker to make decisions and perform tasks such as operating to avoid the obstacle 56 or stopping the tractor 1 and removing the obstacle 56, and detection of the obstacle 56 and measurement of the distance can provide information for decision-making (assistance and support for decision-making) for the worker's decisions and tasks.

[0037] The ridge edge detection means 259 detects the ridges 57 in the field based on the image captured by the camera 51. The ridges 57 are detected based on the image captured by the camera 51 by image analysis and image recognition using AI that has been trained in advance. The furrow distance calculation means 260 calculates the distance (furrow distance) from the work implement 18 to the ridge 57 detected by the ridge edge detection means 259. Specifically, for the detected ridge 57, the distance between the ridge 57 and the nearest one of the two ends of the work implement 18 is calculated as the furrow distance. The furrow distance is detected by image analysis and image recognition using AI that has been trained in advance based on the image captured by the camera 51.

[0038] The traveling tendency determination means 261 determines whether the traveling vehicle body (vehicle main body) 1a is approaching or moving away from the ridge 57 based on the time history (profile data) of the distance between the ridge 57 and the end of the work implement 18. That is, the furrow distance is acquired at predetermined time intervals (sampling intervals) and stored as history information, and if the difference in furrow distance becomes smaller over time, the traveling tendency is determined to be "approaching" the ridge 57, and if the difference in furrow distance becomes larger, the traveling tendency is determined to be "moving away" from the ridge 57. Note that if the difference in furrow distance falls within a predetermined range, the distance from the ridge 57 is determined to be "constant," which is the traveling tendency. In the embodiment, the furrow distance is measured as the distance between the work implement 18 and the ridge 57, but this is not limited to this. It is also possible to install cameras 51 at the front and rear of the traveling body 1a to measure the furrow distance in front of the traveling body 1a and the furrow distance behind it, and determine the inclination and tendency of the direction of travel from the two furrow distances, or determine the tendency from the average history of the two furrow distances.

[0039] The ridge-edge steering correction means 262 corrects the steering of the traveling vehicle body 1a during traveling according to the traveling tendency when steering at the ridge (when the ridge 57 is detected) based on information about the traveling route. The positioning means 252 using the GNSS method may generate measurement errors of several meters, and may not be able to measure errors of several tens of centimeters. In particular, errors are likely to become large when GNSS radio wave reception is intermittent due to bad weather or obstacles. In the embodiment, when providing guidance (navigation) along the traveling route during manual traveling or when the traveling tendency is "constant" during automatic traveling, the ridge-edge steering correction means 262 provides traveling and steering guidance or automatic steering of the traveling vehicle body 1a based on the positioning results of the positioning means 252, and does not correct steering.

[0040] On the other hand, when the driving tendency is "approaching," although it cannot be measured by the positioning means 252 with respect to the planned driving route, there is a possibility that the vehicle is approaching the ridge 57, and therefore the ridge-edge steering correction means 262 of the embodiment corrects the steering in a direction away from the ridge 57 relative to the driving direction along the driving route. This makes it possible to prevent the traveling vehicle body 1a from contacting the ridge 57. Regarding the steering correction, during manual driving, a guidance (navigation) display is displayed to steer in a direction away from the ridge 57, urging the operator to correct the steering, and during automatic driving, automatic steering control is performed via the steering motor 204 in a direction away from the ridge 57. In addition, if the driving tendency is "moving away," the positioning means 252 cannot measure the planned driving route, but there is a possibility that the vehicle is moving away from the ridge 57, so steering is corrected in a direction closer to the ridge 57 relative to the driving direction along the driving route.

[0041] Therefore, in the tractor 1 of the embodiment, steering correction is performed according to the driving tendency, and the error between the actual driving path of the traveling body 1a and the planned driving path, which cannot be measured by the positioning means 252, can be reduced based on the driving tendency measured from the image of the camera 51. Therefore, the traveling body 1a can travel along the edge of rice paddies with high accuracy. In the conventional technology in which positioning is performed only by the positioning means 252, automatic travel along the edge of a field is difficult because, due to measurement accuracy, there is a risk of the tractor running over or damaging the edge of a field, 57. In contrast, the tractor 1 of the embodiment, which can travel along the edge of a field with high accuracy by also using images from the camera 51, is capable of automatic travel along the edge of a field, and work efficiency can be improved.

[0042] In the tractor 1 of the embodiment having the above configuration, a virtual line 53 is generated based on an image captured by the camera 51, and the distance to an obstacle 56 inside the virtual line 53 is measured. Therefore, the distance to the obstacle 56 is measured based on the image captured by the camera 51, and measurement is possible over a longer distance than when an obstacle sensor is used. Furthermore, the field of view of the camera 51 is generally wider than the range of the obstacle sensor, and can be further widened by using a wide-angle lens, etc. Therefore, in the embodiment, the obstacle 56 can be detected over a wider range and at a longer distance than in conventional technology.

[0043] Furthermore, in the tractor 1 of the embodiment, a virtual line 53 corresponding to the width L1 of the work implement 18 is displayed as a composite image 54, allowing the operator to perform work while checking the composite image 54. Therefore, when using a tiller attached to the rear of the tractor 1 as the work implement 18 and performing tilling work while moving backward, the operator can perform the work while checking the composite image 54 and confirming obstacles 56 and ridges 57 in the direction of travel (rear). Furthermore, when using a chemical sprayer attached to the front of the tractor 1 as the work implement 18 and performing chemical spraying work while moving forward, the operator can perform the work while checking the composite image 54 and confirming obstacles 56 and ridges 57 in the direction of travel (forward). Furthermore, when the work implement 18 is deformed, the virtual line 53 is updated in accordance with the deformation, and the creation of the composite image 54, the detection of the obstacle 56, and the calculation of the distance are also performed again. Therefore, the appropriate virtual line 53, composite image 54, etc. can be updated as needed depending on the work, facilitating the operator's confirmation and work.

[0044] (Example of change) In the above embodiment, the tractor 1 is used as an example of a work vehicle, but the present invention is not limited to this. The present invention can be applied to any work vehicle having a work implement, such as a rice planter, a lawn mower, a seedling transplanter, a vegetable harvester, or a combine harvester. For example, in a seedling transplanter for growing potato seedlings or the like, it is possible to detect obstacles and ridges based on images captured by a camera. At this time, images captured by the camera showing the seedling supply status can be analyzed and identified, and if the supply is not completed in time, the vehicle speed can be reduced to prevent missing plants. When capturing images of the supply status with a camera, it is preferable to attach a stay to the side of the side clutch case to secure the camera, as this shortens the distance to the seedling transplanting section and the seedling transplanting section and the side clutch case rise and fall in tandem, thereby maintaining the relative positions of the camera and the seedling transplanting section and facilitating image identification.

[0045] In the combine harvester, it is also possible to detect obstacles and ridges based on images taken by the camera. It is also possible to perform image analysis and image recognition on images of crops in the field taken by the camera 51, and change the settings of the work machine (harvesting machine) depending on the state of lodging of the crops. Furthermore, the rice transplanter can also detect obstacles and ridges based on images captured by a camera. At this time, if lodging has occurred in a specific area of ​​the field based on work data (past data) from the previous year's combine harvest, it is possible to narrow the spacing between rice plants when transplanting in that area, reduce the amount of seedlings removed, and perform other controls. Note that if the past data from combine harvesting includes information about a disaster such as a typhoon passing before harvesting, it is also possible to avoid using the data on lodging. [Explanation of symbols]

[0046] 1...Work vehicle, 1a...Vehicle body, 18...Work equipment, 51...imaging device, 53...imaginary line, 54…Synthetic image, 56...obstacles, 57...Round, 200...control means, L1...Width of the implement.

Claims

1. a work implement (18) supported on the vehicle body (1a) and used to perform work on a farm field; an imaging device (51) that captures an image including at least a part of the work machine (18) within an imaging range; a control means (200) that determines the width (L1) of the work machine (18) based on an image captured by the imaging device (51), creates a composite image (54) by combining a virtual line (53) corresponding to the width (L1) of the work machine (18) with the captured image, detects an obstacle (56) that exists inside the width (L1) of the work machine (18) in the traveling direction of the work machine (18) based on the virtual line (53), and measures the distance to the obstacle (56); A work vehicle comprising:

2. The control means (200) identifies a ridge (57) in the field based on the image captured by the imaging device (51), detects the distance between the ridge (57) and the vehicle body (1a), determines whether the vehicle body (1a) is approaching or moving away from the ridge (57) based on the time history of the distance between the ridge (57) and the vehicle body (1a), and corrects steering while the vehicle body (1a) is traveling according to the result of the determination.

2. The work vehicle according to claim 1.

3. the control means (200) for varying the virtual line (53) in accordance with a variation in the width (L1) of the work machine (18) and creating a composite image (54) using the varied virtual line (53); 3. The work vehicle according to claim 1, further comprising:

Images