Automated driving support system

Abstract

To enable stable, automatic movement of a work vehicle equipped with work devices. [Solution] The automatic driving support system includes an input device that inputs vehicle information relating to a work vehicle having a running device for moving the vehicle body and capable of being equipped with a work device for performing work, target information relating to a target point in a predetermined area, and area information relating to obstacles present in the area; an information processing device that creates a cost map of the area showing the distribution of costs, setting a first predetermined value as the degree of difficulty of passage for the vehicle body, running device, and work device attached to the work vehicle at locations where obstacles exist, and setting costs smaller than the first predetermined value at locations without obstacles, such that the cost decreases as the distance from the obstacle increases; and a vehicle control device that controls the driving of the work vehicle so that the vehicle body, running device, and work device pass through locations where a second predetermined value or less, which is smaller than the first predetermined value, is set, and executes automatic driving of the work vehicle toward the target point.

Description

Technical Field

[0001] The present invention relates to an automatic driving support system that supports automatic driving while avoiding collisions between a work vehicle and obstacles.

Background Art

[0002] As a system for supporting automatic driving (autonomous driving) of a work vehicle, for example, a system disclosed in Patent Document 1 is known. The system disclosed in Patent Document 1 is a movement control system that controls the movement of a moving body (vehicle) to a destination, and includes acquisition means for acquiring teacher data including trajectory data regarding the trajectory actually traveled by the moving body and movement situation information regarding the situation when the moving body moves along the trajectory, a calculation means for calculating a cost function such that the trajectory actually traveled by the moving body has the smallest cost by inverse reinforcement learning based on the teacher data, and a trajectory calculation means for generating a cost map showing a distribution such as a safety margin regarding the distance from an obstacle by inputting at least one of the position, speed, and surrounding information of the moving body into the cost function, and calculating the trajectory having the smallest cost based on the cost map.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Unlike ordinary vehicles (so-called automobiles) as exemplified in Patent Document 1, work vehicles are equipped with work devices (such as implements) for performing tasks. The types and sizes of work devices vary, and they protrude outward from the body of the work vehicle in the width direction and rearward. As a result, when the work vehicle is in motion, the work device may come into contact with or collide with obstacles in the surrounding area, impairing its drivability. However, conventionally, work devices have not been taken into consideration when creating cost maps.

[0005] In view of the above problems, the present invention aims to enable stable automatic driving of a work vehicle equipped with a work device. [Means for solving the problem]

[0006] An automatic driving support system according to one aspect of the present invention includes: an input device for inputting vehicle information relating to a work vehicle having a driving device for moving its body and capable of being equipped with a work device for performing work, target information relating to a target point in a predetermined area, and area information relating to obstacles in the area; an information processing device for creating a cost map of the area showing the distribution of costs, where a first predetermined value is set as the degree of difficulty of passage of the vehicle body, the driving device, and the work device attached to the work vehicle at the location where the obstacle exists, and the cost smaller than the first predetermined value is set at the location where there is no obstacle, with the cost decreasing as the distance from the obstacle increases; and a vehicle control device for controlling the driving of the work vehicle so that the vehicle body, the driving device, and the work device pass through locations where the cost is set to be less than or equal to a second predetermined value which is smaller than the first predetermined value, and executing automatic driving of the work vehicle toward the target point. [Effects of the Invention]

[0007] According to the present invention, it becomes possible to stably and automatically drive a work vehicle equipped with a work device. [Brief explanation of the drawing]

[0008] [Figure 1]This is a schematic diagram of an example of an automated driving assistance system. [Figure 2] This is a side view of an example of a work vehicle. [Figure 3] This is a block diagram showing an example of the electrical configuration of a work vehicle. [Figure 4A] This is a diagram showing an example of the first cost map. [Figure 4B] This figure shows an example of the first cost map and travel route. [Figure 5A] This figure shows an example of a second cost map. [Figure 5B] This figure shows an example of a second cost map and travel route. [Figure 6] This is a front view of an example of a work vehicle and surrounding objects. [Figure 7] This figure shows an example of the cost distribution around static and dynamic obstacles. [Figure 8] This figure shows an example of the cost distribution around a dynamic obstacle. [Figure 9A] A flowchart illustrating an example of the operation of an automated driving assistance system. [Figure 9B] A flowchart illustrating an example of the operation of an automated driving assistance system. [Figure 10] This flowchart shows an example of image generation processing for weeds and unpaved roads. [Figure 11] This is a flowchart illustrating an example of the image generation process for a field. [Figure 12] This is a simplified diagram of an example of a camera image. [Figure 13A] This figure shows an example of generating an annotation image. [Figure 13B] This figure shows an example of generating an annotation image. [Figure 13C] This figure shows an example of generating an annotation image. [Figure 13D] This figure shows an example of generating an annotation image. [Modes for carrying out the invention]

[0009] Hereinafter, embodiments of the present invention will be described while referring to the drawings. For convenience, the same reference numerals are given to the same configurations and corresponding configurations.

[0010] FIG. 1 is a schematic configuration diagram of an example of an automatic driving support system 100. The automatic driving support system 100 is a system that supports the automatic driving (autonomous driving) of the work vehicle 1. The automatic driving support system 100 includes a work vehicle 1, a management server 20, and a terminal device 30. In FIG. 1, one work vehicle 1, one management server 20, and one terminal device 30 are shown respectively, but the number of work vehicles 1, management servers 20, and terminal devices 30 included in the automatic driving support system 100 may be one or two or more respectively.

[0011] The work vehicle 1 is composed of an agricultural machine (also referred to as an autonomous driving type agricultural machine) capable of autonomously traveling and working. In the present embodiment, the work vehicle 1 is composed of a tractor which is an example of an agricultural machine performing agricultural work in a field. The work vehicle 1 may be composed of an agricultural machine other than a tractor, a construction machine, or a work machine that performs work while traveling outside the field.

[0012] The work vehicle 1 includes an information processing device (computer) 11, a vehicle control device 12, and an input device 13. The vehicle control device 12 is composed of an ECU (electronic control unit) including a processor and a memory. The vehicle control device 12 is a controller that controls the operation of each part of the work vehicle 1. The input device 13 is an input interface for inputting information to the work vehicle 1 and the automatic driving support system 100. The input device 13 includes a user interface 13a and a communication device (communication interface) 13b. Various information can be input and output through the user interface 13a. The communication device 13b includes a communication circuit for wireless communication.

[0013] The management server 20 is a server or a computer provided in a management center or a cloud system. The management server 20 includes a processor, a memory, and a larger capacity than the memory The system is equipped with a storage device 21 and a communication device 23. A database 22 is built into the storage device 21. Various types of information are stored in the database 22. The management server 20 can communicate with the information processing device 11, the vehicle control device 12, and the terminal device 30 via the communication device 23 through a wide-area network such as a mobile phone network and the internet. The management server 20 is an input device that can input information to the work vehicle 1.

[0014] Terminal device 30 is a computer used by the administrator or other users of the automated driving support system 100. Terminal device 30 is equipped with a processor, memory, user interface 32, and communication device 33. Various types of information can be input and output via the user interface 32. Terminal device 30 can communicate with the management server 20, information processing device 11, and vehicle control device 12 via a wide area network using the communication device 33. Terminal device 30 is an input device that can input information to the management server 20, the work vehicle 1, and the automated driving support system 100.

[0015] The information processing device 11 and vehicle control device 12 of the work vehicle 1 can communicate with the management server 20 and terminal device 30 via a wide-area network using the communication device 13b. Alternatively, the information processing device 11 and vehicle control device 12 may communicate with the management server 20 and terminal device 30 via a narrow-area network such as a wireless LAN using the communication device 13b.

[0016] Figure 2 is a side view of an example of work vehicle 1. The direction indicated by arrow A1 in Figure 2 is the front of work vehicle 1. The direction indicated by arrow A2 is the rear of work vehicle 1. The direction indicated by arrow Z1 is above work vehicle 1. The direction indicated by arrow Z2 is below work vehicle 1. The directions perpendicular to arrows A1, A2, Z1, and Z2 are the width directions (left and right directions) of work vehicle 1. In Figure 2, the near side is the left side of work vehicle 1, and the far side is the right side of work vehicle 1.

[0017] The work vehicle 1 comprises a body 3, a prime mover 4, a running gear 5, a transmission 6, a braking system 7, and a steering system 8. The running gear 5 is provided on the left and right sides of the body 3, respectively. The running gear 5 supports the body 3 and moves it. In this embodiment, the running gear 5 is a wheeled running gear having front wheels 5F and rear wheels 5R. Alternatively, the running gear 5 may be a tracked running gear, or a composite running gear having wheels and tracks.

[0018] The prime mover 4, transmission 6, and steering system 8 are mounted on the vehicle body 3. The prime mover 4 consists of an engine (diesel engine, gasoline engine) or an electric motor. The transmission 6 consists of an HST (Hydro Static Transmission) or an HMT (Hydro Mechanical Transmission). The transmission 6 changes the propulsion force of the running gear 5 by performing a gear change operation, and also switches the running gear 5 between forward and reverse. The braking system 7 brakes the running gear 5. The steering system 8 steers the vehicle body 3 by changing the direction of the front wheels 5F.

[0019] A cabin 10 is mounted on top of the vehicle body 3. The cabin 10 is equipped with a driver's seat 14 and a steering wheel 8a. The steering wheel 8a is included in the steering system 8. The cabin 10 is also equipped with operating devices such as levers, pedals, and switches for driving and operating the work vehicle 1. The work vehicle 1 is a tractor capable of unmanned automatic driving and work, but it can also be manually driven by an operator seated in the driver's seat 14 operating the operating devices.

[0020] A bonnet 15 is provided at the front of the cabin 10. The bonnet 15 is attached to the vehicle body It is attached to 3. A storage room (not shown) is formed between the bonnet 15 and the vehicle body 3. This storage room houses not only the prime mover 4, but also a cooling fan, radiator, battery, etc. (not shown).

[0021] A coupling device 9 is provided at the rear of the vehicle body 3. The coupling device 9 is composed of, for example, a three-point linkage mechanism, but may also be composed of drawers. A work device 2 is detachably attached to the coupling device 9. That is, the work device 2 is mounted (equipped) at the rear of the vehicle body 3. By connecting the work device 2 to the coupling device 9 and driving the running gear 5, the work vehicle 1 (vehicle body 3) can be moved, and the work device 2 can be moved. The coupling device 9 can also be used to raise and lower the work device 2, and to change the posture (vertical position and orientation) of the work device 2.

[0022] The work apparatus 2 consists of a tilling apparatus for tilling, a fertilizer spreading apparatus for spreading fertilizer, a pesticide spreading apparatus for spraying pesticides, a harvesting apparatus for harvesting, a mowing apparatus for cutting grass, etc., a spreading apparatus for spreading grass, etc., a grass collecting apparatus for collecting grass, etc., and a shaping apparatus for shaping grass, etc.

[0023] The work devices 2 that can be used with the work vehicle 1 include work devices 2 that are wider than the width of the work vehicle 1 (vehicle body 3), towable work devices 2 with wheels, and mounted work devices 2 that are supported by the work vehicle 1 via a coupling device 9. The mounted work devices 2 include work devices 2 that perform work while on the ground and work devices 2 that perform work while not on the ground. When a work device 2 that performs work while on the ground is connected to the coupling device 9, when performing work, the work device 2 is lowered by the coupling device 9 and set to a grounded position. When not performing work, such as when the work vehicle 1 is moving, the work device 2 is raised by the coupling device 9 and set to a non-grounded position (a position somewhat high above the ground).

[0024] Figure 3 is a block diagram showing an example of the electrical configuration of the work vehicle 1. In addition to the configuration described above, the work vehicle 1 is equipped with a position detection device 13c, a sensing device 13d, a sensor unit 16, an operation unit 17, and a storage device 18. The position detection device 13c and the sensing device 13d are included in the input device 13, along with the user interface 13a and the communication device 13b.

[0025] The information processing device 11, vehicle control device 12, input device 13, sensor unit 16, and storage device 18 are electrically connected by an in-vehicle network such as CAN, LIN, or FlexRay, which is built into the work vehicle 1. The operation unit 17 is connected to the vehicle control device 12.

[0026] The internal memory of the information processing device 11 and the vehicle control device 12 includes volatile memory and non-volatile memory. The storage device 18 is composed of a memory drive such as an SSD (Solid State Drive). Various information, data, and software programs for automatically driving the work vehicle 1 are stored in these internal memories and storage devices 18. The information processing device 11 and the vehicle control device 12 also store information in the internal memory or storage device 18 as appropriate.

[0027] The user interface 13a is, for example, a touch panel or tablet-type terminal device (computer) with a display. The user interface 13a is installed near the driver's seat 14 inside the cabin 10 (Figure 2). The information processing device 11 and the vehicle control device 12 output (display) various information about the work vehicle 1 stored in their internal memory via the user interface 13a. The driver of the work vehicle 1 and others also input various information via the user interface 13a. The user interface 13a is both an input device and an output device (output interface).

[0028] The communication device 13b includes not only a communication circuit for wirelessly communicating with the management server 20 and the terminal device 30, but also an interface for communicating with the information processing device 11 and the vehicle control device 12 via the vehicle network of the work vehicle 1. The communication device 13b is both an input device and an output device.

[0029] The position detection device 13c consists of a positioning device including a GNSS receiver and is installed on top of the cabin 10 as shown in Figure 2. The position detection device 13c uses a satellite positioning system to detect its own position (positioning information including latitude and longitude). Specifically, the position detection device 13c receives signals transmitted from positioning satellites (position of the positioning satellite, transmission time, correction information, etc.) and detects its own position based on these signals. The position detection device 13c may also detect a position corrected based on correction signals from a base station (reference station) capable of receiving signals from positioning satellites as its own position.

[0030] Furthermore, the position detection device 13c may have inertial measuring devices (inertial sensors) such as a gyro sensor and an acceleration sensor. In this case, the position detection device 13c may correct the position (latitude, longitude) detected based on the signal received from the positioning satellite using the inertial measuring devices, and detect the corrected position as its own position.

[0031] The information processing device 11 and the vehicle control device 12 consider the position detected by the position detection device 13c to be the position of the vehicle body 3 of the work vehicle 1. That is, the position detection device 13c detects the position of the vehicle body 3 (work vehicle 1). Furthermore, the information processing device 11 and the vehicle control device 12 calculate the positions of the vehicle body 3, the running gear 5, and the work equipment 2, respectively, based on the position detected by the position detection device 13c and pre-stored information (external shape information) indicating the size of the vehicle body 3, the running gear 5, and the work equipment 2 of the work vehicle 1.

[0032] The sensing device 13d includes a laser sensor, an ultrasonic sensor, and a camera (imaging device). The laser sensor, ultrasonic sensor, and camera are installed at appropriate locations such as the front, rear, left and right sides, top, and bottom of the work vehicle 1 to detect the conditions around the work vehicle 1 and the work device 2. At least one of the laser sensor, ultrasonic sensor, and camera may be included in the sensing device 13d. In addition, detection means such as other sensors may be included in the sensing device 13d.

[0033] The sensing device 13d senses the area around the work vehicle 1 using a laser sensor, an ultrasonic sensor, and a camera. Specifically, the sensing device 13d senses within a predetermined first distance range from the work vehicle and detects objects, including obstacles and non-obstacle objects, that exist within that range. The sensing device 13d also calculates the distance to the object from the detection results of the laser sensor and ultrasonic sensor. An obstacle is defined as a tangible or intangible object that hinders the movement of the work vehicle 1. Objects and obstacles include predetermined natural objects, artificial objects, ditches, depressions, people, animals, etc., that occupy space. For example, the ground and pathways are included as objects, regardless of whether they are paved or not. Areas such as fields and ground where the work vehicle 1 is prohibited from driving are included as obstacles. Areas such as fields and ground where the work vehicle 1 is permitted to drive are not included as obstacles.

[0034] The sensor unit 16 includes various sensors installed on various parts of the work vehicle 1 and the work device 2. The information processing device 11 and the vehicle control device 12 determine the operating status of the work vehicle 1 and the work device 2 based on the output signals from the various sensors of the sensor unit 16. The operating status of the work vehicle 1 determined by the information processing device 11 and the vehicle control device 12 includes the drive and stop status of each part of the work vehicle 1, the direction of travel, travel speed, acceleration, and posture (pitch angle, roll angle, yaw angle (direction, orientation)) of the work vehicle 1 (vehicle body 3), and the operating status of the work device 2. This includes the driving and stopping states and postures (at least the vertical position) of each part of the work device 2.

[0035] As another example, the information processing device 11 and the vehicle control device 12 may detect the position of the vehicle body 3 at predetermined intervals using the position detection device 13c, and based on the time-series data of the position of the vehicle body 3, detect (calculate) the direction of travel, travel speed, acceleration, and posture of the work vehicle 1 (vehicle body 3). Alternatively, the information processing device 11 and the vehicle control device 12 may detect the direction of travel, travel speed, acceleration, and posture of the work vehicle 1 from the measurement results of the inertial measuring device of the position detection device 13c. Alternatively, a rotation speed sensor may be provided to detect the rotation speed of the wheels 5F and 5R of the running device 5, or the rotation speed and direction of rotation of the running motor that rolls the wheels 5F and 5R, and the information processing device 11 and the vehicle control device 12 may detect the direction of travel, travel speed, and acceleration of the work vehicle 1 based on the output signal of the rotation speed sensor.

[0036] The operation unit 17 includes multiple operating devices, multiple drive circuits, and multiple actuators for operating the prime mover 4, running gear 5, transmission 6, braking gear 7, steering gear 8, and coupling gear 9, respectively. The multiple operating devices include multiple operating members operated by the driver of the work vehicle 1, and multiple sensors for detecting at least one of the operation status, direction of operation, and amount of operation of the multiple operating members.

[0037] Each of the multiple actuators in the operation unit 17 is a hydraulic actuator, such as a hydraulic motor and a hydraulic cylinder. To operate these multiple hydraulic actuators, the operation unit 17 includes a hydraulic circuit (drive circuit), which includes a hydraulic pump, a hydraulic motor, a hydraulic pilot-operated valve, and an electromagnetic control valve.

[0038] As another example, at least one of the multiple actuators of the operation unit 17 may be an electric actuator such as a servo motor and a servo cylinder. The operation unit 17 includes an electrical circuit (drive circuit) to operate the electric actuator, and this electrical circuit may include semiconductor elements.

[0039] The vehicle control device 12 drives or operates the corresponding prime mover 4, running gear 5, transmission 6, braking system 7, steering system 8, and coupling device 9 by operating a plurality of actuators included in the operation unit 17. The vehicle control device 12 also controls the operation of the prime mover 4, running gear 5, transmission 6, braking system 7, and steering system 8 via the operation unit 17 to drive the work vehicle 1 and control its movement (driving speed and steering). In other words, the vehicle control device 12 controls the operation of the prime mover 4, running gear 5, transmission 6, braking system 7, and steering system 8 via the operation unit 17 to perform automatic driving, which automatically controls the start and stop of the work vehicle 1, changes in driving speed, and steering. The vehicle control device 12 also raises and lowers the work device 2 connected to the coupling device 9 by operating the coupling device 9 with the corresponding actuator.

[0040] The storage device 18 of the work vehicle 1 stores vehicle information relating to the work vehicle 1. The vehicle information includes identification information for the work vehicle 1 and information regarding the vehicle body 3, the running gear 5, and the work device 2. Specifically, the identification information for the work vehicle 1 is, for example, the model number of the work vehicle 1. The information regarding the vehicle body 3 and the running gear 5 includes information indicating the size (external dimensions) of the vehicle body 3, the type and size of the running gear 5, the relative position of the running gear 5 to the vehicle body 3, and the height (vertical position) of the vehicle body 3 and the running gear 5 from the bottom surface of the running gear 5.

[0041] The five types of running gear include wheeled running gear where the landing part is wheels, tracked running gear where the landing part is tracks, and combined running gear where the landing part is both wheels and tracks. If the type of running gear 5 is a wheeled or composite running gear, the information indicating the size of the running gear 5 includes the width and diameter of the wheels and the tread width of the pair of left and right wheels. If the type of running gear 5 is a tracked or composite running gear, the information indicating the size of the running gear 5 includes the size of the track in the three orthogonal axes (front-rear, left-right, and up-down directions of the vehicle body 3) and the distance between the pair of left and right tracks.

[0042] The information regarding the work device 2 included in the vehicle information contains multiple device information entries set for each of the multiple work devices 2 available on the work vehicle 1. Each of the multiple device information entries includes information indicating the type and size (external dimensions) of the work device 2, the relative position of the work device 2 with respect to the vehicle body 3, and the height of the work device 2 from the bottom surface of the running gear 5.

[0043] For example, the operator may directly input vehicle information using the user interface 13a. Alternatively, the operator may input vehicle information using the user interface 32 of the terminal device 30, and then input the vehicle information from the terminal device 30 to the in-vehicle network of the work vehicle 1 via wireless, wired, or storage medium. The information processing device 11 may then store the vehicle information input in this manner in the storage device 18. Furthermore, the information processing device 11 may transmit the input vehicle information to the management server 20 via the communication device 13b, and the management server 20 may receive the vehicle information and store it in the database 22.

[0044] Alternatively, the terminal device 30 may transmit vehicle information to the management server 20 via the communication device 33 using the user interface 32, and the management server 20 may receive the vehicle information via the communication device 23 and store it in the database 22. The information processing device 11 may then communicate with the management server 20 via the communication device 23 to receive (acquire) vehicle information corresponding to the work vehicle 1 via the communication device 23 and store the vehicle information in the storage device 18.

[0045] Furthermore, after any of the work devices 2 are connected to the coupling device 9 of the work vehicle 1, if the operator inputs information indicating the connected work device 2 via the user interface 13a, the information processing device 11 may select the device information corresponding to that information from the vehicle information stored in the storage device 18.

[0046] Alternatively, if the work device 2 is equipped with an electronic control unit including a CPU and memory, the electronic control unit may be connected to the vehicle network of the work vehicle 1 wirelessly or via a wired connection, and identification information of the work device 2 may be input from the electronic control unit to the vehicle network. In such cases, the information processing unit 11 may select device information corresponding to the identification information from the vehicle information stored in the storage device 18. Alternatively, device information of the work device 2 may be input from the electronic control unit to the vehicle network, and the information processing unit 11 may select the device information and store it in the storage device 18.

[0047] The storage device 18 of the work vehicle 1 also stores one or more area information items relating to one or more predetermined areas (regions) on which the work vehicle 1 travels. The area information includes map information of the predetermined area, and this map information includes information about pathways and obstacles (and other objects) that exist in the predetermined area. The pathway information includes information indicating the location (coordinates), size (length, width), and gradient of pathways such as paved roads and unpaved farm roads that exist in the predetermined area. The obstacle information includes information indicating the type, size, location (coordinates), and height of the obstacles from the ground that exist in the predetermined area.

[0048] Area information may, for example, be generated by terminal device 30, stored in database 22 of management server 20, acquired by communication device 13b, and stored in storage device 18 by information processing device 11.

[0049] For example, map information for multiple areas is stored in the database 22 of the management server 20. When an operator specifies an area by inputting information indicating the desired area via the user interface 32 of the terminal device 30, the terminal device 30 communicates with the management server 20 via the communication device 33 to obtain (receive) the map information corresponding to that information from the database 22 via the management server 20, and outputs the obtained map information via the user interface 32. As a result, a map of the desired area is displayed on the display of the user interface 32, and objects representing known paths and obstacles are displayed on the map.

[0050] If the operator does not see an object representing a new passage or obstacle that they are aware of on the map of the area displayed by the user interface 32, the operator inputs information about the new passage or obstacle through the user interface 32. Based on the input information, the terminal device 30 displays an object representing the new passage or obstacle on the map of the area displayed on the user interface 32.

[0051] Alternatively, the operator may set (input) the target location that the work vehicle 1 is heading to on the displayed area map using the user interface 32. In this case, the terminal device 30 will display an object representing the set target location on the area map using the user interface 32.

[0052] Then, when the operator performs a predetermined confirmation operation via the user interface 32, the terminal device 30 incorporates the newly entered information regarding the passage, obstacles, and target locations into the area map information, generates area information including the map information, and transmits this area information to the management server 20 via the communication device 33. The management server 20 stores the area information received from the terminal device 30 in the database 22. After this, the area information may be updated in the terminal device 30 using the same procedure as described above, and the updated area information may be stored in the database 22. The information regarding the target location includes information indicating the location (coordinates) of the target location set in the predetermined area.

[0053] In the work vehicle 1, when the operator inputs information indicating the desired area via the user interface 13a to specify the area, the information processing device 11 communicates with the management server 20 via the communication device 13b to obtain (receive) area information corresponding to the desired area from the management server 20, and stores the obtained area information in the storage device 18.

[0054] Alternatively, area information may be input from the terminal device 30 to the in-vehicle network of the work vehicle 1 wirelessly, via wired, or via a storage medium, without going through the management server 20, and the information processing device 11 may store the area information in the storage device 18. Alternatively, the operator may obtain map information of a desired area from the database 22 using the user interface 13a of the work vehicle 1, the information processing device 11, and the communication device 13b, etc., generate area information, store the area information in the storage device 18, and also store it in the database 22. Furthermore, among the information included in the area information, target information related to the target location may be input by the operator using the user interface 13a of the work vehicle 1.

[0055] Furthermore, the area information may include identification information for the device that generated the area information, identification information for the device to which the area information was updated, and time information such as a timestamp indicating the date and time the area information was generated and the date and time it was updated.

[0056] Furthermore, the area information may include information indicating the planned route of the work vehicle 1 heading towards the target location. For example, the information processing device 11 may plan the planned route based on the area information and include information indicating that planned route in the area information. Alternatively, the terminal device 30 may use the area information to... The terminal device 30 may then plan a route and include information indicating the planned route in the area information, or the information processing device 11 may receive the information indicating the planned route planned by the terminal device 30 and add it to the area information stored in the storage device 18. Alternatively, the operator may plan a route and input information indicating the planned route using either the user interface 32 of the terminal device 30 or the user interface 13a of the work vehicle 1, and either the terminal device 30 or the information processing device 11 may include the information indicating the planned route in the area information.

[0057] When planning a route, for example, a vehicle model, showing a plan view of the work vehicle 1 and the work device 2 attached to the work vehicle 1, is placed on a map of the area, and a route is created so that the vehicle model moves toward the target point while avoiding collisions with obstacles according to predetermined conditions. The predetermined conditions include minimizing at least one of the time and distance to reach the target point. In addition, the predetermined conditions may also include, for example, that the work vehicle 1 travels only in areas where it is permitted to travel. The areas where the work vehicle 1 is permitted to travel are predetermined locations and include not only the paths shown on the area map information, but also land other than paths.

[0058] The starting point of the planned route may be the current position of the work vehicle 1 (vehicle body 3) detected by the position detection device 13c, or it may be a pre-set home position such as the work vehicle 1's garage. The ending point of the planned route is the target point, which is, for example, a field where work is performed by the work vehicle 1. In this embodiment, the predetermined area is an area that includes fields and areas other than fields. The planned route is a route that the work vehicle 1 takes to the target point by traveling through areas other than fields in the area without performing work with the work device 2. The same applies to other routes described later.

[0059] When an operator specifies an area by inputting information indicating the desired area via the user interface 13a, the information processing device 11 and the vehicle control device 12 read the area information and vehicle information corresponding to the input information from the storage device 18, and the information processing device 11 displays a map of the area, showing the target point and the planned route to the target point, via the user interface 13a based on the area information.

[0060] When the operator views a map of the area and performs a predetermined start operation to automatically drive the work vehicle 1 toward the target point using at least one of the user interface 13a and the operating device of the operation unit 17, the vehicle control device 12 starts the automatic driving of the work vehicle 1 based on the planned route.

[0061] More specifically, the vehicle control device 12 controls the operation of the engine 4, running gear 5, transmission 6, braking gear 7, and steering gear 8 via the operation unit 17 based on the planned route and the vehicle body 3 (work vehicle 1) detected by the position detection device 13c, to drive the work vehicle 1, and also controls the movement of the work vehicle 1 (driving speed, steering, etc.) to move along the planned route, thereby performing automatic driving. In addition, the vehicle control device 12 refers to the vehicle information and the sensing results of the sensing device 13d, and performs automatic driving based on this information as well, controlling the driving, stopping, driving speed, and steering of the work vehicle 1 to avoid collisions with obstacles detected by the sensing device 13d.

[0062] Furthermore, when a desired area is specified by the user interface 13a and a map of that area is displayed by the user interface 13a, the operator may set a target point on the area map using the user interface 13a. The information processing device 11 may then create (plan) a route from the current position of the vehicle 3 to the target point, and store the target information related to the target point and the information indicating the planned route in the storage device 18, associating them with the area information.

[0063] As described above, when the operator specifies a desired area and performs a predetermined driving start operation, the information processing device 11 creates a cost map based on vehicle information and area information, which shows the distribution of costs representing the degree of difficulty of passage for the vehicle body 3, running gear 5, and work equipment 2 in the area relative to obstacles. At this time, the information processing device 11 sets a first predetermined value as the cost at the location where an obstacle exists, and sets a cost smaller than the first predetermined value at the location where there is no obstacle, with the cost decreasing as the distance from the obstacle increases. The information processing device 11 also creates a first cost map showing the distribution of costs representing the degree of difficulty of passage for the running gear 5, and a second cost map showing the distribution of costs representing the degree of difficulty of passage for the vehicle body 3 and work equipment 2.

[0064] Figure 4A shows an example of the first cost map MP1. Figure 5A shows an example of the second cost map MP2. More specifically, Figures 4A and 5A show the cost distribution for a portion of the first cost map MP1 and the second cost map MP2 (around work vehicle 1). The same applies to cost maps MP1 and MP2 in Figures 4B and 5B, which will be described later.

[0065] Furthermore, the first cost map MP1 in Figure 4A shows the distribution of costs, which represent the degree of difficulty in the passage of the landing portion 5t of the wheels 5F and 5R of the running gear 5 of the work vehicle 1, as shown in Figure 6. The first cost map MP1 is used to determine where in the area the wheels 5F and 5R should pass without colliding (contacting) with obstacles.

[0066] The second cost map MP2 in Figure 5B shows the distribution of costs, which represent the difficulty of passage for the vehicle body 3 of the work vehicle 1 and the work device 2, which extends laterally beyond the vehicle body 3 as shown in Figure 6. Furthermore, since the vehicle body 3 is not always on the ground and the work device 2 is in a position where it does not land when the work vehicle 1 is automatically traveling toward the target point, the second cost map MP2 shows the distribution of costs, which represent the difficulty of passage for the vehicle body 3 and work device 2 when they are not on the ground. The second cost map MP2 is used to determine where in the area the work device 2 and vehicle body 3 should pass without colliding with obstacles.

[0067] For example, during the automatic driving of the work vehicle 1, the information processing device 11 generates a sensing image of the area around the work vehicle 1 based on sensing data output from a sensing device 13d that senses within a certain distance (e.g., 20 to 50 m) from the work vehicle 1. Sensing data is a type of sensing result and area information. The sensing image is an image generated based on an imaging signal output from a camera included in the sensing device 13d, for example. More specifically, the sensing image is an image of the area in front of the work vehicle 1, based on an imaging signal output from a camera that images the area in front of the work vehicle 1 (direction of travel).

[0068] Furthermore, the information processing device 11 may generate sensing images in the corresponding direction based on imaging signals output from cameras capturing images of at least one of the rear, left, and right sides of the work vehicle 1, in addition to the sensing image of the front of the work vehicle 1. Alternatively, the information processing device 11 may generate an around-view image of the work vehicle 1 as a sensing image based on imaging signals output from multiple cameras capturing images of the front, rear, left, and right sides of the work vehicle 1.

[0069] The information processing device 11 creates a first cost map MP1 and a second cost map MP2 based on the sensing image. In this case, the first cost map MP1 and the second cost map MP2 may be created using AI (artificial intelligence). The AI ​​may be installed in the information processing device 11, or it may be installed in the management server 20 or the terminal device 30. If the AI ​​is installed in the management server 20 or the terminal device 30, the information processing device 11 communicates directly or indirectly with the management server 20 or the terminal device 30 via the communication device 13b. Create the first cost map MP1 and the second cost map MP2.

[0070] For example, the information processing device 11 uses AI to segment the sensing image (camera image) and generate an annotation image. The information processing device 11 then sets costs for the obstacles and their surroundings shown in the annotation image, and creates a first cost map MP1 and a second cost map MP2 based on the set costs and the annotation image. Furthermore, when the information processing device 11 displays (outputs) the first cost map MP1 and the second cost map MP2 on the display of the user interface 13a (and / or user interface 32), it displays the first cost map MP1 and the second cost map MP2 as shown in Figures 4 and 5, and draws a vehicle model M1 representing the work vehicle 1 on the first cost map MP1 and the second cost map MP2.

[0071] As shown in Figures 4 and 5, the information processing device 11 sets a maximum value of "100" as a cost (first predetermined value) at the location where obstacle Q1 exists in area E1. The information processing device 11 also sets a cost of "100" for the prohibited area X1 located in area E1. That is, in cost maps MP1 and MP2, the ranges where a cost of "100" is set indicate the locations of obstacle Q1 and prohibited area X1.

[0072] The prohibited area X1 is an area where entry by the work vehicle 1 is prohibited. Information indicating the location of the prohibited area X1 is included in the area information. For example, the information processing device 11 may identify a field from the sensing results of the sensing device 13d, the sensing image, or the annotation image, and determine that field to be the prohibited area X1. In this case, if a field is set as the target location, the information processing device 11 will not determine that field to be the prohibited area X1.

[0073] The information processing device 11 sets costs ranging from "99" to "0", which are less than "100", around the obstacle Q1 and the prohibited area X1, such that the costs decrease as the distance from the obstacle Q1 and the prohibited area X1 increases. Below Figures 4 and 5, the correspondence between the cost values ​​and their display formats in cost maps MP1 and MP2 (i.e., the legend) is shown.

[0074] As shown in Figures 4 and 5, the information processing device 11 sets a cost of "99" for an inscribed area that is within a first distance from the obstacle Q1 and the prohibited area X1. The inscribed area may be defined as the radius of a circular area set on the work vehicle 1 (running unit 5, body 3, and work device 2) to determine (predict) the possibility of contact between the work vehicle 1 (running unit 5, body 3, and work device 2) and the obstacle Q1, for example, by setting the first distance. The areas where costs of "100" and "99" are set are areas where the work vehicle 1 (running unit 5, body 3, and work device 2) is prohibited from passing.

[0075] Furthermore, the information processing device 11 sets a cost of "98" to "1" according to the distance from obstacle Q1 and prohibited area X1 within a second distance range that is outside the first distance range from obstacle Q1 and prohibited area X1 but is longer than the first distance range. The range in which a cost of "98" to "1" is set is a range for keeping the running device 5, vehicle body 3, and work device 2 away from obstacle Q1 and prohibited area X1, but is also a range in which the running device 5, vehicle body 3, and work device 2 may pass.

[0076] Furthermore, the range in which the cost is set from "98" to "1" is the inflation range (inflation width) for inflating the cost. In this embodiment, the information processing device 11 expands the inflation range based on a predetermined exponential function, but the information processing device 11 may expand the inflation range based on other functions (for example, a linear function). Alternatively, the information processing device 11 may expand the inflation range based on an exponential function corresponding to the type of obstacle Q1 and prohibited area X1.

[0077] The information processing device 11 sets a cost of "0" outside the second distance range from the obstacle Q1 and the prohibited area X1. The range in which a cost of "0" is set is the range in which the running device 5, vehicle body 3, and work device 2 can pass through stably. In other words, the range in which a cost of "0" is set is also the range in which the running device 5, vehicle body 3, and work device 2 may pass through. The cost values ​​and setting ranges for each cost described above are examples and are not limited.

[0078] The information processing device 11 determines specific obstacles that may collide with the vehicle body 3, the running gear 5, and the working gear 2, based not only on the horizontal position of the vehicle body 3, the running gear 5, and the working gear 2 of the work vehicle 1, but also on their height from the ground (vertical position). The information processing device 11 may also consider the cabin 10 mounted on top of the vehicle body 3 as part of the vehicle body 3 when determining specific obstacles. The information processing device 11 then sets costs for the positions where specific obstacles are present and where they are not, as described above.

[0079] As a result, for example, if the height (height from the ground) of obstacle Q1b (Q1) on the passage J1 shown in Figure 4 is lower than the height of the vehicle body 3 and the work device 2, the information processing device 11 determines that obstacle Q1b is a specific obstacle for the traveling device 5. The information processing device 11 then sets a cost of "100" for the location where obstacle Q1b exists in the first cost map MP1, and sets costs of "99" to "1" around obstacle Q1b. Alternatively, if the information processing device 11 does not determine that obstacle Q1b is a specific obstacle for the vehicle body 3 and the work device 2, in the second cost map MP2 in Figure 5, it does not set a cost of "100" indicating the location where obstacle Q1b exists, nor a cost of "99" indicating the inscribed range within a first distance from obstacle Q1b, but sets costs of "98" to "1" for the location where obstacle Q1b exists and its surroundings.

[0080] Furthermore, if, for example, the obstacle Q1c (Q1) shown in Figure 5 is a tree branch protruding into the passage J1 from the side of the passage J1, and is above the running gear 5 and at the same height (height from the ground) as at least one of the vehicle body 3 and the work device 2, the information processing device 11 determines that the obstacle Q1c is a specific obstacle for at least one of the vehicle body 3 and the work device 2. The information processing device 11 then sets a cost of "100" for the location where the obstacle Q1c is located in the second cost map MP2, and sets costs of "99" to "1" around the obstacle Q1c (at least on the passage J1 side). Alternatively, if the information processing device 11 does not determine that the obstacle Q1c is a specific obstacle for the running gear 5, in the first cost map MP1 in Figure 4, it does not set a cost of "100" indicating the location where the obstacle Q1c is located and a cost of "99" indicating the inscribed range, but sets costs of "98" to "1" for the location where the obstacle Q1c is located and its surroundings.

[0081] Furthermore, if the obstacle Q1a (Q1) on the passage J1 shown in Figures 4 and 5 is an object such as another vehicle that has landed and has the same height as the body 3 and work device 2 of the work vehicle 1, the information processing device 11 determines that the obstacle Q1a is a specific obstacle for the running device 5, the body 3, and the work device 2. The information processing device 11 then sets a cost of "100" for the location where the obstacle Q1a is located in the first cost map MP1 and the second cost map MP2, and sets costs of "99" to "1" around the obstacle Q1a. In this case, the location and range for which costs of "100" to "1" are set for the obstacle Q1c in the first cost map MP1 will be the same as the location and range for which costs of "100" to "1" are set for the obstacle Q1c in the second cost map.

[0082] Furthermore, as shown in Figure 6, if an obstacle Q1d (Q1), such as an inclined surface protruding above the work device 2, exists around the work vehicle 1, the information processing device 11 determines that the obstacle Q1d is a specific obstacle to the running device 5, the vehicle body 3, and the work device 2. The information processing device 11 sets a cost of "100" for the location where the obstacle Q1d exists in the first cost map MP1 and the second cost map MP2, and sets costs of "99" to "1" around the obstacle Q1d. In this case, the location where costs of "100" to "1" corresponding to the obstacle Q1d in the first cost map MP1 are set is closer to the work vehicle 1 than the location where costs of "100" to "1" corresponding to the obstacle Q1d in the second cost map MP2 are set, and the range where costs of "100" to "1" corresponding to the obstacle Q1d in the first cost map MP1 are set is wider than the range where costs of "100" to "1" corresponding to the obstacle Q1d in the second cost map MP2 are set.

[0083] Furthermore, if an obstacle Q1e (Q1), such as a depression, ditch, or cliff, is located below the running gear 5 and exists around the work vehicle 1, the information processing device 11 determines that the obstacle Q1e is a specific obstacle for the running gear 5, but not for the vehicle body 3 or the work gear 2. Then, in the first cost map MP1, the information processing device 11 sets a cost of "100" at the location where the obstacle Q1e exists (including the side of the obstacle Q1e that slopes downwards), and sets costs of "99" to "1" around the obstacle Q1e. In addition, in the second cost map MP2, the information processing device 11 does not set costs of "100" and "99" at the location where the obstacle Q1e exists, nor does it set costs of "98" to "1" around the obstacle Q1e.

[0084] The information processing device 11 may determine, based on area information, whether the obstacle Q1 is a movable dynamic obstacle Q1y or an immovable static obstacle Q1z. Furthermore, as shown in Figure 7, the information processing device 11 may set the cost of a predetermined area around a dynamic obstacle Q1y to be higher than the cost of a predetermined area around a static obstacle Q1z.

[0085] For example, if the type of obstacle Q1 indicated by the area information, or the type of obstacle (object) Q1 determined from the sensing results of the sensing device 13d, is another vehicle, a person, or an animal, the information processing device 11 determines that the obstacle Q1 is a dynamic obstacle Q1y. Also, if the type of obstacle Q1 is an object that does not move on its own or real estate such as a natural object, the information processing device 11 determines that the obstacle Q1 is a static obstacle Q1z.

[0086] The information processing device 11 sets a cost of "100" for the location where the static obstacle Q1z exists and for the location where the dynamic obstacle Q1y exists. The information processing device 11 also sets a cost of "99" to "86" within a range (inscribed range) of a first distance d1 from the static obstacle Q1z, and a cost of "85" to "1" within a range of a second distance d2 that is outside the range of the first distance d1 and is longer than the first distance d1. The information processing device 11 also sets a cost of "99" to "86" within a range of a second distance d2 from the dynamic obstacle Q1y, and a cost of "85" to "1" within a range of a third distance d3 that is outside the range of the second distance d2 and is longer than the second distance d2.

[0087] As a result, the cost set within the range from a first distance d1 to a second distance d2 around the dynamic obstacle Q1y becomes greater than or equal to the cost set within the range from a first distance d1 to a second distance d2 around the static obstacle Q1z. In addition, the inflation range of costs "99" to "1" corresponding to the dynamic obstacle Q1y becomes wider than the inflation range of costs "99" to "1" corresponding to the static obstacle Q1z.

[0088] Furthermore, the information processing device 11 predicts the direction of movement F1 of the dynamic obstacle Q1y based on the sensing results of the sensing device 13d, and may set the cost of the portion in the direction of movement F1 higher than the cost of the portion in the direction of movement F1 within a predetermined range around the dynamic obstacle Q1y that is not in the direction of movement F1. In the example of Figure 8, the information processing device 11 predicts the direction of movement F1 within the range from the second distance d2 to the third distance d3 around the dynamic obstacle Q1y, and sets the cost of the portion in the direction of movement F1 higher. A cost of "98" is set for the portion R2 that is not in the direction of movement F1, and a cost of "97" to "1" is set for the portion R2 that is not in the direction of movement F1. In addition, the information processing device 11 may monitor (detect) at least one of the orientation and position of the dynamic obstacle Q1y from the sensing results, and predict the direction of movement F1 of the dynamic obstacle Q1y from the time-series data of the monitored results.

[0089] The information processing device 11 may, while the work vehicle 1 is performing automatic driving, determine, based on sensing data output from the sensing device 13d, the type of object different from the obstacle Q1 present around the work vehicle 1, and the distance from that object to the obstacle Q1 indicated by the area information. The information processing device 11 may then set a cost for the object according to the type of object and the distance to the obstacle Q1. For example, object cost information indicating the cost value to be set for each type of object is stored in the storage device 18. After determining the type of object and the distance from the object to the obstacle Q1, if the distance is less than or equal to a predetermined value, the information processing device 11 sets the cost value corresponding to the type of object at the location where the object is located.

[0090] Furthermore, if the information processing device 11 detects the ground as an object based on the sensing data output from the sensing device 13d, it may determine at least one of the following from the map of area E1 included in the area information and the sensing data: whether the ground is paved, the degree of the ground's slope, and the direction of the ground's slope. Based on the result of this determination and the distance from the detected ground to the obstacle Q1, the information processing device 11 may set a cost for the ground from "0" to "99" as the distance from the ground to the obstacle Q1 decreases. Also, if the ground is not paved, the information processing device 11 sets a cost for the ground from a third predetermined value greater than "0" (for example, cost "1" in Figures 4A and 4B) to "99" as the distance from the ground to the obstacle Q1 decreases.

[0091] Furthermore, if the ground detected as an object is a flat surface without incline, the information processing device 11 sets a cost of "0" to "99" for the ground as the distance from the ground to the obstacle Q1 decreases. Also, if the ground is incline, the information processing device 11 calculates the degree of incline of the ground based on the sensing data, sets a cost of "1" to "99" for the ground as the degree of incline increases, and significantly changes the cost set for the ground as the distance from the ground to the obstacle Q1 decreases.

[0092] Furthermore, if the ground detected as an object is a downhill slope that slopes downwards as it moves away from the work vehicle 1, the information processing device 11 may set a cost of "1" to "98" for the downhill slope. Also, if the ground is an uphill slope that slopes upwards as it moves away from the work vehicle 1, the information processing device 11 may set a cost of any value from a fourth predetermined value greater than "1" to "98" for the uphill slope.

[0093] Furthermore, the information processing device 11 may change the cost set for the ground as described above, depending on whether the running device 5 has tracks or not. For example, when a running device with tracks travels on paved ground (road), its drivability, such as speed, is reduced compared to when a running device 5 with only wheels and no tracks travels on paved ground. Also, when a running device 5 with only wheels and no tracks travels on unpaved ground, its drivability, such as propulsion, is reduced compared to when a running device with tracks travels on unpaved ground.

[0094] Therefore, the information processing device 11 determines whether the running gear 5 has tracks based on the vehicle information and whether the ground is paved based on the area information. If the information processing device 11 determines that the running gear 5 does not have tracks (i.e., it has wheels), it does not change the cost of paved ground, but increases the cost of unpaved ground by a predetermined value. In addition, if the information processing device 11 determines that the running gear 5 has tracks, it may reduce the cost of unpaved ground by a predetermined value without changing the cost of paved ground.

[0095] Furthermore, if traffic regulations stipulate that a vehicle must travel on either the left or right side of the road, the information processing device 11 may set the cost for one side of the road to "0" or greater, and the cost for the other side to a fifth predetermined value greater than "0".

[0096] Furthermore, the information processing device 11 may change the cost set around the obstacle Q1 according to the type of work device 2 attached to the work vehicle 1. For example, based on vehicle information, the information processing device 11 determines whether the work device 2 is a mounted work device that is mounted on the vehicle body 3 via a coupling device 9, or a towed work device that is towed by the vehicle body 3. If the information processing device 11 determines that the work device 2 is a towed work device, it expands the inflation range of the cost "98" to "1" set around the obstacle Q1 by a predetermined amount, and sets a cost with a larger value at the position where the cost "98" to "1" was set before the expansion. Alternatively, if the information processing device 11 determines that the work device 2 is a mounted work device, it does not have to change the cost set around the obstacle Q1, or it may decrease the cost by a predetermined amount.

[0097] Furthermore, the information processing device 11 may create the first cost map and the second cost map, as described above, while the work vehicle 1 is performing automatic driving, but limited to the area around the work vehicle 1 that can be sensed by the sensing device 13d.

[0098] Alternatively, the information processing device 11 may also create a first cost map and a second cost map for other parts of area E1 that cannot be sensed by the sensing device 13d. In this case, the information processing device 11 sets costs for the locations where obstacle Q1 exists and the locations where obstacle Q1 does not exist, based on the vehicle information and area information, as described above, and associates information indicating the values ​​and ranges of the set costs with the map information of area E1 included in the area information to create a first cost map and a second cost map.

[0099] Furthermore, the information processing device 11 may, in addition to vehicle information and area information, determine specific obstacles among the multiple obstacles Q1 indicated in the area information that are located around the work vehicle 1 (within a certain distance range) and have the potential to collide with at least one of the vehicle body 3, the running gear 5, and the work equipment 2, based on the sensing results of the sensing device 13d and the current vehicle body 3 detected by the position detection device 13c. The information processing device 11 may set the cost of the location where the specific obstacle exists to a first predetermined value "100", and set costs "99" to "0" around the specific obstacle so that the cost decreases as the distance from the specific obstacle increases.

[0100] As described above, the information processing device 11 creates cost maps MP1 and MP2 based on area information (including information on obstacles and objects in the area input by the sensing device 13d) while the automatic driving of the work vehicle 1 is in progress (after it has started). The information processing device 11 then generates control information based on the created cost maps MP1 and MP2 and outputs this control information to the vehicle control device 12. Based on this control information, the vehicle control device 12 controls the driving of the work vehicle 1 so that the vehicle body 3, the driving device 5, and the work device 2 pass through locations where the cost is set to be less than or equal to a second predetermined value which is less than the first predetermined value cost "100", and continues the automatic driving of the work vehicle 1 toward the target point.

[0101] The second predetermined value is the upper limit of the cost of moving the vehicle body 3, running gear 5, and work device 2. For example, the second predetermined value is set to a cost of "85", and below the second predetermined value, the cost is... It is set to "85" to "0". This range of costs below the second predetermined value is just an example and is not limited to it. A range of costs smaller than "99" may also be set as below the second predetermined value.

[0102] The control information output by the information processing device 11 may include, for example, information indicating cost maps MP1 and MP2. In this case, the vehicle control device 12 controls the movement of the work vehicle 1 based on the cost maps MP1 and MP2 included in the control information, the position of the vehicle body 3 detected by the position detection device 13c, the position of the target point, and the planned route, so that the vehicle body 3 does not deviate from the planned route as much as possible, and the vehicle body 3, the running device 5, and the work device 2 pass through positions where the cost is set to "85" to "0", thereby executing automatic movement of the work vehicle 1.

[0103] The first cost map MP1 shown in Figure 4B and the second cost map MP2 shown in Figure 5B show the path L2 (shown as a dashed line) along which the vehicle body 3 moves, and the paths L2a and L2b (shown as dashed lines) along which the left and right ends of the work device 2 move, as controlled by the vehicle control device 12. These are drawn on the first cost map MP1 and the second cost map MP2 when the information processing device 11 displays (outputs) the first cost map MP1 and the second cost map MP2 on the display of the user interface 13a (and / or user interface 32). Path L2 is also the path along which the work vehicle 1 travels while avoiding collision with obstacle Q1 (and entry into prohibited area X1). At least one of the vehicle control device 12 and the information processing device 11 calculates paths L2, L2a, and L2b based on cost maps MP1 and MP2, vehicle information, and target information.

[0104] Furthermore, the control information output by the information processing device 11 may include the travel route of the work vehicle 1 in place of or in addition to the cost maps MP1 and MP2. In this case, the information processing device 11 determines (calculates) a travel route L2 that causes the work vehicle 1 (vehicle body 3) to travel toward the target point, based on the cost maps MP1 and MP2, vehicle information, and target information, such that the vehicle body 3, travel device 5, and work device 2 pass through positions where the cost (cost "85" to "0") is set to be less than or equal to a second predetermined value. In Figure 4B, etc., the direction in which the target point is located is indicated by the arrow Y1.

[0105] Furthermore, the information processing device 11 determines the travel path L2 such that the sum of multiple costs set at multiple locations through which at least one of the vehicle body 3, the running gear 5, and the work device 2 passes before the work vehicle 1 reaches the target point approaches a predetermined target value. The target value may be set to a low first target value so that the vehicle body 3, the running gear 5, and the work device 2 do not approach the obstacle Q1. Alternatively, the target value may be set to a higher second target value, greater than the first target value, so that the vehicle body 3, the running gear 5, and the work device 2 do not move too far from the obstacle Q1 and reduce the travel efficiency of the work vehicle 1. In addition, instead of the information processing device 11, the vehicle control device 12 may determine the travel path L2 as described above, and in this case as well, the target value may be set as described above.

[0106] The information processing device 11 outputs control information indicating the travel route L2 determined as described above to the vehicle control device 12. The vehicle control device 12 controls the movement of the work vehicle 1 based on the travel route L2 indicated by the control information and performs (continues) automatic driving.

[0107] The travel path L2 is a path for the work vehicle 1 (vehicle body 3, running gear 5, and work device 2) to avoid collision with obstacle Q1, and is a path that replaces a part of the planned path L1 (shown as a dashed line) as shown in Figures 4B and 5B. For this reason, after the vehicle control device 12 performs automatic driving of the work vehicle 1 based on the travel path L2, it performs (continues) automatic driving of the work vehicle 1 based on the planned path L1.

[0108] Figures 9A and 9B are flowcharts illustrating an example of the operation of the automated driving support system 100, showing the operation of the information processing device 11 and the vehicle control device 12 described above. Each step in Figures 9A and 9B is executed by at least one of the information processing device 11 and the vehicle control device 12 according to a software program stored in the internal memory of the information processing device 11 and the vehicle control device 12.

[0109] Information indicating one of the areas is input via the user interface 13a (and / or user interface 32, hereinafter the same) of the work vehicle 1, thereby specifying the area (S1 in Figure 9A). Then, the information processing device 11 and the vehicle control device 12 read the vehicle information and the area information corresponding to the input information from the storage device 18 (S2). Then, the information processing device 11 displays a map of the area via the user interface 13a based on the area information (S3). At this time, the roads and obstacles Q1 in the area, the target point set in the area, and the planned route L1 to the target point are displayed on the map of the area.

[0110] Next, a predetermined driving start operation to automatically drive the work vehicle 1 toward the target point is performed by at least one of the user interface 13a and the operating device of the operation unit 17 (S4). Then, the vehicle control device 12 starts the automatic driving of the work vehicle 1 based on the planned route L1, etc. (including the position of the vehicle body 3 detected by the position detection device 13c) (S5). The information processing device 11 also creates a first cost map MP1 and a second cost map MP2 around the work vehicle 1 (S6). The information processing device 11 may also display (output) the created first cost map MP1 and second cost map MP2 on the display of the user interface 13a.

[0111] Then, if there are no obstacles Q1 on the planned route L1 in the direction of travel of the work vehicle 1 or in the vicinity of the planned route L1 (within a range of a second distance from the planned route L1) (S7:NO in Figure 9B), the vehicle control device 12 continues (executes) the automatic driving of the work vehicle 1 based on the planned route L1 (S10).

[0112] On the other hand, if an obstacle Q1 exists on or near the planned route L1 in the direction of travel of the work vehicle 1 (S7: YES), the information processing device 11 and the vehicle control device 12 refer to the first cost map MP1 and the second cost map MP2 to determine whether there is a location where the cost is set to be less than or equal to a second predetermined value, allowing the vehicle body 3, running gear 5, and work device 2 to pass.

[0113] At this time, if there is a location where the cost is set to be less than or equal to a second predetermined value and the vehicle body 3, running gear 5, and work device 2 can pass through (S8: YES), at least one of the information processing device 11 and the vehicle control device 12 determines (calculates) a travel path L2 that will take the vehicle body 3, running gear 5, and work device 2 to the location where the cost is set to be less than or equal to the second predetermined value (S9). Then, the vehicle control device 12 controls the movement of the work vehicle 1 based on the travel path L2, etc., and continues (executes) automatic driving (S10). The information processing device 11 may also draw the travel path L2 on cost maps MP1 and MP2 displayed on the user interface 13a display.

[0114] Subsequently, if the vehicle body 3 has not yet reached the target point (S14:NO), and the predetermined stop operation to stop the automatic movement of the work vehicle 1 has not been performed (S15:NO), the information processing device 11 recreates the first cost map MP1 and the second cost map MP2 around the work vehicle 1 (S6 in Figure 9A), and repeats the subsequent steps in Figure 9B.

[0115] Furthermore, if there is an obstruction on or near the planned route L1 in the direction of travel of the work vehicle 1 If an obstacle Q1 exists (S7: YES), and there is another obstacle Q1 or a prohibited area X1 near the obstacle Q1, the information processing device 11 and the vehicle control device 12 determine that there is no location where the cost is set to be less than or equal to a second predetermined value and the vehicle body 3, running gear 5, and work device 2 can pass (S8: NO). In this case, the information processing device 11 further determines whether there is a location around the obstacle Q1 on or near the planned route L1 where the cost can be corrected.

[0116] The information processing device 11 refers to the first cost map MP1 and the second cost map MP2 and determines that there is a location where the cost can be corrected if a cost less than "99" (and cost "100") is set around an obstacle Q1 located on or near the planned route L1 (S11: YES), and corrects the cost set at that location to a smaller value (S12). At this time, the information processing device 11 readjusts the cost to a second predetermined value (cost "85") or a cost smaller than the second predetermined value at locations around obstacle Q1 located on or near the planned route L1 where a cost of "98" to "86" is set in the first cost map MP1 and the second cost map MP2.

[0117] After this, the information processing device 11 determines again whether there is a position where the cost is set to be less than or equal to a second predetermined value, which allows the vehicle body 3, the running gear 5, and the work device 2 to pass through. If such a position exists (S8: YES), the steps from step S9 onward are executed.

[0118] Furthermore, if, even after cost correction, there is no position where the cost is set to be less than or equal to the second predetermined value that would allow the vehicle body 3, running gear 5, and work device 2 to pass (S8: NO), the information processing device 11 re-determines whether there is a position where the cost can be corrected around the obstacle Q1 located on or near the planned route L1. In this case, if there is no position where the cost can be corrected if there is no position where the cost is less than "99" and greater than the second predetermined value ("98" to "86") located on or near the planned route L1 and around the obstacle Q1, it is determined that there is no position where the cost can be corrected (S11: NO).

[0119] The information processing device 11 then outputs error information via the user interface 13a indicating that the work vehicle 1 cannot move due to the presence of an obstacle Q1 (S13). At this time, the information processing device 11 may also display a message indicating the error information on the display of the user interface 13a. Alternatively, the information processing device 11 may transmit the error information to the terminal device 30 via the communication device 13b, causing the terminal device 30 to display a message indicating the error information on the display of the user interface 32. The information processing device 11 also outputs a command to stop automatic driving to the vehicle control device 12, and the vehicle control device 12 stops the automatic driving of the work vehicle 1 in accordance with the command (S16).

[0120] Furthermore, after step S10, when the position of the vehicle body 3 reaches the target point Pg (S14:YES), the information processing device 11 outputs a stop command, and the vehicle control device 12 stops the automatic driving of the work vehicle 1 in accordance with the stop command (S16). Also, if a predetermined driving stop operation is performed by at least one of the user interface 13a and the operation device of the operation unit 17 before the position of the vehicle body 3 reaches the target point Pg (S14:NO) (S15:YES), the information processing device 11 outputs a stop command, and the vehicle control device 12 stops the automatic driving of the work vehicle 1 (S16).

[0121] The above-described embodiment shows an example in which cost maps MP1 and MP2 are created while the work vehicle 1 is automatically driving, but it is not limited to this. The information processing device 11 may create cost maps MP1 and MP2 before the start of the automatic driving of the work vehicle 1. Alternatively, the information processing device 11 may plan (determine) the driving route L2 based on cost maps MP1 and MP2 and vehicle information before the start of the automatic driving of the work vehicle 1.

[0122] When annotation images that form the basis of cost maps MP1 and MP2 are generated by AI based on images captured by the camera included in the sensing device 13d (sensing images), if the number of data (number of images) used to train the AI ​​is small, the field recognition rate and the accuracy of the annotation images will be low. In particular, weeds and unpaved roads (unpaved pathways and roads) surrounding the field may be mistakenly identified as fields by the AI. To address this, for example, the information processing device 11 generates images of many fields including weeds and unpaved roads from many other images and uses them to train the AI.

[0123] To this end, images (textures) of weeds and unpaved roads are first prepared. Figure 10 is a flowchart showing an example of the image generation process for weeds and unpaved roads. In this embodiment, each step in Figure 10 is executed by an AI provided in, for example, an information processing device 11, according to a predetermined software program, but it may also be executed by an AI or processor such as a CPU provided in another device. The same applies to each step in the flowchart of Figure 11, which will be described later.

[0124] When camera image data and annotation image data pre-generated from the camera image are input from an open dataset or custom dataset (S21 in Figure 10), the AI ​​plays back the camera image and annotation image. The AI ​​then searches the annotation image data for weed labels indicating weeds and unpaved road labels indicating unpaved roads.

[0125] If data for a weed label or an unpaved road label exists in the annotation image data (S22:YES), the AI ​​detects the size (number of pixels) of each side of the weed or unpaved road texture in the annotation image corresponding to the label (S23). If the size of at least one side of the texture is greater than or equal to a predetermined value (predetermined pixels) (S24:YES), the AI ​​detects the drawing position of the texture in the annotation image, extracts an image from the camera image at the same position as the drawing position (S25), and saves the extracted image as the weed or unpaved road texture in a predetermined storage device (e.g., storage device 18) (S26).

[0126] On the other hand, if the annotation image data does not contain data for weed labels and unpaved road labels (S22: NO), the AI ​​will not save the weed or unpaved road textures based on the camera image. Also, even if the annotation image data contains data for weed labels or unpaved road labels (S22: YES), if the size of each side of the texture corresponding to that label is less than a predetermined value (S24: NO), the AI ​​will not save the weed or unpaved road textures based on the camera image.

[0127] As described above, after the textures of weeds and unpaved roads are prepared, an image of the field, including weeds and unpaved roads, is generated from other images, similar to the image captured by the camera included in the sensing device 13d of the work vehicle 1. Figure 11 is a flowchart of an example of the field image generation process. Figure 12 is a simplified diagram of an example of a camera image. Figures 13A to 13D show examples of annotation image generation.

[0128] When data from an open dataset or custom dataset, such as a camera image with a central passage J1 as shown in Figure 12, and data from an annotation image pre-generated from that camera image are input (S31 in Figure 11), the AI ​​plays back the camera image and the annotation image. The reason why camera image data with a central passage J1 as shown in Figure 12 is input is because it is similar to the image captured by the camera included in the sensing device 13d while the work vehicle 1 is in motion. In other words, the passage is also captured in the center of the image captured by the camera included in the sensing device 13d. Note that the camera image in Figure 12 The image shows weed-covered areas G1 and other land H3 on both sides of the pathway J1, but a camera image in which weed-covered areas G1 are not visible may also be input. In addition, land H3 may be a field, or land other than a field where structures exist.

[0129] The AI ​​searches for road label data from the input annotation image data. In this process, the AI ​​treats both the paved road label indicating a paved path (road) J1 and the unpaved road label indicating an unpaved path J1 as road labels.

[0130] If road label data exists in the annotation image data (S32: YES), the AI ​​detects the location of the road (pathway) in the annotation image corresponding to the road label and assigns a predetermined road mask K1 to the corresponding location in the camera image corresponding to that location (S33 in Figure 11, Figure 13A).

[0131] The AI ​​then checks whether a predetermined obstacle exists in the left range of the camera image, up to a position moved a predetermined number of pixels to the left from the road mask, and in the right range, up to a position moved a predetermined number of pixels to the right from the road mask. If no obstacle exists (S34: NO in Figure 11), a predetermined field mask K2 is applied to the left and right ranges (S35). If an obstacle exists in at least one of the left or right ranges (S34: YES), a predetermined weed mask K3 is applied to the position in the left and right ranges where no obstacles exist (S36, Figure 13B).

[0132] After step S35 or step S36 in Figure 11, the AI ​​checks whether there are obstacles in the leftmost and rightmost ranges of the camera image, respectively, from the applied mask to the leftmost and rightmost ends. If there are no obstacles (S37: NO), the AI ​​applies the field mask K2 to the leftmost and rightmost ranges (S38, Figure 13C). If there are obstacles in at least one of the leftmost and rightmost ranges (S37: YES in Figure 11), the AI ​​applies the field mask K2 to the position where there are no obstacles in the leftmost and rightmost ranges (S39). In other words, in steps S34 to S39, the AI ​​considers the parts on both the left and right sides of the path J1 captured in the camera image as fields or weed-covered areas.

[0133] The AI ​​then applies texture images corresponding to each mask K1 to K3 attached to the camera image (S40). At this time, if the label corresponding to the road mask is an unpaved road label, the AI ​​applies a predetermined unpaved road texture image to the road mask, and if the label corresponding to the road mask is a paved road label, the AI ​​applies a predetermined paved road texture image to the road mask. In the example shown in Figure 13D, the paved road texture image T1 is applied to the road mask K1. The AI ​​also applies a predetermined weed texture image T3 to the weed mask K3 attached to the camera image, and a predetermined field texture image T2 to the field mask K2. Furthermore, if an obstacle is present, the AI ​​applies an obstacle texture image corresponding to the obstacle label indicating the obstacle to the location where the obstacle is present.

[0134] The AI ​​then converts the camera image, to which texture images T1 to T3 corresponding to each mask K1 to K3 have been applied as described above, into an annotation image (as shown in Figure 13D), and saves the data of the annotation image to a predetermined storage device (S41 in Figure 11). This generates an annotation image in which a weed field is interposed between the pathway (paved road, unpaved road) and the field. In other words, a sample of the annotation image to be generated (an image for AI training) is created from the image captured by the camera included in the sensing device 13d of the work vehicle 1. Note that in the annotation image shown in Figure 13D, corresponding texture images have also been applied to objects other than the field, weeds, and paved road.

[0135] After this, the AI ​​uses the texture image data saved in the generation process of Figure 10 and the raw data of Figure 11. Based on the annotation image data saved during the processing, subsequent camera images are converted into annotation images to learn the field image. Alternatively, the operator may compare the annotation image saved by the AI ​​with the corresponding camera image to correct the positions of unpaved roads, weeds, and field texture images in the annotation image, and overwrite the memory device with the corrected annotation image.

[0136] In the embodiment described above, an example was shown in which the information processing device 11 and the vehicle control device 12 are installed on the work vehicle 1, but the invention is not limited to this. The information processing device 11 and the vehicle control device 12 may be installed on the management server 20, or on a remote device (terminal device) that remotely operates or monitors the work vehicle 1, or on a computer installed elsewhere than the work vehicle 1, such as a terminal device 30. In other words, the information processing device 11 and the vehicle control device 12 may be installed on at least one of the work vehicle 1, a server that can communicate with the work vehicle 1, and a terminal device. Furthermore, the information processing device 11 and the vehicle control device 12 may be installed on separate devices or machines.

[0137] Furthermore, input devices for inputting vehicle information and area information (user interfaces 13a, 32, communication device 13b, position detection device 13c, and sensing device 13d) may be provided on the work vehicle 1, management server 20, terminal device 30, remote device, and other devices or machines. Alternatively, a single computer may be configured to operate as both the information processing device 11 and the vehicle control device 12. The management server 20 and terminal device 30 may also be omitted from the automated driving support system 100.

[0138] The automated driving support system 100 of this embodiment, as described above, has the following configuration and achieves the following effects.

[0139] [Item 1] The automatic driving support system 100 of this embodiment includes input devices 13a, 32, 13b, 13c, and 13d (user interfaces 13a, 32, communication device 13b, position detection device 13c, sensing device 13d) that input vehicle information relating to a work vehicle 1 having a driving device 5 for moving the vehicle body 3 and capable of being fitted with a work device 2 for performing work, target information relating to a target point in a predetermined area E1, and area information relating to an obstacle Q1 if present in area E1, and the degree of difficulty of passage for the vehicle body 3, the driving device 5, and the work device 2 attached to the work vehicle 1. The system includes an information processing device 11 that creates cost maps MP1 and MP2 of area E1 showing the distribution of costs, by setting a first predetermined value as a cost at the location where an obstacle Q1 exists, and setting costs smaller than the first predetermined value at locations where there is no obstacle Q1, such that the cost decreases as the distance from obstacle Q1 increases; and a vehicle control device 12 that controls the movement of the work vehicle 1 so that the vehicle body 3, running gear 5, and work device 2 pass through locations where a second predetermined value or less, which is smaller than the first predetermined value, is set, and executes automatic movement of the work vehicle 1 toward the target point.

[0140] According to the configuration described in item 1 above, cost maps MP1 and MP2 are created that show the distribution of costs, which represent the degree of difficulty of passage for the vehicle body 3, the running gear 5, and the work device 2. Based on these cost maps MP1 and MP2, the movement of the work vehicle 1 is controlled so that it moves toward the target point while avoiding collisions with obstacle Q1, and automatic driving is performed. As a result, the work vehicle 1 equipped with the work device 2 can be driven automatically and stably.

[0141] [Item 2] In the automatic driving support system 100 described in Item 1 above, the information processing device 11 creates cost maps MP1 and MP2 based on area information input by input devices 13a to 13d and 32 at at least one of the following times: before the start of automatic driving of the work vehicle 1 and during its execution, and outputs control information based on the cost maps MP1 and MP2 to control the vehicle. The device 12 controls the movement of the work vehicle 1 based on the control information to perform automatic driving.

[0142] According to the configuration described in item 2 above, cost maps MP1 and MP2 are created at least either before the start of automatic driving of the work vehicle 1 or during its execution, based on the presence of obstacles Q1. The driving of the work vehicle 1 is then controlled based on these cost maps MP1 and MP2 to perform automatic driving. As a result, the work vehicle 1 can be driven automatically in a stable manner while more reliably avoiding collisions between the vehicle body 3, the driving device 5, and the work device 2 with obstacles Q1 in the surrounding area.

[0143] [Item 3] In the automatic driving support system 100 described in Item 2 above, the information processing device 11 determines a driving path L2 for the work vehicle 1 to travel toward a target point, based on cost maps MP1 and MP2, vehicle information and target information, such that the vehicle body 3, driving device 5 and work device 2 pass through positions where the cost is set to be less than or equal to a second predetermined value, and outputs control information indicating the driving path L2. The vehicle control device 12 controls the driving of the work vehicle 1 based on the driving path L2 to perform automatic driving.

[0144] According to the configuration described in item 3 above, the work vehicle 1 automatically travels along the travel path L2, and the travel device 5, vehicle body 3, and work device 2 can avoid colliding with obstacle Q1 with ample margin, reducing the frequency of sudden evasive maneuvers such as steering, deceleration, and stopping to avoid such collisions. As a result, the work vehicle 1 can be automatically driven stably and efficiently toward the target point.

[0145] [Item 4] In the automated driving support system 100 described in Item 3 above, the information processing device 11 determines the driving path L2 such that the sum of multiple costs set at multiple locations through which at least one of the vehicle body 3, the driving device 5, and the working device 2 passes before the work vehicle 1 reaches the target point approaches the target value.

[0146] According to the configuration described in item 4 above, the travel path L2 of the work vehicle 1 can be appropriately determined while balancing the stable travel of the work vehicle 1, which prevents the vehicle body 3, running gear 5, and work device 2 from colliding with the obstacle Q1, with the efficient travel of the work vehicle 1, which prevents the vehicle body 3, running gear 5, and work device 2 from moving unnecessarily far from the obstacle Q1.

[0147] [Item 5] In the automated driving support system 100 described in any of the above items 1 to 4, the vehicle information includes information indicating the size of the vehicle body 3, the running gear 5, and the work device 2, the relative positions of the running gear 5 and the work device 2 with respect to the vehicle body 3, and the heights of the vehicle body 3, running gear 5, and work device 2 from the bottom surface of the running gear 5, respectively. The area information includes information indicating the location where the obstacle Q1 exists and the height of the obstacle Q1 from the ground, respectively. The information processing device 11 creates cost maps MP1 and MP2, which are the distribution of costs representing the degree of difficulty of passage for the running gear 5 and the distribution of costs representing the degree of difficulty of passage for the vehicle body 3 and the work device 2.

[0148] According to the configuration described in item 5 above, a first cost map MP1 is created that shows the distribution of costs, which is the degree of difficulty of passing obstacle Q1 with respect to the grounded running device 5, in both the planar and vertical directions, and a second cost map MP2 is created that shows the distribution of costs, which is the degree of difficulty of passing obstacle Q1 with respect to the grounded vehicle body 3 and work device 2, as well as the work device 2 that protrudes in the width direction from the vehicle body 3. Therefore, based on the first cost map MP1 and the second cost map MP2, the work vehicle 1 can be driven automatically and stably while more reliably avoiding collisions between the vehicle body 3, running device 5, and work device 2 with surrounding obstacle Q1. Furthermore, for example, if there is an obstacle Q1 that is lower in height from the ground than the work device 2 that protrudes from the width direction from the vehicle body 3, the running device 5 and vehicle body 3 will collide with the obstacle Q1. This allows the work vehicle 1 to be driven so that the work device 2 passes over the obstacle Q1, while avoiding the obstacle, thereby increasing the degree of freedom in the position where the work vehicle 1 can travel.

[0149] [Item 6] In the automatic driving support system 100 described in any of Items 2 to 5 above, the information processing device 11 determines, at least either before the start of automatic driving of the work vehicle 1 or during its execution, a specific obstacle Q1 from among a plurality of obstacles Q1 indicated by area information that may collide with at least one of the vehicle body 3, the driving device 5, and the work device 2, sets the cost of the location where the specific obstacle Q1 is located to a first predetermined value, and sets costs around the specific obstacle Q1 such that the cost decreases as the distance from the specific obstacle Q1 increases.

[0150] According to the configuration described in item 6 above, cost maps MP1 and MP2 are created based on obstacle Q1, limited to obstacle Q1 that may cause collision, at least at either before the start of automatic driving of the work vehicle 1 or during its execution. This improves the reliability of cost maps MP1 and MP2, allowing the work vehicle 1 to be driven automatically more stably and efficiently, and also reduces the processing load on the information processing device 11.

[0151] [Item 7] In the automated driving support system 100 described in any of the above items 1 to 6, the area information includes information indicating the location of a prohibited area X1 that prohibits the entry of a work vehicle 1 located in area E1, and the information processing device 11 sets the cost of the prohibited area X1 to a first predetermined value and sets costs around the prohibited area X1 such that the cost decreases as the distance from the prohibited area X1 increases.

[0152] According to the configuration described in item 7 above, the work vehicle 1 can be automatically and stably driven toward the target point while avoiding the driving device 5, the vehicle body 3, and the work device 2 entering the prohibited area X1.

[0153] [Item 8] In the automatic driving support system 100 described in any of the above items 1 to 7, if there is no position where the vehicle body 3, the driving device 5, and the work device 2 can pass through and where a cost of 2 or less is set, the information processing device 11 corrects the cost of the position around the obstacle Q1, which is set to a value smaller than the 1st predetermined value, to a smaller value.

[0154] According to the configuration described in item 8 above, the work vehicle 1 can be driven automatically and stably by bringing at least one of the vehicle body 3, the running gear 5, and the work device 2 close to the obstacle Q1 without colliding with it, and the degree of freedom of the position in which the work vehicle 1 can travel can also be increased.

[0155] [Item 9] In the automatic driving support system 100 described in Item 8 above, the information processing device 11 outputs error information via the user interfaces 13a and 32 and causes the vehicle control device 12 to stop automatic driving in at least one of the following cases: when the cost cannot be corrected, or when, even after correcting the cost, there is no position where the cost is set to a second predetermined value or less that the vehicle body 3, the driving device 5, and the work device 2 can pass through.

[0156] According to the configuration described in item 9 above, it is possible to reliably prevent at least one of the vehicle body 3, the running gear 5, and the work device 2 from colliding with the obstacle Q1, and to notify the user that the automatic movement of the work vehicle 1 has stopped.

[0157] [Item 10] In the automatic driving support system 100 described in any of the above items 1 to 9, the information processing device 11 determines whether the obstacle Q1 is a movable dynamic obstacle Q1y or an immovable static obstacle Q1z based on area information, and the surroundings of the static obstacle Q1z The cost of a predetermined range around the dynamic obstacle Q1y is set higher than the cost of a predetermined range.

[0158] According to the configuration described in item 10 above, the distance (spacing) between the vehicle body 3, running gear 5, and working device 2 to a dynamic obstacle Q1y can be made longer than the distance (spacing) between the vehicle body 3, running gear 5, and working device 2 to a static obstacle Q1z. Therefore, when the working vehicle 1 is traveling alongside the dynamic obstacle Q1y, even if the dynamic obstacle Q1y suddenly starts moving, it is possible to avoid the dynamic obstacle Q1y colliding with any of the vehicle body 3, running gear 5, or working device 2.

[0159] [Item 11] In the automatic driving support system 100 described in Item 10 above, the input devices 13a to 13d, 32 include a sensing device 13d that senses the area around the work vehicle 1, and the information processing device 11 predicts the direction of movement F1 of the dynamic obstacle Q1y based on the sensing results of the sensing device 13d, and sets the cost of the portion R1 that is in the direction of movement F1 higher than the cost of the portion R2 that is not in the direction of movement F1 within a predetermined range around the dynamic obstacle Q1y.

[0160] According to the configuration described in item 11 above, when the work vehicle 1 is traveling alongside the dynamic obstacle Q1y, even if the dynamic obstacle Q1y suddenly starts moving, it is possible to further avoid the dynamic obstacle Q1y colliding with the vehicle body 3, the running gear 5, or the work device 2.

[0161] [Item 12] In the automatic driving support system 100 described in any of the above items 1 to 10, the input devices 13a to 13d, 32 include a sensing device 13d that senses the surroundings of the work vehicle 1, and the information processing device 11 sets a cost for an object based on the sensing data output from the sensing device 13d during the execution of automatic driving of the work vehicle 1, according to the type of object present around the work vehicle 1 and the distance from the object to the obstacle Q1.

[0162] According to the configuration described in item 12 above, the work vehicle 1 can be driven automatically in a stable and efficient manner, taking into account the objects and obstacles Q1 present around the work vehicle 1.

[0163] [Item 13] In the automated driving support system 100 described in Item 12 above, when the information processing device 11 detects the ground as an object, it determines at least one of the following: whether the ground is paved, the degree of inclination, and the direction of inclination. Based on the result of this determination and the distance from the ground to the obstacle Q1 indicated by the area information, it sets a cost for the ground.

[0164] According to the configuration described in item 13 above, a cost can be set for the ground on which the work vehicle 1 travels, taking into account the ease of travel for the work vehicle 1 and the obstacles Q1 present in the surrounding area. Based on the cost set for the ground, the work vehicle 1 can be driven automatically in a stable and efficient manner.

[0165] [Item 14] In the automated driving support system 100 described in Item 13 above, the information processing device 11 determines whether the driving device 5 has tracks based on vehicle information, determines whether the ground is paved based on area information, and changes the cost set for the ground according to the result of the determination.

[0166] According to the configuration described in item 14 above, for example, if the running gear 5 has tracks, the cost of unpaved ground can be reduced by a predetermined value, and the work vehicle 1 can be guided so that the running gear 5 travels on that ground. Also, if the running gear 5 does not have tracks and only has wheels 5F and 5R, the cost of unpaved ground can be increased by a predetermined value, and the work vehicle 1 can be guided so that the running gear 5 travels on other paved ground. This also makes it possible to improve the autonomous driving performance of work vehicle 1.

[0167] [Item 15] In the automated driving support system 100 described in any of the above items 1 to 14, the work vehicle 1 has a coupling device 9 for connecting the work device 2, and the information processing device 11 determines, based on vehicle information, whether the work device 2 is a mounted work device that is mounted on the vehicle body 3 via the coupling device 9, or a towed work device that is towed by the vehicle body 3, and changes the cost set around the obstacle Q1 according to the result of the determination.

[0168] According to the configuration described in item 15 above, costs can be set around the obstacle Q1 depending on whether the work device 2 connected to the work vehicle 1 is a mounted work device or a towed work device. Furthermore, when the work vehicle 1 travels to the side of the obstacle Q1, the work device 2 can more reliably avoid colliding with the obstacle Q1.

[0169] Although the present invention has been described above, the embodiments disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of the present invention is indicated by the claims rather than by the foregoing description, and all modifications within the meaning and scope equivalent to the claims are intended to be included. [Explanation of symbols]

[0170] 1. Work vehicles 2. Working equipment 3. Vehicle Body 5. Traveling device 9 Coupling device 11 Information Processing Devices 12 Vehicle control system 13a User interface (input device) 13b Communication device (input device) 13c Position detection device (input device) 13b Sensing device (input device) 32. User Interface (Input Device) 100 Automated Driving Assistance Systems MP1 1st Cost Map MP2 Second Cost Map E1 Area F1 Movement direction X1 Prohibited area L2 Driving Route Q1, Q1a~Q1e Obstacles Q1y Dynamic Obstacle Q1z Static Obstacle

Claims

1. An input device for inputting vehicle information relating to a work vehicle having a running gear for moving the vehicle body and capable of being fitted with work equipment for performing work, target information relating to a target point in a predetermined area, and area information relating to obstacles in the said area, An information processing device creates a cost map of the area showing the distribution of costs, where a first predetermined value is set at the location where the obstacle exists, and costs smaller than the first predetermined value are set at locations without obstacles, such that the cost decreases as the distance from the obstacle increases. A vehicle control device that controls the movement of the work vehicle so that the vehicle body, the running gear, and the work device pass through a position where the cost is set to be less than or equal to a second predetermined value which is smaller than the first predetermined value, and executes automatic movement of the work vehicle toward the target point. An automated driving assistance system equipped with [features].

2. The information processing device creates the cost map based on the area information input by the input device at at least one of the following times: before the start of the automatic driving of the work vehicle and during its execution, and outputs control information based on the cost map. The automatic driving support system according to claim 1, wherein the vehicle control device controls the movement of the work vehicle based on the control information to perform the automatic driving.

3. Based on the cost map, the vehicle information, and the target information, the information processing device determines a travel path for the work vehicle toward the target point such that the vehicle body, the running gear, and the work device pass through positions where the cost is set to be less than or equal to the second predetermined value, and outputs the control information indicating the travel path. The automatic driving support system according to claim 2, wherein the vehicle control device controls the movement of the work vehicle based on the travel path to perform the automatic driving.

4. The automated driving support system according to claim 3, wherein the information processing device determines the driving path such that the sum of the multiple costs set at multiple locations through which at least one of the vehicle body, the driving device, and the working device passes before the work vehicle reaches the target point approaches the target value.

5. The vehicle information includes information indicating the size of the vehicle body, the running gear, and the working device, the relative positions of the running gear and the working device with respect to the vehicle body, and the heights of the vehicle body, the running gear, and the working device from the bottom surface of the running gear, respectively. The area information includes information indicating the location where the obstacle exists and the height of the obstacle from the ground, respectively. The automated driving support system according to claim 1, wherein the information processing device creates a cost map, which includes a first cost map showing the distribution of costs representing the degree of difficulty of passage for the driving device, and a second cost map showing the distribution of costs representing the degree of difficulty of passage for the vehicle body and the work device.

6. The automated driving support system according to claim 2, wherein the information processing device determines, at least either before the start of the automated driving of the work vehicle and during its execution, a specific obstacle among the plurality of obstacles indicated in the area information that may collide with at least one of the vehicle body, the driving device, and the work device, sets the cost of the location where the specific obstacle is located to a first predetermined value, and sets the cost around the specific obstacle such that the cost decreases as the distance from the specific obstacle increases.

7. The area information includes information indicating the location of prohibited areas where entry of the work vehicles is prohibited within the area. The automated driving support system according to claim 1, wherein the information processing device sets the cost of the prohibited area to a first predetermined value, and sets the cost around the prohibited area such that the cost decreases as the distance from the prohibited area increases.

8. The automatic driving support system according to claim 1, wherein if there is no position where the vehicle body, the running gear, and the work device can pass and the cost is set to a second predetermined value or less, the information processing device corrects the cost of a position around the obstacle, which is set to a value smaller than the first predetermined value, to a smaller value.

9. The automatic driving support system according to claim 8, wherein the information processing device outputs error information via a user interface and causes the vehicle control device to stop the automatic driving in at least one of the following cases: when the cost cannot be corrected, or when, even after correcting the cost, there is no position where the cost is set to be less than or equal to the second predetermined value that the vehicle body, the running gear, and the work device can pass through.

10. The automated driving support system according to claim 1, wherein the information processing device determines, based on the area information, whether the obstacle is a movable dynamic obstacle or an immovable static obstacle, and sets the cost of a predetermined range around a dynamic obstacle to be higher than the cost of a predetermined range around a static obstacle.

11. The input device includes a sensing device that senses the area around the work vehicle. The automatic driving support system according to claim 10, wherein the information processing device predicts the direction of movement of the dynamic obstacle based on the sensing results of the sensing device, and sets the cost of the portion in the direction of movement to be higher than the cost of the portion in the direction of movement of a predetermined range around the dynamic obstacle that is not in the direction of movement.

12. The input device includes a sensing device that senses the area around the work vehicle. The automated driving support system according to claim 1, wherein the information processing device sets the cost for an object based on sensing data output from the sensing device during the execution of the automated driving of the work vehicle, according to the type of object present around the work vehicle and the distance from the object to the obstacle.

13. The automatic driving support system according to claim 12, wherein when the information processing device detects the ground as the object, it determines at least one of the presence or absence of pavement, the degree of inclination, and the direction of the ground, and sets the cost on the ground according to the result of the determination and the distance from the ground to the obstacle indicated by the area information.

14. The automated driving support system according to claim 13, wherein the information processing device determines whether the running device has tracks based on the vehicle information, determines whether the ground is paved based on the area information, and changes the cost set on the ground according to the result of the determination.

15. The aforementioned work vehicle has a coupling device for connecting the aforementioned work equipment, The information processing device determines, based on the vehicle information, whether the work device is a mounted work device that is mounted on the vehicle body via the coupling device, or a towed work device that is towed by the vehicle body, and changes the cost set around the obstacle according to the result of the determination, as described in claim 1.

Images