Work vehicles

Abstract

To provide a work vehicle capable of performing avoidance travelling for bypassing an obstacle, which can reduce unnecessary travelling.SOLUTION: A work vehicle 10 includes a vehicle body 11, a detection device for detecting an obstacle around the vehicle body 11, a discrimination processing unit for discriminating a type of a detection object on the basis of a result of the detection by the detection device, and a travelling control unit that causes the vehicle body to stop travelling or causes the vehicle body to execute avoidance travelling for bypassing the detection object according to the type of the detection object. The travelling control unit executes the avoidance travelling in which an avoidance distance between the detection object and a start position of the avoidance travelling is smaller when the detection object is an animal other than a human than when the detection object is a human.SELECTED DRAWING: Figure 10

Description

[Technical Field]

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

[0002] There is a demand for work vehicles to be equipped with various safety features. Patent Document 1 discloses a work vehicle that has a function to notify those around it that work is in progress. The work vehicle has a sensor that detects objects located around the work radius, and a notification unit that outputs a warning sound or announcement. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Korean Patent No. 10-2087714 Summary of the Invention [Problem to be solved by the invention]

[0004] For example, a work vehicle that performs agricultural work travels through a field to perform tasks such as plowing or harvesting crops, and also travels on farm roads (roads) to reach the field where the work is carried out. Furthermore, such work vehicles may be manually driven by a driver, or may be automatically driven without the driver's control, or remotely driven. For example, in the case of automatic driving, if an obstacle is present around the vehicle body, the vehicle is controlled to perform an avoidance drive that detours around the obstacle for safety reasons.

[0005] There are cases where there are multiple birds and other animals in the fields and farm roads on which work vehicles travel. If the work vehicle were to uniformly take evasive action every time it detected an animal as an obstacle, it would not be able to travel along the intended route, which would reduce work efficiency and make driving control more complicated. Furthermore, birds may not flee even when a work vehicle approaches to a certain extent, but may only flee when the work vehicle approaches significantly. For this reason, even if a work vehicle detects nearby birds and begins to maneuver to avoid them in advance, this may be in vain.

[0006] Therefore, the present disclosure provides a work vehicle that is capable of avoiding obstacles by detouring around them, thereby reducing unnecessary travel. [Means for solving the problem]

[0007] The work vehicle of the present disclosure comprises a vehicle body, a detection device that detects obstacles around the vehicle body, a discrimination processing unit that determines the type of detection target based on the detection results of the detection device, and a driving control unit that can perform avoidance driving by stopping the vehicle body or causing the vehicle body to detour around the detection target depending on the type of detection target, and the driving control unit performs avoidance driving in which the avoidance distance between the detection target and the start position of the avoidance driving is smaller when the detection target is an animal other than a human than when the detection target is a human. [Effects of the Invention]

[0008] The work vehicle of the present disclosure is capable of avoiding obstacles by detouring around them, thereby reducing unnecessary driving. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is a side view showing an embodiment of a work vehicle. [Figure 2] FIG. 2 is a block diagram showing the system configuration of the work vehicle. [Figure 3] FIG. 3 is an explanatory diagram showing a work vehicle traveling autonomously on a road outside a farm field. [Figure 4] FIG. 4 is an explanatory diagram of a target route when automatic driving is performed in a farm field. [Figure 5] FIG. 5 is a flowchart showing an example of control for automatic driving. [Figure 6] FIG. 6 is an explanatory diagram of a work vehicle that travels by automatic driving. [Figure 7] FIG. 7 is a flow diagram illustrating a case where avoidance travel is performed during operation of a work vehicle. [Figure 8] FIG. 8 is a flow diagram illustrating avoidance driving including a warning operation. [Figure 9] FIG. 9 is a flow diagram illustrating avoidance travel accompanied by deceleration driving. [Figure 10] FIG. 10 is an image diagram of map information when avoidance driving is performed midway along the target route. DETAILED DESCRIPTION OF THE INVENTION

[0010] <Summary of Embodiments of the Present Disclosure> The following provides an outline of embodiments of the present disclosure. (1) The work vehicle of this embodiment comprises a vehicle body, a detection device that detects obstacles around the vehicle body, a discrimination processing unit that discriminates the type of detection target based on the detection results of the detection device, and a driving control unit that can perform avoidance driving by stopping the vehicle body or causing the vehicle body to detour around the detection target depending on the type of detection target, and the driving control unit performs avoidance driving in which the avoidance distance between the detection target and the start position of the avoidance driving is smaller when the detection target is an animal other than a human than when the detection target is a human.

[0011] The work vehicle changes the timing to start evasive driving depending on the detected target. Non-human animals may not flee even when the work vehicle approaches to a certain extent, but may only flee when the work vehicle is in close proximity. The work vehicle performs evasive driving in accordance with the behavior of such animals. As a result, it is possible to reduce unnecessary driving.

[0012] (2) Preferably, the discrimination processing unit is capable of distinguishing a static object that does not take evasive action on its own as a type of detection object, and the driving control unit performs evasive driving with a shorter evasive distance when the detection object is an animal other than a human than when the detection object is the static object. According to the above configuration, when the detection target is a stationary object, it is possible to prevent interference with the stationary object by evasive driving. As mentioned above, non-human animals may only flee (take evasive action) when a work vehicle is located in close proximity, and they will flee relatively quickly. Therefore, when the detection target is a non-human animal, it is possible to prevent interference with the animal even if evasive driving is performed with a short evasive distance.

[0013] (3) Preferably, the discrimination processing unit is capable of discriminating between devices capable of taking evasive action on their own as the type of detection target, and the driving control unit performs evasive driving with a shorter evasive distance when the detection target is an animal other than a human than when the detection target is the device. Even if the device is capable of evasive action on its own, it may not be able to achieve it immediately. Therefore, with the above configuration, when the detection target is the device, the avoidance distance becomes longer than when it is an animal other than a human, making it possible to prevent interference with the device. In contrast, as mentioned above, non-human animals may only flee (take evasive action) when a work vehicle is in close proximity, and they may flee relatively quickly. Therefore, when the detected target is a non-human animal, it is possible to prevent interference with the animal even if evasive driving is performed with a short evasive distance.

[0014] (4) The agility and avoidance methods vary depending on the type of animal. Therefore, in the work vehicle of any one of (1) to (3), preferably, the discrimination processing unit is capable of discriminating the type of animal other than a human as the type of detection target, and the travel control unit executes different avoidance travel depending on the type of animal. According to the above configuration, an avoidance running action appropriate for the type of animal is executed.

[0015] (5) Birds can fly away quickly. Therefore, in the work vehicle of (4), preferably, the discrimination processing unit can discriminate between birds as a type of detected object, and the traveling control unit, when the detected object is a bird, executes evasive traveling with a shorter evasive distance than when the detected object is something other than a bird. According to the above configuration, an appropriate avoidance run is performed against birds.

[0016] (6) Preferably, the work vehicle of any one of (1) to (5) above has an alarm device that can execute an alarm operation depending on the type of the detection target. According to the above configuration, the warning action can make it possible to evacuate humans and non-human animals, thereby increasing the number of occasions where evasive driving is not required. Examples of the warning action include the generation of sound or ultrasonic waves, and light emission.

[0017] (7) Preferably, the work vehicle of (6) has a behavior detection unit that detects the behavior of the detection target based on the detection results of the detection device, and when the behavior detection unit detects the behavior of the detection target moving away from the vehicle body, it stops one or both of the evasive driving and the warning operation. According to the above configuration, unnecessary avoidance driving and warning operations are not performed.

[0018] (8) Preferably, in the work vehicle of (6) or (7) above, the warning device issues the warning, and then the driving control device executes the avoidance driving. According to the above configuration, if the detected object is an animal, an alarm is first issued. If the animal does not escape, the vehicle can take evasive action, thereby reducing unnecessary evasive action by the vehicle body.

[0019] (9) Preferably, in the work vehicle of any one of (1) to (8), the driving control unit is capable of decelerating the vehicle body depending on the type of the detection object, and when the detection object is an animal other than a human, the driving speed after deceleration is higher or the rate of deceleration is smaller than when the detection object is a human. As mentioned above, non-human animals may only flee when a work vehicle is nearby, and they tend to flee relatively quickly. Therefore, with this configuration, when the detection target is a non-human animal, it is possible to prevent interference with the animal without significantly slowing down the vehicle body.

[0020] <Details of the embodiment of the present disclosure> Hereinafter, details of embodiments of the present disclosure will be described with reference to the drawings. Note that at least some of the embodiments described below may be combined in any manner. [About work vehicles] FIG. 1 is a side view showing one embodiment of a work vehicle. The work vehicle 10 shown in FIG. 1 is a tractor, and is a vehicle that can be used to perform agricultural work. FIG. 1 shows a state in which a work implement 50 is coupled to the tractor. The technology of the present disclosure is not limited to tractors and can be applied to other types of work vehicles. For example, the work vehicle may be a harvester, rice transplanter, riding tiller, vegetable transplanter, mower, seed sower, fertilizer applicator, etc.

[0021] The work vehicle 10 of this embodiment has the function of operating in both a manual driving mode operated by a driver and an automatic driving mode not operated by a driver. The work vehicle 10 is capable of automatic driving and manual driving both within a field and on roads (farm roads) outside the field.

[0022] Manual driving is driving in which the operation (including driving) of the work vehicle 10 is performed by manual operation of a driver seated in the driver's seat 20 of the work vehicle 10. Autonomous driving is driving in which the operation (including driving) of the work vehicle 10 is performed by the functions of the control device 70 that the work vehicle 10 has, without manual operation by the driver.

[0023] The automatic driving is performed not only in an unmanned state where no driver is seated in the driver's seat 20, but also in an manned state where a driver is seated in the driver's seat 20. The automatic driving is realized by the function of the control device 70 (a control unit 73 for driving control, which will be described later). The control device 70 can control at least one of the steering required for the movement of the work vehicle 10, adjustment of the movement speed, and starting and stopping of movement. In the case of autonomous driving, in addition to the driving control of the work vehicle 10, the operation control of the work implement 50 is also performed without the driver's operation. In other words, the work vehicle 10 drives autonomously, while the work implement 50 performs work autonomously.

[0024] As will be explained later, the work vehicle 10 has a positioning device 37 that includes a GNSS receiver. A control device 70 (a control unit 73 for driving control, which will be explained later) causes the work vehicle 10 to travel automatically based on the position of the work vehicle 10 identified by the positioning device 37 and a target route that is stored in advance in a storage device 79 (see FIG. 2).

[0025] Autonomous driving also includes cases where the work vehicle 10 travels autonomously without human intervention in controlling its movement, sensing the surrounding environment using a detection device, which will be described later. Autonomous driving includes not only movement of the work vehicle 10 toward a destination along a predetermined route (the target route), but also movement of the work vehicle 10 following a tracking target. During such automatic driving, as will be described later, obstacle detection and obstacle avoidance travel are performed.

[0026] The work vehicle 10 can also be operated remotely by someone other than the driver seated in the driver's seat 20. For this purpose, the work vehicle 10 has a communication device 16 (see FIG. 2). The work vehicle 10 is remotely operated using wireless communication between the work vehicle 10 and a management computer (not shown) at a management center. During remote operation, the driver's seat 20 may be unmanned, but may also be manned. Remote operation may take priority over manual operation.

[0027] The directions of the work vehicle 10 are defined below. The work vehicle 10 has a driver's seat 20. The front-rear, left-right, and up-down directions of the work vehicle 10 are defined based on the driver seated in the seat 20c of the driver's seat 20. In other words, the direction forward for the driver is "forward," and the direction behind is "rear." The right direction for the driver is "right," and the left direction is "left." The front-rear and left-right directions are parallel to the ground, and are perpendicular to the front-rear and left-right directions. The up-down direction is perpendicular to both the front-rear and left-right directions. The left-right direction is sometimes referred to as the "vehicle width direction." The front direction is the "traveling direction" of the work vehicle 10. It should be noted that the work vehicle 10 does not have to have a driver's seat 20. In this case, the direction in which work proceeds for the work vehicle 10 is the "forward" direction, and the opposite direction is the "rear" direction. Facing the direction in which work proceeds, the right direction of the work vehicle 10 is the "right" direction, and the left direction is the "left" direction.

[0028] The work vehicle 10 has a vehicle body 11, a prime mover 12, a transmission 13, a traveling device 14, a steering device 15, a coupling device 40, and a control device 70. The vehicle body 11 has a chassis 21 that forms the framework of the vehicle, a body 22 that forms the exterior, and a driver's seat 20. The driver's seat 20 is provided with a steering wheel 30 operated by the driver, and an operation unit (operation interface) 31 that includes an operation terminal operated by the driver and a group of operation switches.

[0029] The prime mover 12 is an engine or a motor, and in this embodiment, is a diesel engine. The traveling device 14 has front wheels 14a and rear wheels 14b. The rotational force of the prime mover 12 is changed in speed by the transmission 13, and the rotational force is transmitted to the wheels, causing the work vehicle 10 to travel. When the work vehicle 10 travels within a field to perform work, the traveling device 14 may have crawlers as one or both of the front and rear wheels.

[0030] The steering device 15 has a steering shaft 25 that is rotated by the steering wheel 30. The steering device 15 changes the rolling direction of the wheels (front wheels 14a) and changes the traveling direction of the work vehicle 10. The steering device 15 has an assist mechanism (power steering device). The assist mechanism uses hydraulics or electricity to assist the driver's operation force on the steering wheel 30. In the case of autonomous driving, the assist mechanism performs steering and changes the traveling direction.

[0031] The transmission 13 is configured to have a plurality of gears, etc. The transmission 13 changes the propulsive force and travel speed of the work vehicle 10. The transmission 13 can also switch the work vehicle 10 between forward and reverse travel. The work vehicle 10 has a power take-off mechanism (hereinafter referred to as the "PTO mechanism"). In this embodiment, the transmission 13 has the PTO mechanism. The PTO mechanism has a PTO shaft 17, which is one of the output shafts of the transmission 13. The PTO shaft 17 rotates due to the power of the prime mover 12. The rotational force of the PTO shaft 17 operates various drive parts of the work implement 50. The PTO shaft 17 is the output shaft for operating the work implement 50.

[0032] The coupling device 40 couples the work implement 50 to the vehicle body 11. The coupling device 40 is mounted on the rear of the vehicle body 11 (chassis 21). The coupling device 40 has a lifting link mechanism that supports the work implement 50 so that it can be raised and lowered. The lifting link mechanism is configured, for example, by a three-point link mechanism. The work implement 50 can be attached to and detached from the work vehicle 10 by the coupling device 40. The lifting link mechanism changes the height position of the work implement 50 and changes the posture of the work implement 50 using an actuator such as a hydraulic device.

[0033] The work implement 50 shown in Fig. 1 is a rotary tiller. However, the work implement 50 is not limited to this and may be, for example, a seed sower, a fertilizer applicator, a transplanter, a brush cutter, a rake, a grass collector, a harvester, etc. The coupling device 40 connects the desired work implement 50 to the work vehicle 10. The work vehicle 10 tows the work implement 50 and causes the work implement 50 to perform a predetermined task. The coupling device 40 may be provided at the front of the vehicle body 11.

[0034] The work vehicle 10 has an imaging device. In this embodiment, the imaging device is a camera 35. The cameras 35 are installed, for example, on the front, rear, left and right sides of the work vehicle 10, and capture images of the environment around the work vehicle 10. The camera 35 is, for example, a CCD camera equipped with a CCD image sensor, or a CMOS camera equipped with a CMOS image sensor. The camera 35 has a processing circuit that processes signals output from the image sensor, and image information of the surroundings is obtained by the processing circuit.

[0035] When the work vehicle 10 travels on a road or in a field, the camera 35 is used not only to recognize white lines, signs, indications, etc., but also to recognize surrounding obstacles. The camera 35 may be either a visible light camera that generates a visible light image or an infrared camera that generates an infrared image, or both. In the case of an infrared camera, it becomes easier to detect objects (obstacles) at night.

[0036] The image information acquired by the camera 35 is transmitted to the control device 70. In addition to being used to control automatic driving, the image information is also used in manual driving. As will be explained later, the control device 70 uses the image information in combination with other information (sensor data, described later) to detect obstacles around the work vehicle 10 and can cause the work vehicle 10 to avoid those obstacles.

[0037] The work vehicle 10 has a three-dimensional range sensor. The three-dimensional range sensor in this embodiment is a LiDAR (Light Detection And Ranging) sensor 36. The LiDAR sensor 36 is disposed, for example, at the lower front part of the vehicle body 11. The LiDAR sensor 36 may be disposed in another position. The LiDAR sensor 36 acquires and outputs sensor data indicating the distance and direction of each measurement point on a surrounding object, and sensor data indicating the two-dimensional or three-dimensional coordinate values ​​of each measurement point on a surrounding object. The sensor data acquired by the LiDAR sensor 36 is transmitted to the control device 70.

[0038] The sensor data from the LiDAR sensor 36 is used to detect surrounding obstacles. By using the sensor data in combination with other information (the image information), the control device 70 can detect obstacles around the work vehicle 10 and control the work vehicle 10 to avoid the obstacles. The camera 35 and the LiDAR sensor 36 function as obstacle sensors (detection devices) that detect obstacles in the surrounding area.

[0039] The sensor data of the LiDAR sensor 36 is also used for other purposes. The control device 70 can execute the process of generating an environmental map based on the sensor data, using an algorithm such as SLAM (Simultaneous Localization and Mapping). The process of generating an environmental map may be executed by a computer such as an external management device that is capable of communicating with the work vehicle 10.

[0040] The work vehicle 10 has a positioning device 37. The positioning device 37 receives satellite signals transmitted from multiple GNSS satellites and performs positioning based on the satellite signals. GNSS is a general term for satellite positioning systems such as GPS (Global Positioning System), QZSS (Quasi-Zenith Satellite System: for example, "Michibiki"), GLONASS (Russia), Galileo (Europe), and BeiDou (China).

[0041] The positioning device 37 has a receiver 37a that receives satellite signals, and a processor (arithmetic processing unit) 37b. The receiver 37a has an antenna that receives signals from GNSS satellites. The processor 37b calculates and determines the position (coordinates) of the work vehicle 10 based on the signals received by the antenna. The receiver 37a is placed, for example, above the driver's seat 20. Information indicating the position of the work vehicle 10 is transmitted to the control device 70 and used for automatic driving, etc. The positioning device 37 corrects or complements the position information of the work vehicle 10 based on satellite signals, using data acquired by the camera 35 and the LiDAR sensor 36. The position of the work vehicle 10 is identified with higher accuracy.

[0042] The work vehicle 10 has an inertial measurement unit 38. The inertial measurement unit 38 has a three-axis gyro sensor and a three-directional acceleration sensor. The inertial measurement unit 38 detects the tilt and movement of the work vehicle 10. The signals acquired by the inertial measurement unit 38 are transmitted to the control device 70. The detection signals from the inertial measurement unit 38 are used to complement the position information of the work vehicle 10. This improves positioning accuracy.

[0043] The work vehicle 10 has an alarm device 19. The alarm device 19 is a device that issues an alarm to animals around the work vehicle 10. The alarm device 19 has, for example, a speaker or buzzer that generates sound, an oscillator that generates ultrasonic waves, and a floodlight that emits light. The work vehicle 10 may have one or more of the speaker, buzzer, oscillator, and lighting device. The alarm operation, such as sound generation by the alarm device 19, is controlled by a control device 70 (for example, a control unit 74 for arithmetic processing). Light emission as an alarm operation is effective at night.

[0044] [System configuration of work vehicle 10] FIG. 2 is a block diagram showing the system configuration of the work vehicle 10. The control device 70 is configured by a control unit (computer) including a processor (arithmetic processing unit) and memory consisting of RAM, ROM, etc. The processor reads and executes a computer program from the memory, thereby performing each function of the control device 70. The control device 70 may be configured by one control unit (ECU: Electronic Control Unit) or by multiple control units. When the control device 70 is configured by multiple control units, information can be communicated between these control units.

[0045] In this embodiment, the control device 70 has a control unit 71 for speed control, a control unit 72 for steering, a control unit 73 for driving control, and a control unit 74 for arithmetic processing. The control unit 73 for driving control will be referred to as the driving control unit 73 hereinafter. The control unit 74 for arithmetic processing will be referred to as the arithmetic processing unit 74 hereinafter.

[0046] The control device 70 includes a storage device (storage unit) 79, which is made up of a nonvolatile memory or the like that stores various types of information. Various computer programs for causing the control unit to function are stored in the storage device 79. The storage device 79 stores map information that can be used for autonomous driving, a trained model (described later), a database, and the like.

[0047] The speed control control unit 71 generates drive signals and supplies them to the prime mover 12, the transmission 13, and the brake device, thereby controlling the working vehicle 10 to adjust its traveling speed and stop it. The steering control unit 72 provides the generated steering signal to the steering device 15. The control unit 72 controls the hydraulic device or electric motor of the auxiliary mechanism of the steering device 15 based on the measurement value of the rotation sensor of the steering shaft 25, thereby controlling the steering of the work vehicle 10.

[0048] [Operation control unit 73] The driving control unit 73 performs overall control over driving, including automatic driving, of the work vehicle 10. Control by the driving control unit 73 may be achieved in cooperation with a speed control control unit 71 and a steering control unit 72. The driving control unit 73 is capable of performing control over each of automatic driving, remote driving, and manual driving.

[0049] The driving control unit 73 can also perform control to cause the vehicle body 11 to make an avoidance drive around an obstacle present around the work vehicle 10, or control to stop the vehicle body 11. The driving control unit 73 intervenes in the avoidance drive or stop drive during any of the automatic driving, remote driving, and manual driving. In the specific examples described below, a case where the avoidance drive intervenes in the automatic driving will be mainly described.

[0050] As will be explained later in detail, the type of detected object that is an obstacle is identified based on the detection results of the camera 35 and the like that the work vehicle 10 has. Avoidance driving and stopping of driving are performed depending on the type of detected object. If it is determined that interference with the detected obstacle cannot be avoided even when avoidance driving is performed, the driving control unit 73 stops the driving of the vehicle body 11. In this way, the driving control unit 73 functions as a driving control unit that can selectively stop the vehicle body 11 from moving or perform avoidance driving that detours the vehicle body 11 around the detected object, depending on the type of detected object.

[0051] [Analysis processing unit 74] The arithmetic processing unit 74 detects obstacles around the vehicle body 11 based on the detection results of a detection device including one or both of the camera 35 and the LiDAR sensor 36. The arithmetic processing unit 74 determines the type of obstacle to be detected based on the detection results. This determination is performed, for example, using a trained model. For this purpose, machine learning is performed in advance using information on the detection results of the detection target by the detection device and the type of the detection target (label) as a data set. A trained model is generated using the information on the detection results as input data and the type of the detection target as output data. The trained model is stored in the storage device 79.

[0052] The arithmetic processing unit 74 can distinguish between humans and non-human animals as types of detection targets. The arithmetic processing unit 74 can distinguish between static objects that do not take evasive action on their own, such as plants and artificial pillars, as types of detection targets. The arithmetic processing unit 74 can distinguish between devices that are capable of taking evasive action (evasion action) on their own, such as other agricultural machinery, unmanned robots, automobiles, etc. as types of detection targets.

[0053] The arithmetic processing unit 74 can distinguish between non-human animals and the types of animals, such as birds, dogs, weasels, wild boars, deer, etc. A different trained model may be used to distinguish between types of animals. As described above, the arithmetic processing unit 74 functions as a discrimination processing unit that discriminates the type of detection target (obstacle) based on the detection results of a detection device such as the camera 35.

[0054] The arithmetic processing unit 74 can further detect the behavior of the detection target based on the detection results of a detection device such as the camera 35. If the detection target is an animal, the animal will take action to escape (avoid) from the work vehicle 10. The arithmetic processing unit 74 determines whether the detection target, such as an animal, is moving or not, based on one or both of the image information from the camera 35 and the sensor data from the LiDAR sensor. In addition to determining whether or not movement has occurred, the arithmetic processing unit 74 can also determine the direction of movement. In other words, it can determine whether the detection target is moving in a direction away from the work vehicle 10.

[0055] In this way, the arithmetic processing unit 74 functions as a behavior detection section that detects the behavior of the detection target. The determination processing section and the behavior detection section may be realized by separate control units.

[0056] [Automatic driving of work vehicle 10] The work vehicle 10 requires a target route for autonomous driving. The target route is generated before autonomous driving begins. The target route is generated by a control unit of a control device 70 possessed by the work vehicle 10. Alternatively, the target route may be generated by a computer of a management device other than the work vehicle 10. In this case, the work vehicle 10 acquires information about the target route via the communication device 16.

[0057] FIG. 3 is an explanatory diagram showing a work vehicle 10 traveling autonomously on a road 8 outside a field 7. When the work vehicle 10 travels autonomously on a road 8, the work vehicle 10 requires environmental map information. The environmental map is map information of the environment in which the work vehicle 10 travels, and includes map information such as the field 7 and the roads 8 for traveling between the fields 7. A target route G is generated based on the map information. In other words, when an arrival destination is set in the map information, a route from the current position of the work vehicle 10 to that destination is generated as the target route G. In the case of FIG. 3, a target route G along the road 8 has been generated. Information about the target route G is stored in a storage device 79 together with the map information.

[0058] 4 is an explanatory diagram of a target route G when autonomous driving is performed within a field 7. The field 7 includes a work area 7a where the work vehicle 10 performs work, and a headland 7b located near the outer periphery of the field 7. Which areas of the field 7 correspond to the work area 7a and which correspond to the headland 7b is set by the user. The target route G in the field 7 includes multiple parallel main routes P1 and a turning route P2 connecting the two main routes P1. The main routes P1 are located in the work area 7a, and the turning route P2 is located in the headland 74. Although the main route P1 shown in FIG. 4 is straight, the main route P1 may include a curved portion.

[0059] The distance between the dashed lines in Figure 4 represents the working width of the work implement 50. The working width is set in advance and stored in the storage device 79. The working width is input to the control device 70 by the user operating the operation unit 31, which includes an operation terminal. The working width may be automatically recognized when the work implement 50 is connected to the work vehicle 10 and input to the control device 70. The distance between adjacent main paths P1 is set to match the working width. The target route G is created so as to pass through (cover) the entire work area 7a. The work vehicle 10 automatically travels along the target route G, repeatedly making round trips from the work start point S to the work end point G. Note that the target route G shown in Figure 4 is an example.

[0060] Automatic driving becomes possible when the target route G for the outside of the field 7 or the inside of the field 7 is generated. Fig. 5 is a flowchart showing an example of control for automatic driving. While the work vehicle 10 is traveling, the driving control unit 73 performs automatic steering by executing the processing of steps S101 to S105 shown in Fig. 5. The traveling speed of the work vehicle 10 is maintained at a preset value, but may be automatically changed midway.

[0061] While the work vehicle 10 is traveling, the driving control unit 73 acquires position information of the work vehicle 10 obtained by the positioning device 37 (step S101). The position information is information that indicates the current position of the work vehicle 10. The driving control unit 73 calculates the deviation E between the current position of the work vehicle 10 and the target route G (step S102). 6 is an explanatory diagram of a work vehicle 10 traveling by autonomous driving. Deviation E is the difference, for example, distance, between the current position of the work vehicle 10 and the target route G. The driving control unit 73 determines whether deviation E exceeds a threshold value (step S103 in FIG. 5).

[0062] If the deviation E exceeds the threshold value ("Yes" in step S103), the operation control unit 73 outputs a command signal to the steering control unit 72 to reduce the deviation E. Upon receiving the command signal, the steering control unit 72 changes the control parameters to be given to the steering device 15, and changes the steering angle (step S104). The steering angle is changed until the deviation E becomes equal to or less than the threshold value. In step S103, if the deviation E is equal to or smaller than the threshold value, step S104 is skipped.

[0063] In this embodiment (see FIG. 6), the steering device 15 is controlled based on the deviation E between the current position of the work vehicle 10 and the target route G, but the steering device 15 may also be controlled by adding the deviation in heading. For example, the driving control unit 73 determines whether the angular difference between the heading of the work vehicle 10 identified by one or both of the positioning device 37 and the inertial measurement unit 38 and the direction of the target route G exceeds a preset threshold. Note that this angular difference is the deviation in heading. If the angular difference exceeds the threshold, the control parameters of the steering device 15 (for example, the steering angle) are changed according to the deviation.

[0064] 5 is a step for determining whether or not the operation control unit 73 has received a command to end the operation. The command to end the operation is issued, for example, in the following cases. 1) The administrator (user) remotely instructs the system to stop automatic operation. 2) The work vehicle 10 reaches the destination (end position).

[0065] If the operation control unit 73 does not receive the command to end the operation ("No" in step S105), the operation control unit 73 returns to step S101 and executes the same processes thereafter. The operation control unit 73 repeats the operations from step S101 to S105 until it receives the command to end the operation.

[0066] If the driving control unit 73 detects an obstacle at any step during the processing of steps S101 to S105 for automatic driving, it can execute avoidance driving to avoid the obstacle. That is, while the work vehicle 10 is traveling, the camera 35 captures images of the surrounding environment, and the LiDAR sensor 36 senses surrounding objects. The arithmetic processing unit 74 detects obstacles present around the work vehicle 10 based on the detection results of the camera 35 and the LiDAR sensor 36, and determines the type of obstacle. The driving control unit 73 performs avoidance driving depending on the type of obstacle.

[0067] When avoidance travel begins, the work vehicle 10 travels along a travel route that differs from the target route G. As will be explained later, this travel route changes depending on the type of obstacle. 5 shows a case where actual avoidance driving is started after step S104. When the detour driving is completed, the driving control unit 73 returns the work vehicle 10 to the driving mode before the avoidance driving. In other words, the driving control unit 73 drives the vehicle main body 11 so that the position of the work vehicle 10 returns to the target route G. Thereafter, the process of automatic driving along the target route G (steps S101 to S105) is executed.

[0068] [Regarding avoidance driving of the work vehicle 10] For example, during automatic driving, the driving control unit 73 can execute avoidance driving, which causes the vehicle body 11 to detour around a detected object, depending on the type of detected object. Avoidance driving is driving in which a steering signal is output to the steering device 15, causing the vehicle body 11 to travel in either the left or right direction while maintaining a distance from the detected object, and detouring around the detected object. The distance between the position of the vehicle body 11 and the detected object is referred to as the "avoidance distance." The avoidance distance at the start position of the avoidance driving and the avoidance distance at a position midway through the avoidance driving may be the same or different. The avoidance distance is also referred to as the avoidance margin.

[0069] The driving control unit 73 can execute a plurality of patterns of avoidance driving. For the avoidance driving, an avoidance distance is set arbitrarily. The setting may be performed by the driving control unit 73 or the arithmetic processing unit 74. If the avoidance distance is small, the timing to start avoidance driving with respect to the detected object will be late, that is, the avoidance driving will start at a position relatively close to the detected object. When the avoidance distance is large, the timing to start avoidance driving with respect to the detected object becomes earlier, that is, the avoidance driving starts at a position relatively far from the detected object. The avoidance distance is changed depending on the type of detection target.

[0070] When the detour around the detected target is completed, the driving control unit 73 returns the work vehicle 10 to the driving mode before the avoidance driving. In other words, the driving control unit 73 drives the vehicle main body 11 so as to return to the target route set for automatic driving. Thereafter, automatic driving along the target route continues.

[0071] The avoidance driving may be automatically performed not only during automatic driving but also during manual driving. In other words, even if the driver seated in the driver's seat 20 is operating the steering wheel 30 while driving, if an obstacle is detected in the surroundings and avoidance driving becomes necessary, the driving control unit 73 may intervene in the manual driving with the avoidance driving. Once the detour around the detected object is completed, the vehicle returns to manual driving mode. Alternatively, an operation switch for canceling the avoidance driving may be provided on the driver's seat 20, and the vehicle returns to manual driving mode when the driver operates the operation switch.

[0072] [Specific example of avoidance travel of work vehicle 10] 7 is a flow diagram illustrating a case where avoidance driving is performed while the work vehicle 10 is driving. The following explanation will be given for a case where the work vehicle 10 is mainly driven autonomously. The explanation will be given for a case where the work vehicle 10 is mainly driving on a road (farm road).

[0073] While the work vehicle 10 is traveling, the camera 35 photographs the surrounding environment and acquires image information from moment to moment. The LiDAR sensor 36 senses surrounding objects and acquires three-dimensional point cloud data from moment to moment (step ST1 in FIG. 7). The control device 70 acquires the image information and three-dimensional point cloud data that are the outputs of the camera 35 and the LiDAR sensor 36, respectively. The arithmetic processing unit 74 detects obstacles present around the work vehicle 10 based on the detection results of the camera 35 and the LiDAR sensor 36, and if detected, determines the type of obstacle (step ST2 in FIG. 7).

[0074] For example, obstacles are detected using three-dimensional point cloud data, and the type of the obstacle is determined using image information. The type of obstacle is determined using the trained model as described above. The arithmetic processing unit 74 may perform sensor fusion processing to combine image information from the camera 35 with three-dimensional point cloud data from the LiDAR sensor 36 to detect obstacles and determine their type. The LiDAR sensor 36 is suitable for detecting objects with irregular contours, but is not suitable for detecting objects with small features. Therefore, by combining image information from the camera 35 with the output of the LiDAR sensor 36, highly accurate determination is possible.

[0075] If the result of the determination is that the detected object is an animal other than a human, the type of animal is determined (step ST3-1 in FIG. 7). As a result, the case where the detected object is a "bird" ("Yes" in step ST4-1 in FIG. 7) will be described. In this case, the driving control unit 73 executes avoidance traveling with a short avoidance distance between the detected object (bird) and the start position of the avoidance traveling (step ST5-1 in FIG. 7).

[0076] On the other hand, if the result of determining the type of animal is that the detected object is an "animal other than a bird" ("No" in step ST4-1 in FIG. 7), the driving control unit 73 executes avoidance driving for the detected object (animal other than a bird) at an avoidance distance (small-2) that is larger than the avoidance distance (small-1) when the detected object is a "bird" (step ST5-2 in FIG. 7). However, the avoidance distance (small-2) in this case is shorter than when the detected object is a human.

[0077] The agility and avoidance method differ depending on the type of animal. Therefore, the driving control unit 73 performs different avoidance maneuvers depending on the type of animal. Here, different avoidance maneuvers are avoidance maneuvers with different avoidance distances. The arithmetic processing unit 74 can distinguish a bird as the type of detected object (step ST4-1 in FIG. 7). Birds can fly away and escape quickly. Therefore, when the detected object is a bird, the driving control unit 73 performs an avoidance maneuver with a shorter avoidance distance than when the detected object is something other than a bird (step ST5-1 in FIG. 7).

[0078] A case where the result of the determination (step ST2 in FIG. 7) is that the detected object is a human will be described. The driving control unit 73 executes avoidance driving for the detected object (human) (step ST5-5). In this case, the avoidance distance is longer than when the detected object is an "animal other than a human."

[0079] This is because the driving control unit 73 performs evasive driving in accordance with the habits of animals (other than humans). In other words, there are cases where non-human animals do not flee even when the work vehicle 10 approaches to a certain extent, but only flee when the work vehicle 10 is in close proximity. Therefore, the work vehicle 10 performs evasive driving in accordance with the habits of such animals. In other words, when the detected target is an animal other than a human, the driving control unit 73 performs evasive driving in which the avoidance distance between the detected target and the start position of the evasive driving is shorter than when the detected target is a human.

[0080] A case will be described where, as a result of the determination (step ST2 in FIG. 7), the detected object is a static object that does not take evasive action on its own, such as a plant or an artificial pillar. The driving control unit 73 executes evasive driving with a shorter avoidance distance for the detected object (static object) than when the detected object is a human (step ST5-3). However, the avoidance distance in this case is longer than when the detected object is a "non-human animal." The driving control unit 73 executes evasive driving with a shorter avoidance distance when the detected object is a non-human animal than when the detected object is a static object such as the above.

[0081] A case will be described where, as a result of the determination (step ST2 in FIG. 7), the detected object is a device capable of taking evasive action on its own, such as another agricultural machine, an unmanned robot, or an automobile. The driving control unit 73 executes evasive driving with a shorter avoidance distance for the detected object (device) than when the detected object is a human (step ST5-4). However, the avoidance distance in this case is longer than when the detected object is a "non-human animal." The driving control unit 73 executes evasive driving with a shorter avoidance distance when the detected object is a non-human animal than when the detected object is a device such as the above.

[0082] [Evasive driving including warnings] Figure 8 is a flow diagram illustrating evasive driving including an alarm operation. The arithmetic processing unit 74 identifies the type of detection target (step ST2 in Figures 7 and 8). Control of evasive driving including an alarm operation can be added to the "case of human" and "case of non-human animal" of the evasive driving flow shown in Figure 7 (step ST2 and thereafter in Figure 7).

[0083] If the detection target is a human or a non-human animal, the alarm device 19 executes an alarm operation (step ST11 in FIG. 8). If the detection target is a static object such as a plant or an artificial pillar, the alarm device 19 does not execute an alarm operation. The warning action allows humans and non-human animals to escape, increasing the chances of avoiding the need for evasive maneuvers. The alarm operation is one of sound generation, ultrasonic wave generation, and light emission, or a combination of at least two of these. Different types of animals are repelled by different alarms. Therefore, the alarm device 19 may change the type of alarm operation depending on the type of animal and execute the alarm operation.

[0084] After issuing the warning, the arithmetic processing unit 74 detects the behavior of the detection target based on the detection results of a detection device such as the camera 35 (step ST12 in FIG. 8). If the arithmetic processing unit 74 detects the behavior of the detection target moving away from the vehicle body 11 (if "Yes" in step ST13 in FIG. 8), it cancels the warning (step ST14-1 in FIG. 8). On the other hand, if the arithmetic processing unit 74 does not detect the behavior of the detection target moving away from the vehicle body 11 (if "No" in step ST13 in FIG. 8), the driving control unit 73 starts avoidance driving at a predetermined timing.

[0085] During avoidance driving (step ST14-2 in FIG. 8), the arithmetic processing unit 74 detects the behavior of the detection target based on the detection results of a detection device such as the camera 35 (step ST15 in FIG. 8). When the arithmetic processing unit 74 detects the behavior of the detection target moving away from the vehicle main body 11 (if "Yes" in step ST16 in FIG. 8), it stops the avoidance driving that it has started (step ST17-1 in FIG. 8) and returns to the driving mode before the avoidance driving (step ST18 in FIG. 8). In other words, it stops the avoidance driving midway, and drives the vehicle main body 11 from the stopped state to return to the route set for autonomous driving. Thereafter, autonomous driving continues.

[0086] As described above, in the case of the work vehicle 10 of this embodiment, the alarm device 19 executes an alarm operation according to the type of detected object (step ST11). If the alarm device 19 executes an alarm operation and no movement of the detected object moving away from the work vehicle 10 is detected (if "No" in step ST13), avoidance driving is executed. In this way, if the detected object is an animal, the alarm operation is executed first. If the animal still does not escape, avoidance driving is executed. As a result, it is possible to reduce unnecessary avoidance driving of the vehicle body 11.

[0087] [Regarding slowing down] As described above, the driving control unit 73 has a function as a driving control unit that enables the execution of avoidance driving. The driving control unit 73 can also decelerate the vehicle body 11 in conjunction with the avoidance driving, depending on the type of detected object. The deceleration driving is performed during either or both of the driving immediately before the avoidance driving and the driving during the avoidance driving.

[0088] The deceleration driving will be explained in detail. Fig. 9 is a flow diagram explaining the avoidance driving accompanied by the deceleration driving. As mentioned above, animals other than humans may only flee when the work vehicle 10 is located in the immediate vicinity, and they will also flee relatively quickly.

[0089] If the detection target is a human (step ST21 "Yes" in Figure 9), the driving control unit 73 controls one or both of the prime mover 12 and the transmission 13 to perform deceleration driving, and the vehicle body 11 travels at a low speed at a first speed (step ST22-1 in Figure 9). If the detection target is an animal other than a human ("No" in step ST21 in FIG. 9), the driving control unit 73 controls one or both of the prime mover 12 and the transmission 13 so that the running speed after deceleration is higher than when the detection target is a human. If the detection target is an animal other than a human, deceleration driving is performed, but the running speed is higher than the first speed (step ST22-2 in FIG. 9).

[0090] Alternatively, if the detection target is a human ("Yes" in step ST21 in FIG. 9), the operation control unit 73 performs deceleration operation using a negative first acceleration with a large absolute value. When the detection target is an animal other than a human ("No" in step ST21 in FIG. 9), the driving control unit 73 controls one or both of the prime mover 12 and the transmission 13 so that the rate of deceleration is smaller than when the detection target is a human. In other words, when the detection target is an animal other than a human, deceleration driving is performed with a negative acceleration whose absolute value is smaller than the first acceleration. Note that the speed of the vehicle body 11 after deceleration driving may be the same when the detection target is a human and when the detection target is a non-human animal.

[0091] The above-described controls for avoidance travel may be executed when the work vehicle 10 is traveling on a road outside a field, or may be executed when the work vehicle 10 is traveling in a field.

[0092] [Examples of evasive driving] As described above, in the case of autonomous driving, a target route G is set in the map information stored in the storage device 79. The work vehicle 10 travels so that the target route G coincides with the position of the vehicle body 11 measured by the positioning device 37, which includes a GNSS receiver. FIG. 10 is an image diagram of map information M when an evasive maneuver V is executed midway along the target route G. The work vehicle 10 travels along the target route G. When the work vehicle 10 detects an obstacle X ahead on the target route G and an evasive maneuver V is necessary, the driving control unit 73 (or the arithmetic processing unit 74) sets an obstacle area A1 and a buffer area A2 on the map information M. The obstacle area A1 is an area centered around the detection target X. The buffer area A2 is an area surrounding the obstacle area A1.

[0093] The driving control unit 73 performs avoidance driving so that the work vehicle 10 avoids at least the obstacle area A1. The driving control unit 73 may also perform avoidance driving so that the work vehicle 10 avoids the obstacle area A1 and the buffer area A2. When the detection target that becomes the obstacle X is an animal other than a human, the obstacle area A1 is set to be narrow. When the detection target that becomes the obstacle X is a human, the obstacle area A1 is set to be wider than when the detection target that becomes the obstacle X is an animal other than a human. The buffer area A2 may have a fixed size from the edge of the obstacle area A1, or may vary in size in the same way as the obstacle area A1.

[0094] This allows for avoidance running in which the avoidance distance between the detected object (obstacle X) and the start position of the avoidance running is smaller when the detected object (obstacle X) is an animal other than a human than when the detected object (obstacle X) is a human.

[0095] [Regarding the work vehicle 10 of this embodiment] As described above, the work vehicle 10 of this embodiment has the vehicle body 11 and a detection device (camera 35 and LiDAR sensor 36) that detects obstacles around the vehicle body 11. The work vehicle 10 also has an arithmetic processing unit 74 and a driving control unit 73. The arithmetic processing unit 74 determines the type of detected object based on the detection results of the detection device. The driving control unit 73 is capable of stopping the vehicle body 11 from traveling or executing avoidance driving that causes the vehicle body 11 to detour around the detected object, depending on the type of detected object. When the detected object is a non-human animal, the driving control unit 73 executes avoidance driving in which the avoidance distance between the detected object and the start position of the avoidance driving is shorter than when the detected object is a human.

[0096] According to the work vehicle 10 of this embodiment, the timing to start evasive driving is changed depending on the detected target. There are cases where animals other than humans do not flee even when the work vehicle approaches to a certain extent, but only flee when the work vehicle is in close proximity. The work vehicle 10 performs evasive driving in accordance with the behavior of such animals. As a result, it is possible to reduce unnecessary driving.

[0097] 〔others〕 The above-described embodiments are illustrative in all respects and are not limiting. The scope of the present invention is defined by the claims rather than the above-described embodiments, and includes all modifications within the scope equivalent to the configurations described in the claims. [Explanation of symbols]

[0098] 10 Work vehicles 11 Vehicle body 19 Alarm device 35 Camera (detection device) 36 LiDAR sensor (detection device) 73 Control unit for driving control (driving control unit) 74 Control unit for arithmetic processing (discrimination processing unit, behavior detection unit)

Claims

1. The vehicle body, a detection device for detecting obstacles around the vehicle body; a discrimination processing unit that discriminates the type of the detection target based on the detection result of the detection device; a travel control unit that enables the vehicle body to perform avoidance travel by detouring around the detection target in accordance with the type of the detection target; Equipped with the avoidance traveling is traveling in one of the left and right directions of the vehicle body while maintaining a distance from the detection target, and detouring around the detection target, The traveling control unit When the detection target is an animal other than a human, the avoidance running is performed with a smaller avoidance distance between the detection target and the start position of the avoidance running than when the detection target is a human. Work vehicle.

2. the discrimination processing unit is capable of discriminating, as a type of detection target, a stationary object that does not take evasive action by itself; The traveling control unit When the detection target is an animal other than a human, the avoidance traveling is performed with the avoidance distance being shorter than when the detection target is the stationary object. The work vehicle according to claim 1 .

3. the discrimination processing unit is capable of discriminating, as a type of detection target, a device that is capable of taking evasive action by itself; The traveling control unit When the detection target is an animal other than a human, the avoidance travel is performed with the avoidance distance being shorter than when the detection target is the device. The work vehicle according to claim 1 or 2.

4. the discrimination processing unit is capable of discriminating the type of animal other than a human as the type of detection object; the running control unit executes the avoidance running differently depending on the type of animal. The work vehicle according to claim 1 or 2.

5. the discrimination processing unit is capable of discriminating between birds as types of detection objects, the traveling control unit, when the detection object is a bird, executes the avoidance traveling with the avoidance distance being shorter than when the detection object is something other than a bird. The work vehicle according to claim 4.

6. an alarm device that can execute an alarm operation depending on the type of the detection target; The work vehicle according to claim 1 .

7. a behavior detection unit that detects the behavior of the detection target based on the detection result of the detection device; When the behavior detection unit detects a behavior in which the detection target moves away from the vehicle body, the behavior detection unit stops one or both of the avoidance driving and the warning operation. The work vehicle according to claim 6.

8. The warning device executes the warning operation, and then the travel control unit executes the avoidance travel. The work vehicle according to claim 6 or 7.

9. The traveling control unit The vehicle body can be decelerated depending on the type of the detection target, When the detection target is an animal other than a human, the running speed after deceleration is higher or the rate of deceleration is smaller than when the detection target is a human. The work vehicle according to claim 1 or 2.

Images