Work vehicle

Abstract

To provide a working vehicle capable of accurately performing control according to a state of a field.SOLUTION: A work vehicle according to an aspect of an embodiment includes a controller and a storage unit. The storage unit stores a first state model and a second state model, the first state model learning control data affected by disturbance when a work vehicle without a work implement actually travels in a field, the second state model learning control data affected by disturbance when a work vehicle with a work implement actually travels in the field, the second state model outputting second control data when the work vehicle travels next time or later. The controller generates final control data by weighting and integrating the first control data output from the first state model and the second control data output from the second state model, and controls the work vehicle according to the final control data. The weighting is changed according to the state of the work vehicle during travel.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a work vehicle. [Background technology]

[0002] Conventionally, there has been known a control device that is provided with a behavior detection sensor that detects behavior and controls a work vehicle based on the detected value of the behavior detection sensor. For example, Patent Document 1 discloses a technology in which a behavior detection sensor detects the condition of a field, such as unevenness, and controls the work vehicle according to the detected field condition (see, for example, Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2010-227009 Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the conventional technology, the detection value of the behavior detection sensor may deviate from the actual state of the field because it is affected by engine vibration and vibration caused by accelerator work. In other words, in the conventional technology, there is a risk that control according to the state of the field cannot be performed with high accuracy.

[0005] The present invention has been made in view of the above, and has an object to provide a work vehicle that can perform control in accordance with the state of a farm field with high precision. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems and achieve the object, the work vehicle (1) has a controller (41) and a memory unit (42). The memory unit (42) stores: a first state model, which is a state model that learns control data affected by disturbances when the work vehicle (1) without a work implement (W) actually travels through a field, and outputs first control data for subsequent travels; and a second state model, which is a state model that learns control data affected by disturbances when the work vehicle (1) with a work implement (W) actually travels through a field, and outputs second control data for subsequent travels. The controller (41) generates final control data by weighting and integrating the first control data output from the first state model and the second control data output from the second state model, and controls the work vehicle (1) in accordance with the final control data. The weighting is changed depending on the state of the work vehicle (1) while traveling. [Effects of the Invention]

[0007] The work vehicle according to this embodiment can perform control according to the state of the field with high precision. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 1 is a schematic left side view of a work vehicle according to an embodiment. [Figure 2] FIG. 2 is a block diagram showing an example of a control system of a tractor. [Figure 3] FIG. 3 is a diagram illustrating an example of state model information. [Figure 4] FIG. 4 is a diagram illustrating driving control based on a state model. [Figure 5] FIG. 5 is a diagram illustrating driving control based on a state model. [Figure 6] FIG. 6 is a diagram illustrating driving control based on a state model. DETAILED DESCRIPTION OF THE INVENTION

[0009] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the work vehicle disclosed in the present application will be described in detail with reference to the accompanying drawings. However, the present invention is not limited to the embodiments described below.

[0010] First Embodiment First, the overall configuration of a work vehicle according to a first embodiment will be described with reference to Figure 1. Figure 1 is a schematic left side view of the work vehicle according to this embodiment. Note that this embodiment will be described using a tractor 1 as an example of the work vehicle.

[0011] The tractor 1, which is a work vehicle, is an agricultural tractor that travels by itself to perform work in fields, etc. In addition to carrying out predetermined tasks while being driven by an operator and traveling within a field, the tractor 1 also performs predetermined tasks while traveling automatically within the field, with each part controlled by a control system centered on a control device 40 (see FIG. 2), which will be described later.

[0012] In this embodiment, the forward / rearward direction is the direction of travel of the tractor 1 when traveling straight, with the forward side of the traveling direction being defined as "front" and the rear side being defined as "rear." The traveling direction of the tractor 1 is the direction from the operator's seat 8 (described later) toward the steering wheel 9 when traveling straight.

[0013] The left-right direction is a direction that is horizontally perpendicular to the front-rear direction. Hereinafter, left and right are defined as facing the "front." In other words, when the operator of the tractor 1 is seated in the operator's seat 8 and facing forward, the left-hand side is the "left" and the right-hand side is the "right."

[0014] The up-down direction is a direction parallel to the vertical direction. The front-rear direction, left-right direction, and up-down direction are perpendicular to each other. Note that each direction is defined for the convenience of explanation, and the present invention is not limited to these directions. In addition, in this embodiment, the tractor 1 may be referred to as the "machine body."

[0015] As shown in Fig. 1, the tractor 1 includes a traveling body 2 and a work implement W. The traveling body 2 includes a body frame 3, front wheels 4, rear wheels 5, a bonnet 6, an engine E, a control unit 7, and a transmission case 10. The body frame 3 is the main frame of the traveling body 2.

[0016] The front wheels 4 are a pair of left and right wheels, and are primarily used for steering (steered wheels). The rear wheels 5 are a pair of left and right wheels, and are primarily used for driving (drive wheels). The tractor 1 may be configured to be switchable between two-wheel drive (2WD) in which the rear wheels 5 are driven, and four-wheel drive (4WD) in which both the front wheels 4 and the rear wheels 5 are driven. In this case, both the front wheels 4 and the rear wheels 5 are drive wheels. The traveling body 2 may be equipped with a crawler device instead of wheels (front wheels 4 and rear wheels 5). In this case, the traveling crawler is the drive wheel.

[0017] The hood 6 is provided at the front of the traveling vehicle body 2 so as to be able to be opened and closed freely. The hood 6 can be rotated (opened and closed) in the vertical direction with the rear part as the rotation center. When closed, the hood 6 covers the engine E mounted on the vehicle body frame 3. The engine E is the driving source of the tractor 1 and is a heat engine such as a diesel engine or a gasoline engine.

[0018] The control unit 7 is provided on top of the traveling vehicle body 2 and includes a driver's seat 8, a steering wheel 9, etc. The control unit 7 may be formed by being covered by a cabin 7a provided on top of the traveling vehicle body 2. The driver's seat 8 is a seat for the driver. The steering wheel 9 is operated by the driver when steering the front wheels 4, which are steered wheels. The control unit 7 includes a display unit (meter panel) in front of the steering wheel 9 that displays various information.

[0019] The control unit 7 also includes various operation levers such as a forward / reverse lever, an accelerator lever, a main speed change lever, and an auxiliary speed change lever, as well as various operation pedals such as an accelerator pedal, a brake pedal, and a clutch pedal.

[0020] The transmission case 10 houses a transmission (speed change mechanism). The transmission appropriately reduces the speed of the power (rotational power) transmitted from the engine E and transmits it to the rear wheels 5, which are drive wheels, and a PTO (Power Take-off) shaft.

[0021] A work implement W that performs work in the field is attached to the rear of the traveling body 2, and a PTO shaft that transmits power to drive the work implement W protrudes rearward from the transmission case 10. The PTO shaft transmits rotational power that has been appropriately reduced by the transmission to the work implement W attached to at least the rear of the traveling body 2.

[0022] In addition, a lifting device 12 that raises and lowers the work implement W is provided at the rear of the traveling body 2. The lifting device 12 raises the work implement W to move it to a non-working position. The non-working position is a position where the work implement W is raised when, for example, the traveling body 2 moves backward or turns. In addition, the lifting device 12 lowers the work implement W to move it to a ground work position. The lifting device 12 includes a hydraulic lifting cylinder 121, a lift arm 122, a lift rod 123, a lower link 124, and a top link 125.

[0023] When hydraulic oil is supplied to the lift cylinder 121, the lift arm 122 rotates around the axis AX serving as the rotation fulcrum to raise the work implement W, and when hydraulic oil is discharged from the lift cylinder 121, the lift arm 122 rotates around the axis AX to lower the work implement W. A lift arm sensor 26 that detects the rotation angle of the lift arm 122 is provided at the base of the lift arm 122 (near the axis AX). The height of the work implement W is calculated based on the detection result of the lift arm sensor 26 and the work implement W.

[0024] Furthermore, the lift arm 122 is connected to the lower link 124 via the lift rod 123. In this manner, the lifting device 12 connects the work machine W to the traveling body 2 via the lower link 124 and the top link 125 so that the work machine W can be raised and lowered.

[0025] The work implement W is a machine that performs work in a farm field. In the example shown in Fig. 1, the work implement W is a rotary tiller that performs tilling work in a farm field. The rotary tiller tills the farm field by rotating the tiller tines 61 using power transmitted from the PTO shaft.

[0026] The tractor 1 also includes a control device 40 (see FIG. 2). The control device 40 controls the engine E and also controls the traveling speed of the traveling vehicle body 2. The control device 40 also controls the work implement W.

[0027] The tractor 1 also includes a satellite positioning device 30. The satellite positioning device 30 is provided on top of the traveling vehicle body 2, and measures the vehicle position (latitude and longitude), which is the position of the traveling vehicle body 2. The satellite positioning device 30 is, for example, a GNSS (Global Navigation Satellite System), and can receive radio waves from a navigation satellite S orbiting in the sky to perform positioning and timing.

[0028] Furthermore, in the present disclosure, the work machine W is also provided with a satellite positioning device 30W. The satellite positioning device 30W is provided on top of the work machine W and measures the work machine position (latitude and longitude), which is the position of the work machine W. The satellite positioning device 30W is, for example, a GNSS, and can receive radio waves from a navigation satellite S orbiting in the sky to perform positioning and timing. Note that the work machine position does not necessarily have to be determined by the satellite positioning device 30W, but may also be estimated from the position information of the satellite positioning device 30 attached to the traveling vehicle body 2.

[0029] The tractor 1 also includes a position detection device 20. The position detection device 20 includes a ridge position detection sensor 21. The ridge position detection sensor 21 is, for example, a LiDAR (Light Detection and Ranging, Laser Imaging Detection and Ranging). The ridge position detection sensor 21 is disposed at the front of the traveling body 2, for example, attached to a sensor mounting stay 13 provided at the front, and can detect ridges that exist in front of the traveling body 2.

[0030] Another method is to calculate the distance to a target location (for example, a telegraph pole or public road registered as a reference location) using image analysis from images taken with an imaging device, and then calculate the relative relationship between the distance to the ridge and the vehicle's position, or to calculate the distance directly using image analysis.

[0031] The ridge position detection sensor 21 emits laser light and detects the light reflected from the ridge. The ridge position detection sensor 21 can detect the distance and direction to the ridge by measuring the time it takes to receive the light reflected from the ridge.

[0032] Next, a control system of the tractor 1 centered on the control device 40 will be described with reference to Fig. 2. Fig. 2 is a block diagram showing an example of the control system of the tractor 1. As shown in Fig. 2, the control device 40 includes a controller 41 and a memory unit 42. The controller 41 includes an engine ECU (Electronic Control Unit), a travel system ECU, and a work implement lifting system ECU. The engine ECU controls the rotation speed of the engine E. The travel system ECU controls the rotation of the drive wheels to control the travel speed of the travel vehicle body 2 (see Fig. 1). The work implement lifting system ECU controls the lifting device 12 to control the lifting and lowering of the work implement W.

[0033] The control device 40 is capable of controlling each part electronically, and is equipped with a controller 41 having a CPU (Central Processing Unit) and a memory unit 42 in which necessary data such as various programs and the planned driving route of the traveling vehicle body 2 set in advance for each field is stored.

[0034] As shown in Figure 2, various sensors are connected to the control device 40, such as a global navigation satellite system (GNSS) 30, an engine rotation sensor 23, a vehicle speed sensor 24, a steering angle sensor 25, a position detection device 20 (ridge position detection sensor 21), a lift arm sensor 26, and an inertial measurement unit 60. The engine rotation sensor 23 detects the rotation speed of the engine E. The vehicle speed sensor 24 detects the traveling speed (vehicle speed) of the traveling vehicle body 2 (see Figure 1). The steering angle sensor 25 detects the steering angle of the front wheels 4 (see Figure 1), which are steered wheels. The steering angle sensor 25 detects the turning of the vehicle body.

[0035] The inertial measurement unit 60 is an IMU (Inertial Measurement Unit) that is attached to the traveling vehicle body 2 and measures the inclination of the traveling vehicle body 2. In the present disclosure, an inertial measurement unit 60W that measures the inclination of the work machine W is also attached to the work machine W.

[0036] The control device 40 receives inputs of position information of the traveling vehicle body 2 and work implement W in a field or the like from the satellite positioning devices 30, 30W, the rotation speed of the engine E from the engine rotation sensor 23, the traveling speed of the traveling vehicle body 2 from the vehicle speed sensor 24, the turning angle of the front wheels 4 from the turning angle sensor 25, the ridge position from the ridge position detection sensor 21, and the height of the work implement W from the lift arm sensor 26. When the tractor 1 is caused to travel autonomously, the control device 40 automatically steers the steering wheel 9 by controlling the steering cylinder connected to the steering wheel 9 using the detection results of the turning angle sensor 25 while feeding back the turning angle of the front wheels 4.

[0037] The control device 40 also includes a controller 41 and a storage unit 42. The controller 41 is realized by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit) executing various programs stored in a storage device inside the control device 40 using a RAM or the like as a work area. The controller 41 may also be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a GPGPU (General Purpose Graphic Processing Unit). The controller 41 realizes the functions of, for example, an engine ECU, a travel system ECU, and a work implement lifting system ECU.

[0038] The storage unit 42 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 42 stores state model information 421.

[0039] In addition, the engine ECU of the controller 41 is connected to the engine E, the travel system ECU is connected to the steering device 51, the transmission 52, the braking device 53, etc., and the work implement lifting system ECU is connected to the lifting device 12.

[0040] Of these, the work machine lifting system ECU outputs a work machine lifting signal to the lifting device 12. The lifting device 12 drives the work machine W to lift up and down based on the work machine lifting signal output from the work machine lifting system ECU.

[0041] The control device 40 can also automatically drive the traveling vehicle body 2 while causing the work implement W to perform work. For example, the control device 40 stores a field map for each field in advance in the storage unit 42. The control device 40 also determines a planned travel route R (see FIG. 4) in advance on the field map according to the work to be performed by the work implement W, converts the planned travel route R into data, and stores the data in the storage unit 42. Based on the measurement results of the satellite positioning device 30, the control device 40 controls each part, such as the engine E, steering device 51, transmission 52, braking device 53, and lifting device 12, so that the tractor 1 travels along the stored planned travel route R. In this way, the tractor 1 automatically travels along the planned travel route R. The planned travel route R is set according to the shape and size of the field, the width, length, and number of ridges formed in the field, the type of crop, etc.

[0042] Furthermore, the state model information 421 stored in the storage unit 42 is information on a state model that outputs control data for the planned driving route R. The control data is data that the controller 41 uses to control the engine E, the steering device 51, the transmission 52, the braking device 53, and the lifting device 12. In other words, the controller 41 realizes automatic driving by controlling the engine E, the steering device 51, the transmission 52, the braking device 53, and the lifting device 12 in accordance with the control data output from the state model of the state model information 421.

[0043] Also, in FIG. 2, the control device 40 receives sensor data from the satellite positioning device 30, engine rotation sensor 23, vehicle speed sensor 24, steering angle sensor 25, position detection device 20, lift arm sensor 26, inertial measurement unit 60, etc. as an input group, and receives control data for controlling the steering device 51, transmission 52, braking device 53, and lifting device 12 as an output group. In this input / output, the memory unit 42 that stores information has the function of searching, collating, and synthesizing the sensor data, and in the process of generating state model information 421, artificial intelligence (neural network) may be used to perform calculations that also have a learning function.

[0044] Fig. 3 is a diagram showing an example of the state model information 421. As shown in Fig. 3, the state model information 421 includes items such as "model ID", "user", "work machine", and "control data".

[0045] "Model ID" is identification information that identifies the state model. "User" is identification information that identifies the user (i.e., the worker), and for example, a name or a user ID issued for each user is input. "Work equipment" is information that indicates the type of work equipment W attached to the tractor 1. "Control data" is control data used when traveling along the planned travel route R. Note that while Figure 3 shows a simple string of characters such as "control data #1," in reality, control values ​​for the engine E, steering device 51, transmission device 52, braking device 53, and lifting device 12 are input. Note that an example of control using control data will be described later with reference to Figure 4.

[0046] As shown in FIG. 3, the state model in the state model information 421 is generated for each user and each work machine (including a case where no work machine is present). Specifically, the controller 41 generates the state model by learning control data based on operation information when the user manually drives the tractor 1 to travel along a planned travel route in a field. In other words, when the tractor 1 deviates from the planned travel route R due to the influence of disturbances such as unevenness or slope of the field, the controller 41 generates the state model by learning control data based on operation information performed by the user to manually return the tractor 1 to the planned travel route R. That is, the state model is a model that learns control data affected by disturbances when the tractor 1 actually travels in a field. In other words, the state model is a model that is generated by having artificial intelligence (such as a neural network) learn control data affected by disturbances when the tractor 1 actually travels in a field.

[0047] Next, travel control based on a state model will be described using Figs. 4 to 6. Figs. 4 to 6 are diagrams for explaining travel control based on a state model. Fig. 4 shows an example in which a tractor 1 without a work implement W (i.e., only the traveling body 2) travels along a planned travel route R in a field according to control data output by a state model. Fig. 5 shows an example in which a tractor 1 with a work implement W (i.e., the traveling body 2 and the work implement W) travels along a planned travel route R in a field according to control data output by a state model. Fig. 6 shows an example in which two tractors 1a and 1b with work implement W work together to travel along a planned travel route R in a field according to control data output by a state model. Figs. 4 to 6 also show an example in which the tractors 1, 1a, and 1b travel along straight routes L1 to L4, which are the planned travel route R. Each of Figs. 4 to 6 is set in the same field and has the same planned travel route R. That is, in FIGS. 4 to 6, the influence of disturbances such as uneven shapes and inclinations is also the same.

[0048] Before traveling along the planned travel route R, the worker first inputs information about the user and the presence or absence of a work machine W, and the type of work machine W, to the control device 40. The control device 40 reads out a state model that matches the input information from the state model information 421, and inputs the input information into the state model to obtain control data. The control device 40 then automatically travels along the planned travel route R in accordance with the obtained control data.

[0049] For example, in FIGS. 4 to 6, control points (P1, P51, etc.) are set on the planned driving route R in the acquired control data. A control point indicates a point location where specified control is performed. FIGS. 4 to 6 indicate control points where steering control is performed with a specified operation amount at the point location. For example, control point P1 in FIG. 4 is set to "right 1." This means that at the point location, steering control is performed to turn the steering angle to the right by a predetermined angle x 1. Furthermore, control point P2's "left 2" means that steering control is performed to turn the steering angle to the left by a predetermined angle x 2. In other words, control point P1 has an uneven shape, and this uneven shape causes the aircraft to turn left, so by operating the steering wheel to turn the steering angle to the right, the aircraft does not deviate from the planned driving route R.

[0050] Here, differences in control data depending on whether or not a work implement W is present will be explained using Figures 4 and 5. Control point P1 in Figure 4 and control point P51 in Figure 5 are at the same position, and the operation amount (right 5) when the work implement W is present as shown in Figure 5 is greater than the operation amount (right 1) when the work implement W is not present as shown in Figure 4. This is because when the work implement W is present, the work implement W is susceptible to external disturbances such as unevenness, and if the work implement W sways significantly due to an uneven shape, the traveling vehicle body 2 will also be affected and sway significantly (turn significantly to the left).

[0051] Therefore, in order to travel accurately along the planned travel route R, it is preferable to control the tractor 1 in accordance with the control data shown in Fig. 5 (for example, "Right 5" at control point P51). On the other hand, the larger the operation amount in the control data, the worse the fuel efficiency of the tractor 1 becomes, and in the case of plowing, there are adverse effects such as roughening of the field due to small wheel movements.

[0052] Therefore, in the present disclosure, the control data shown in Fig. 4, i.e., the first control data output from the first state model, and the control data shown in Fig. 5, i.e., the second control data output from the second state model, are weighted and integrated to generate final control data, and the tractor 1 is controlled according to the final control data. Specifically, the final control data is calculated as follows: final control data = first control data × α + second control data × β (α + β = 1).

[0053] In the present disclosure, the weightings (α and β) for the first control data and the second control data can be changed depending on the state of the tractor 1. For example, if the work implement W is a tiller, the controller 41 sets the weighting α for the first control data to be greater than the weighting β for the second control data because the field itself will be tilled even if the work implement W deviates slightly from the planned travel route R. This reduces the amount of operation in the final control data, thereby reducing adverse effects such as a decrease in fuel efficiency of the tractor 1 and field damage. Furthermore, if the tractor 1 deviates significantly from the planned travel route R during work, the controller 41 sets the weighting β for the second control data to be greater than the weighting α for the first control data. This makes it possible to correct the trajectory of the tractor 1 that has deviated significantly from the planned travel route R.

[0054] Furthermore, the controller 41 changes the weightings α and β, for example, when traveling straight and when turning. For example, the controller 41 makes the weighting α greater than β when traveling straight, and makes the weighting β greater than α when turning. The controller 41 also changes the weightings α and β when working and when moving between fields. For example, the controller 41 makes the weighting β greater than α when working, and makes the weighting α greater than β when moving between fields. The controller 41 also makes α greater than β when the work implement W is raised, and makes the weighting β greater than α when the work implement W is lowered.

[0055] Furthermore, the controller 41 detects abnormalities in the traveling body 2 and the work implement W of the tractor 1 based on sensor information and changes the weights α and β according to the detection results. For example, the controller 41 detects an abnormality when the tilt detected by the inertial measurement units 60, 60W is equal to or greater than a threshold. Then, when no abnormality is detected in either the traveling body 2 or the work implement W, the controller 41 weights the second control data more heavily than the first control data. That is, the weight β is made larger than α. Furthermore, when an abnormality is detected in at least one of the traveling body 2 and the work implement W, the controller 41 weights the first control data more heavily than the second control data. That is, the weight α is made larger than β. As a result, when no abnormality is detected, the second control data is prioritized in order to correct the effect of disturbances on the work implement W. When an abnormality is detected, the first control data is prioritized and the operation amount is reduced in order to restore stable traveling.

[0056] Furthermore, when the controller 41 detects an abnormality in at least one of the traveling body 2 and the work implement W, the controller 41 may generate the first control data as final control data. That is, the weighting α is set to 1 and the weighting β is set to zero. This allows the vehicle to quickly return to stable traveling.

[0057] The controller 41 also detects the slip ratio of the tractor 1 while traveling in a field (at the start of work or during work), and if the detected slip ratio is higher than previously detected slip ratios by a threshold or more, corrects the steering angle change amount in the final control data to be smaller (reducing the operation amount). The slip ratio can be calculated from the vehicle speed calculated from changes in the position of the traveling vehicle body 2 detected by the satellite positioning device 30 and the vehicle speed calculated from a rotation speed sensor (not shown) that detects the number of rotations of the wheels. This makes it possible to avoid a situation in a field prone to slippage where turning is impossible due to a large change in the steering angle during control using the final control data. If a large change in the steering angle is desired, the controller 41 may gradually change the steering angle by reducing the amount of change per unit time.

[0058] Furthermore, the controller 41 detects the amount of deviation in the direction of the work machine W per unit time based on the detection results of the satellite positioning device 30, and if the amount of deviation is equal to or greater than a threshold, performs steering control to correct the amount of deviation. That is, for example, if the controller 41 detects a momentary deviation in the direction of the work machine W due to a disturbance at a point where there is no control point in the final control data, the controller 41 performs steering control to correct the deviation in the direction. This makes it possible to avoid a significant deviation from the planned travel route R even if the controller 41 is affected by a disturbance that did not exist when the state model was generated (for example, when stepping on an obstacle such as a large stone).

[0059] Furthermore, the controller 41 detects the inclination per unit time of the work implement W while it is traveling in the field, and if the inclination is equal to or greater than a threshold value, it controls the speed so that the traveling speed is kept constant. In other words, if the inertial measurement unit 60 detects a downward slope or an upward slope, the controller 41 controls the speed so that the speed when traveling on that slope is kept constant. For example, when traveling on a downward slope, the speed increases, so the controller 41 reduces the speed, and when traveling on an upward slope, the speed decreases, so the controller 41 increases the speed. This makes it possible to maintain a constant traveling speed (working speed) even when traveling on a slope.

[0060] If the time period during which the slope is equal to or greater than the threshold continues for a predetermined time or longer, the controller 41 controls the speed so that the actual traveling speed is maintained at the reference speed determined by the final control data. This makes it possible to maintain the traveling speed at the speed determined by the final control data when traveling on a long downward or upward slope.

[0061] Next, cooperative work by the second tractors 1a and 1b will be described with reference to FIG. 6. In FIG. 6, the first tractor 1a works on the straight path L1, then turns and works on the straight path L3, and the second tractor 1b follows the first tractor 1a, works on the straight path L2, then turns and works on the straight path L4. In FIG. 6, the second tractor 1b automatically travels (follows) in accordance with the control data of the first tractor 1a (i.e., the control data shown in FIGS. 4 and 5). Therefore, for example, the first tractor 1a needs to invert the operation amount of the control data output from the state model. For example, when the first tractor 1a works on the straight path L3, its traveling direction is opposite to that of the control data, so the operation directions of the control points P53 (P2) and P54 are changed to the opposite directions. That is, the first tractor 1a changes to "left 5" (see FIG. 5) or "right 5" of the control point P53.

[0062] Furthermore, when the first tractor 1a transmits final control data for its own tractor 1a to the other tractor 1b to have the other tractor 1b work cooperatively, the first tractor 1a corrects the final control data to be transmitted to the other tractor 1b so that the steering angle in the final control data is larger on the uncultivated land side. For example, when transmitting final control data to the second tractor 1b, the first tractor 1a corrects the control data by adding "Left 1" to the control point P100 in FIG. 6 before transmitting the data to the second tractor 1b. This is because the right side of the second tractor 1b has already been worked (plowed) by the first tractor 1a, and therefore the soil in the field is softer on the right side of the second tractor 1b than on the left side (uncultivated land), which could cause the second tractor 1b to drift to the right when traveling along the straight path L2. In this way, the first tractor 1a corrects the final control data to be transmitted to the second tractor 1b, thereby enabling the second tractor 1b to minimize the influence of disturbances caused by the plowing of the first tractor 1a.

[0063] The controller 41 may learn the characteristics of disturbances based on the control data used when learning the state model. For example, the controller 41 learns the location of disturbances in the field and the degree of impact of the disturbances on each work machine W. The controller 41 may also analyze the characteristics of the control data output by the state model for each user and display the analysis results. This allows a user to select, for example, control data that prioritizes fuel efficiency (control data with a small amount of operation) or control data for careful work (control data with little deviation from the planned travel route R) from the characteristics of the control data of state models corresponding to other users.

[0064] As described above, the tractor 1 according to this embodiment includes a controller 41 and a memory unit 42. The memory unit 42 stores: a first state model, which is a state model that learns control data affected by disturbances when the tractor 1 without the work implement W actually travels through a field, and outputs first control data for subsequent travels; and a second state model, which is a state model that learns control data affected by disturbances when the tractor 1 with the work implement W actually travels through a field, and outputs second control data for subsequent travels. The controller 41 generates final control data by weighting and integrating the first control data output from the first state model and the second control data output from the second state model, and controls the tractor 1 according to the final control data. The weighting is changed depending on the state of the tractor 1 while it is traveling. This allows the weighting to be changed depending on the state of the tractor 1 that has been disturbed by the state of the field, enabling accurate control according to the state of the field.

[0065] Further advantages and modifications will readily occur to those skilled in the art. Therefore, the invention in its broader aspects is not limited to the specific details and representative embodiments shown and described above. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents. [Explanation of symbols]

[0066] 1,1a,1b Tractor 2 Running vehicle 3 Body frame 4 front wheels 5 rear wheels 6. Bonnet 7 Control Unit 7a Cabin 8. Cockpit 9. Steering wheel 10 Mission Case 12 Lifting device 13 Sensor mounting stay 20 Position detection device 21 Ridge position detection sensor 23 Engine revolution sensor 24 Vehicle speed sensor 25 Turning angle sensor 26 Lift arm sensor 30,30W satellite positioning device 40 Control device 41 Controller 42 Storage section 51 Steering gear 52 Transmission 53 Braking device 60,60W inertial measurement device 61 Cultivating Claw 121 Lifting cylinder 122 Lift arm 123 Lift Rod 124 Lower Link 125 Top Link 421 State Model Information

Claims

1. A controller and a storage unit are included. The storage unit a first state model that learns control data that is affected by disturbances when a work vehicle without a work implement actually travels through a field, and outputs first control data for the next and subsequent travels; a second state model that learns control data that is affected by disturbances when a work vehicle with a work implement actually travels through a field, and outputs second control data for the next and subsequent travels; Remember, The controller generating definite control data by weighting and integrating the first control data output from the first state model and the second control data output from the second state model, and controlling the work vehicle in accordance with the definite control data; The weighting is It is changed according to the state of the work vehicle while it is traveling. Work vehicle.

2. The controller Detecting abnormalities in the traveling vehicle body and the work implement of the work vehicle based on sensor information; When the abnormality is not detected in either the traveling vehicle body or the work machine, the weighting of the second control data is increased compared to the first control data; When the abnormality is detected in at least one of the traveling vehicle body and the working machine, the weighting of the first control data is increased compared to the second control data. The work vehicle according to claim 1 .

3. The controller When the abnormality is detected in at least one of the traveling vehicle body and the working machine, the first control data is generated as the final control data. The work vehicle according to claim 2 .

4. The controller A slip ratio of the work vehicle is detected when the work vehicle travels in the field, and if the detected slip ratio is higher than a previously detected slip ratio by a threshold value or more, the steering angle change amount in the final control data is corrected to be smaller. The work vehicle according to claim 1 .

5. The controller A deviation amount of the direction of the work implement per unit time while traveling in the field is detected, and if the deviation amount is equal to or greater than a threshold value, steering control is performed to correct the deviation amount. The work vehicle according to claim 1 .

6. The controller The tilt of the work implement per unit time while traveling in the field is detected, and if the tilt is equal to or greater than a threshold value, the speed is controlled so that the traveling speed is kept constant. The work vehicle according to claim 1 .

7. The controller When the time period during which the inclination is equal to or greater than the threshold continues for a predetermined time or longer, the speed is controlled so that the actual traveling speed is maintained at the reference speed determined by the final control data. The work vehicle according to claim 6.

8. The working machine is a tiller, The controller When the final control data of the work vehicle is transmitted to another work vehicle to cause the other work vehicle to perform cooperative work, the final control data is corrected so that the steering angle in the final control data to be transmitted to the other work vehicle is larger on the uncultivated land side. The work vehicle according to claim 1 .

Images