Autonomous driving system for work vehicles

Abstract

To provide a technique associated with an autonomous travel system for a farm work vehicle, enabling the farm work vehicle to autonomously travel along a ridge with high precision, regardless of the shape change of the ridge.SOLUTION: An autonomous travel system for a farm work vehicle makes a farm work vehicle 10, which performs farm work by straddling a ridge 5 in a farm field, autonomously travel along the ridge 5. The autonomous travel system comprises: the farm work vehicle 10 comprising a pair of travel parts 20, provided at intervals so as to travel on travel planes 7I, 7II on both sides of one line of the ridge 5; a ridge tracking LiDAR 50 which is installed on the farm work vehicle 10 so as to incorporate the travel planes 7I, 7II on both sides of the ridge 5, straddled by the farm work vehicle 10, into a detection area; and a controller for, on the basis of information on the two lines of the travel planes 7I, 7II, estimating a center line RCL of the ridge 5 straddled by the farm work vehicle 10, and on the basis of an error between the center line RCL of the ridge 5 estimated and a center line of the farm work vehicle 10, making the farm work vehicle 10 travel along the ridge 5.SELECTED DRAWING: Figure 13

Description

[Technical field]

[0001] The present invention relates to an autonomous driving system for a work vehicle, and more particularly to an autonomous driving system for a work vehicle that causes a work vehicle to autonomously drive along furrows. [Background technology]

[0002] Conventionally, various autonomous driving systems have been proposed that allow agricultural vehicles to automatically follow ridges in a field while performing agricultural work by straddling the ridges in order to reduce labor and improve the efficiency of agricultural work. Some of these autonomous driving systems control the positional relationship between the ridges and the planting unit based on the change in position of a pressure-sensitive ridge detection plate that contacts the ridges, but such autonomous driving systems have problems such as the risk of the ridge detection plate destroying or chipping the ridges, and low detection accuracy due to vibrations caused by contact between the ridge detection plate and the ridges.

[0003] For example, Patent Document 1 proposes a transplanter in which a planting unit is equipped on a running vehicle so that it can be freely adjusted to move left and right, an ultrasonic ridge sensor is provided on the planting unit to detect the distance to the transplanting furrow where the seedlings are transplanted, and the left and right movement of the planting unit is controlled based on the output signal of the ridge sensor, allowing the planting unit to automatically follow the transplanting furrow. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 8-280213 Summary of the Invention [Problem to be solved by the invention]

[0005] The technique disclosed in Patent Document 1 uses an ultrasonic ridge sensor that does not come into contact with the ridges, making it possible to prevent the ridges from being destroyed or scraped.

[0006] However, in the system of Patent Document 1, a ridge sensor is used to detect the distance between the slope of the ridge, and if the shape of the ridge is distorted, it becomes difficult to detect the distance from the ridge, resulting in a problem of low robustness (resistance to external disturbances) against changes in the shape of the ridge.

[0007] The present invention has been made in consideration of these points, and its purpose is to provide a technology in an autonomous driving system for a work vehicle that allows the work vehicle to autonomously drive along ridges with high precision, without being affected by changes in the shape of the ridges. [Means for solving the problem]

[0008] In order to achieve the above-mentioned objective, the autonomous driving system for a work vehicle of the present invention estimates a point on the center line of a group of ridges by extracting only the point cloud on a flat surface extending horizontally from a point cloud consisting of multiple points on the ridges and the driving surface.

[0009] Specifically, the present invention is directed to an autonomous driving system for a work vehicle that autonomously drives a work vehicle along a group of ridges in a field having a group of ridges consisting of multiple ridges and a driving surface formed between adjacent ridges, the work vehicle performing work while straddling the group of ridges.

[0010] This autonomous driving system comprises a distance sensor installed on the work vehicle, an estimation means for estimating the center line of the group of ridges straddled by the work vehicle based on information detected by the distance sensor, and a control means for driving the work vehicle along the group of ridges based on the error between the center line of the group of ridges estimated by the estimation means and the center line of the work vehicle, and is configured to estimate a point on the center line of the group of ridges by extracting only the point clouds on flat surfaces extending horizontally from a point cloud consisting of a plurality of points on the driving surface located on both sides of the ridge, a point cloud consisting of a plurality of points on each slope in the inclined portion of the ridge, and a point cloud consisting of a plurality of points on the top surface of the ridge.

[0011] According to this configuration, the estimation means estimates the center line of the ridge group based on information about a flat surface extending horizontally, which is less susceptible to changes in environment, shape, etc. than the ridges themselves, which are prone to significant changes in environment and shape due to the growth of planted crops, etc. Therefore, even if the bottoms of the ridges, etc., are locally collapsed, the center line of the ridge group can be estimated with high accuracy.

[0012] Furthermore, since the centerline of the ridge group is estimated based on information about a flat surface extending horizontally, the error is halved compared to estimating the centerline of the ridge group based on information about one of the running surfaces, thereby further improving the accuracy of estimating the centerline of the ridge group.

[0013] As described above, according to the present invention, the center line of a group of ridges is estimated based on information about a horizontally extending flat surface which is relatively subject to small changes in environment, shape, etc., making it possible to cause an agricultural vehicle to autonomously drive along the ridges with high accuracy without being affected by changes in the shape of the ridges.

[0014] In addition, the estimation means is configured to exclude the point clouds on each of the inclined slopes on both sides of the ridge group from the point clouds on the ridge group and on each of the two running surfaces on either side of the ridge group, and to exclude the point cloud on the top surface of the ridge group by extracting the point clouds that fall within a predetermined height range that is the height near the lower end of the ridge group, and to estimate that the midpoint between the point closest to the ridge group among the point clouds on one of the running surfaces and the point closest to the ridge group among the point clouds on the other running surface is the point on the center line of the ridge group. Effect of the Invention

[0015] As described above, the autonomous driving system for an agricultural work vehicle according to the present invention makes it possible to cause a work vehicle to autonomously drive along ridges with high precision, without being affected by changes in the shape of the ridges. [Brief description of the drawings]

[0016] [Figure 1]FIG. 1 is a diagram illustrating an autonomous driving system for an agricultural work vehicle according to an embodiment of the present invention. [Diagram 2] FIG. 1 is a side view showing a schematic view of an agricultural work vehicle. [Diagram 3] 1 is a plan view showing a schematic diagram of an agricultural work vehicle in a state where the crawler pitch is widened. FIG. [Figure 4] 1 is a plan view that shows a schematic view of an agricultural work vehicle with a narrowed crawler pitch. FIG. [Diagram 5] 4A and 4B are diagrams for explaining a crawler pitch variable structure. [Figure 6] FIG. 2 is a perspective view showing a schematic installation position of a LiDAR in an agricultural work vehicle. [Figure 7] FIG. 13 is a diagram illustrating the detection direction of a ridge tracing LiDAR. [Figure 8] FIG. 2 is a diagram illustrating a detection region of a LiDAR for detecting a marker. [Figure 9] 1A and 1B are diagrams for explaining the detection area of ​​a LiDAR for ridge tracking, in which FIG. 1A is a side view and FIG. 1B is a plan view. [Figure 10] FIG. 13 is a diagram illustrating a method for estimating the center line of a ridge in an autonomous driving system. [Figure 11] FIG. 13 is a diagram illustrating a method for estimating the center line of a ridge in an autonomous driving system. [Figure 12] FIG. 13 is a diagram illustrating a method for estimating the center line of a ridge in an autonomous driving system. [Figure 13] FIG. 13 is a diagram illustrating a method for estimating the center line of a ridge in an autonomous driving system. [Figure 14] FIG. 13 is a diagram illustrating a method for estimating the center line of a ridge in an autonomous driving system. [Figure 15] 1 is a diagram for explaining a schematic relationship between the arrangement position of a ridge tracing LiDAR and a cloud of detectable points. FIG. [Figure 16] FIG. 13 is a diagram illustrating the relationship between the detection direction of the ridge tracing LiDAR when it is directed straight down and when it is directed diagonally downward and forward. [Figure 17]13A to 13C are diagrams illustrating a method for extracting a flat point cloud according to another embodiment. [Figure 18] FIG. 1 is a diagram for explaining a conventional method for estimating the center line of a ridge. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

[0018] -Autonomous Driving System- 1 is a schematic diagram showing an autonomous driving system 1 for an agricultural work vehicle (work vehicle) 10 according to this embodiment. This autonomous driving system 1 allows the agricultural work vehicle 10 to travel autonomously in a farm field 3 having a plurality of ridges 5 and a running surface 7 formed between adjacent ridges 5, 5, and in a headland 9 extending in the arrangement direction of the ridges 5 on both outer sides in the longitudinal direction of the ridges 5, without requiring operation by an operator or the like.

[0019] More specifically, this autonomous driving system 1 causes an agricultural work vehicle 10, which performs agricultural work while straddling ridges 5 in a farm field 3, to autonomously travel along the longitudinal direction of the ridges 5, and also causes the agricultural work vehicle 10 to autonomously exit from the ridges 5 to the headland 9, travel on the headland 9, and enter the ridges 5 from the headland 9. To enable such autonomous traveling, the autonomous driving system 1 of this embodiment includes, as shown in Fig. 1, markers 80 erected at both longitudinal ends 5f of the ridges 5, the agricultural work vehicle 10, a ridge tracing LiDAR 50, a marker detection LiDAR 60 (see Fig. 3, etc.), and a control device 70 (see Fig. 2).

[0020] <marker> The markers 80 are, for example, plain wood stakes, and are driven into both longitudinal ends 5f of every other ridge 5 as shown in Fig. 1. There are no particular limitations on the material or shape of the markers 80. Also, in this embodiment, the markers 80 are set up on every other ridge 5, but as will be described later, in the case where the agricultural work vehicle 10 straddles the markers 80, the markers 80 may be set up on both longitudinal ends 5f of all the ridges 5.

[0021] <Agricultural vehicles> Fig. 2 is a side view showing the agricultural work vehicle 10, Fig. 3 is a plan view showing the agricultural work vehicle 10 with the crawler pitch widened, and Fig. 4 is a plan view showing the agricultural work vehicle 10 with the crawler pitch narrowed. The agricultural work vehicle 10 performs weeding work while straddling ridges 5 in a field 3, and as shown in Figs. 2 to 4, includes a pair of left and right traveling units 20, a frame unit 30, a frame lifting mechanism 36, a weeding attachment 40, an attachment lifting mechanism 45, a ridge tracing LiDAR 50, a marker detection LiDAR 60, and a control device 70.

[0022] The pair of left and right running parts 20 are provided spaced apart from each other in the width direction of the machine body so as to run on the running surfaces 7, 7 adjacent to each other of one ridge 5. As shown in FIG. 2, each running part 20 has a driving wheel 21, a driven wheel 22, two rollers 23, 24, a frame 25, a crawler belt 26, a traveling motor unit 27, and a traveling battery 29. The driving wheel 21, the driven wheel 22, and the two rollers 23, 24 are supported rotatably by the frame 25. The crawler belt 26 is wound around the driving wheel 21, the driven wheel 22, and the two rollers 23, 24 so as to circumscribe them. The traveling motor unit 27 is an integrated unit of a motor and a reducer, and is disposed in an in-wheel shape on the driving wheel 21. The traveling battery 29 is disposed on the frame 25 above the traveling motor unit 27.

[0023] Each traveling unit 20 is configured to travel on the traveling surface 7 by rotating the crawler belt 26 while being guided by the driven wheels 22 and two rollers 23, 24 due to the driving force of the driving wheels 21 (traveling motor unit 27) driven by power supplied from a traveling battery 29. The forward (normal rotation), reverse (reverse rotation) and stop of each traveling unit 20 are controlled by a control device 70.

[0024] As shown in Figs. 3 and 4, the mount 30 is formed in a substantially rectangular shape in plan view and is provided above the pair of travel parts 20 so as to span the pair of travel parts 20. The mount 30 has a pair of left and right vertical frame members 31 extending in the front-rear direction of the machine body, and first to fourth horizontal frame members 32, 33, 34, and 35 connecting the pair of left and right vertical frame members 31 in the width direction of the machine body. As shown in Fig. 2, the pair of left and right vertical frame members 31 are connected to the frames 25 of the pair of left and right travel parts 20 via a mount lifting mechanism 36. This allows the crawler pitch to be changed by changing the width of the mount 30.

[0025] Fig. 5 is a diagram for explaining the crawler pitch variable structure. As shown in Fig. 5, the first to fourth horizontal frame members 32, 33, 34, 35 have a pair of left and right outer pipe sections 32a, 33a, 34a, 35a and inner pipe sections 32b, 33b, 34b, 35b fitted inside the pair of left and right outer pipe sections 32a, 33a, 34a, 35a. Of the first to fourth horizontal frame members 32, 33, 34, 35, the second horizontal frame member 33 is provided with a width variable screw section 33c.

[0026] In this mounting unit 30, when the width-varying screw portion 33c is manually rotated in one direction, the left and right outer pipe portions 33a approach each other in the second horizontal frame member 33, and the inner pipe portion 33b fits inside the left and right outer pipe portions 33a, as shown by the black arrows in Fig. 5. Accordingly, in the first, third and fourth horizontal frame members 32, 34 and 35, the inner pipe portions 32b, 34b and 35b fit inside the left and right outer pipe portions 32a, 34a and 35a, and the distance in the width direction of the machine body between the pair of left and right vertical frame members 31 becomes relatively narrow as shown in Fig. 4. Thus, since the pair of left and right vertical frame members 31 are connected to the frames 25 of the pair of left and right running portions 20, respectively, when the width-varying screw portion 33c is rotated in one direction, the crawler pitch becomes relatively narrow.

[0027] On the other hand, when the width-varying screw portion 33c is manually rotated in the other direction, the left and right outer pipe portions 33a are separated in the second horizontal frame member 33, and the inner pipe portion 33b is exposed from the left and right outer pipe portions 33a, as shown by the outline arrows in Fig. 5. Accordingly, in the first, third and fourth horizontal frame members 32, 34 and 35, the inner pipe portions 32b, 34b, 35b are exposed from the left and right outer pipe portions 32a, 34a, 35a, and the distance in the machine body width direction between the pair of left and right vertical frame members 31 relatively increases as shown in Fig. 3, and thus the crawler pitch relatively increases.

[0028] As shown in FIG. 2, the platform lifting mechanism 36 has a first link arm 37, a second link arm 38, and an electric hydraulic cylinder 39. The first and second link arms 37, 38 are pivotally connected to each other at the middle portion of the first link arm 37 and the middle portion of the second link arm 38, so that they are substantially X-shaped as shown in FIG. 2. The first link arm 37 has an upper end portion connected to the vertical frame member 31 so as to be rotatable and slidable, and a lower end portion connected to the frame 25 of the traveling section 20 so as to be rotatable and non-slidable. On the other hand, the second link arm 38 has an upper end portion connected to the vertical frame member 31 so as to be rotatable and non-slidable, and a lower end portion connected to the frame 25 of the traveling section 20 so as to be rotatable and slidable. The electric hydraulic cylinder 39 is fixed to the vertical frame member 31, and a tip end of a rod 39a is connected to the upper end portion of the first link arm 37.

[0029] In the thus configured platform lifting mechanism 36, when the rod 39a of the electric hydraulic cylinder 39 advances, the upper end of the first link arm 37 and the upper end of the second link arm 38 move apart in the front-rear direction, and the upper end of the first link arm 37 and the lower end of the second link arm 38, and the lower end of the first link arm 37 and the upper end of the second link arm 38 move closer to each other in the vertical direction, so that the platform unit 30 moves down. On the other hand, when the rod 39a of the electric hydraulic cylinder 39 retracts, the upper end of the first link arm 37 and the upper end of the second link arm 38 move closer to each other in the front-rear direction, and the upper end of the first link arm 37 and the lower end of the second link arm 38, and the lower end of the first link arm 37 and the upper end of the second link arm 38 move apart to each other in the vertical direction, so that the platform unit 30 moves up.

[0030] The weeding attachment 40 is attached to the agricultural work vehicle 10 via an attachment lifting mechanism 45 so that it can be raised and lowered, and is configured to be towed by the agricultural work vehicle 10 to remove weeds that grow between the ridges 5. Specifically, the weeding attachment 40 has a cultivar 41 for removing weeds, a rake 43 for leveling clods of soil that come out of the cultivar 41, and the like.

[0031] <LiDAR for ridge tracing> Fig. 6 is a perspective view showing a schematic view of the installation positions of the LiDARs 50, 60 in the agricultural work vehicle 10. The ridge tracing LiDAR (distance sensor) 50 is a 2D-LiDAR with an aperture angle of 270° suitable for measurement and detection on a plane, and measures the distance to an object by measuring the time it takes for laser light to hit the object and bounce back. As shown in Fig. 6, the ridge tracing LiDAR 50 is attached to the center in the vehicle width direction at the front end of the frame unit 30 (first horizontal frame member 32) via first to third brackets 51, 55, 65.

[0032] More specifically, the first bracket 51 has a rectangular rear wall portion 52 that is bolted to the first horizontal frame member 32 of the mounting unit 30 and extends downward, a rectangular horizontal wall portion 53 that extends forward from the lower end of the rear wall portion 52, and a rectangular front wall portion 54 that extends downward from the front end of the horizontal wall portion 53. The second bracket 55 has a rectangular support wall portion 56 that is parallel to the front wall portion 54, and a mounting wall portion 57 that extends forward from the left end of the support wall portion 56. The third bracket 65 has a rectangular support wall portion 66 that is parallel to the mounting wall portion 57, and a rectangular mounting wall portion 67 that extends from the upper end of the support wall portion 66 to the right in the aircraft width direction. The support wall portion 56 has a circular arc-shaped long hole 56a that extends in the width direction of the body and curves downward, while the mounting wall portion 57 has a circular arc-shaped long hole 57a that extends in the front-rear direction and curves diagonally downward.

[0033] The second bracket 55 is attached to the first bracket 51 so that the left-right angle can be changed by fastening the support wall portion 56 to the front wall portion 54 with a bolt 58 slidably inserted into the long hole 56a. The third bracket 65 is attached to the second bracket 55 so that the front-rear angle can be changed by fastening the support wall portion 66 to the mounting wall portion 57 with a bolt 59 slidably inserted into the long hole 57a. The ridge tracing LiDAR 50 is attached to the upper surface of the mounting wall portion 67 of the third bracket 65. Therefore, the ridge tracing LiDAR 50 is attached to the frame portion 30 via the first to third brackets 51, 55, and 65 so that the left-right angle and the front-rear angle can be changed.

[0034] 7 is a diagram for explaining the detection direction of the ridge tracing LiDAR 50. As shown in FIG. 7, the ridge tracing LiDAR 50 is attached to the center of the vehicle width direction at the front end of the mounting unit 30 with the front-to-rear angle set so that the detection plane DP and the ground G form an angle of 30°. Therefore, the ridge tracing LiDAR 50 is capable of detecting the distance to the ridge 5 and the travel surface 7 in front of the agricultural work vehicle 10. Information about the ridge 5 and the travel surface 7 detected by the ridge tracing LiDAR 50 is input to the control device 70.

[0035] <LiDAR for marker detection> The marker detection LiDAR 60 is a 2D-LiDAR with an aperture angle of 270°, similar to the ridge tracing LiDAR 50. As shown in FIG. 6, the marker detection LiDAR 60 is attached to the front end of the right end of the frame unit 30 in the width direction of the machine body via a lifting bracket 61.

[0036] More specifically, the lifting bracket 61 has a rectangular tubular support part 62 attached by welding or the like to the front end of the right vertical frame member 31 of the mounting part 30, a rectangular tubular support pipe 63 extending vertically and fitted and inserted into the inside of the support part 62, and a mounting plate 64 provided at the lower end of the support pipe 63. The marker detection LiDAR 60 is attached to the upper surface of the mounting plate 64.

[0037] A plurality of recesses 63a are formed on the left side surface of the support pipe 63 in the width direction of the machine body, the recesses 63a being arranged at equal intervals in the vertical direction. Also, two recesses 62a are formed on the left side surface of the support part 62 in the width direction of the machine body, the recesses 62a being arranged at the same intervals in the vertical direction as the recesses 63a. Therefore, when the support pipe 63 is inserted into the support part 62, the two recesses 62a of the support part 62 fit into the recesses 63a of the support pipe 63 arranged in the vertical direction.

[0038] The recess 62a of the support part 62 is formed deep enough to support the weight of the support pipe 63, the mounting plate 64, and the marker detection LiDAR 60 when fitted into the recess 63a of the support pipe 63, while being able to slip out of the recess 63a of the support pipe 63 when an operator moves the support pipe 63 up and down. In other words, the height of the marker detection LiDAR 60 is variable relative to the mount part 30 by moving the support pipe 63 up and down relative to the support part 62. As a result, for example, even when the mount part 30 is raised by the mount lifting mechanism 36, the height of the marker detection LiDAR 60 can be kept constant by pulling down the support pipe 63.

[0039] FIG. 8 is a diagram for explaining the detection area DA of the marker detection LiDAR 60. In this embodiment, the marker detection LiDAR 60 is disposed at the right end in the vehicle width direction at the front end of the agricultural work vehicle 10, and the opening angle is 270°. Therefore, as shown in FIG. 8, the range behind the marker detection LiDAR 60 and to the left in the vehicle width direction is a blind spot, while the other range is the detection area DA. Note that the detection area DA in FIG. 8 is merely a schematic description, and in reality, the marker detection LiDAR 60 is configured to be able to detect objects in a farther range. Information about the object detected by the marker detection LiDAR 60 is input to the control device 70.

[0040] <Control device> 2, the control device 70 is installed, for example, on the stand unit 30. The control device 70 controls the electric hydraulic cylinder 39 of the stand lifting mechanism 36, the attachment lifting mechanism 45, etc., and is configured to control the traveling motor unit 27 of the traveling unit 20 based on the detection results of the ridge tracing LiDAR 50 and the marker detection LiDAR 60 to cause the agricultural work vehicle 10 to travel within the field 3 and to travel and turn on the headland 9.

[0041] For example, if the control device 70 rotates both the travel motor units 27 of the left and right traveling parts 20 in the forward direction, the agricultural work vehicle 10 moves forward, whereas if the control device 70 rotates both the travel motor units 27 of the left and right traveling parts 20 in the reverse direction, the agricultural work vehicle 10 moves backward. Also, if the control device 70 imparts a rotational speed difference to the travel motor units 27 of the left and right traveling parts 20, the agricultural work vehicle 10 turns in the direction in which the slower rotating travel motor unit 27 is provided. Furthermore, if the control device 70 rotates the travel motor unit 27 of one traveling part 20 in the forward direction and rotates the travel motor unit 27 of the other traveling part 20 in the reverse direction, the agricultural work vehicle 10 makes a pivot turn.

[0042] -Autonomous Driving- Next, the autonomous traveling of the agricultural work vehicle 10 in this embodiment will be described.

[0043] As described above, the autonomous driving system 1 of this embodiment (1) causes the agricultural work vehicle 10 to autonomously exit from the ridge 5 to the headland 9, drive on the headland 9, and enter from the headland 9 into the ridge 5, and (2) causes the agricultural work vehicle 10 to autonomously drive along the longitudinal direction of the ridge 5.

[0044] <(1) Autonomous driving and turning on headlands> In this embodiment, when the agricultural work vehicle 10 exits from the ridge 5 onto the headland 9, travels on the headland 9, and then enters the ridge 5 from the headland 9, the agricultural work vehicle 10 travels and turns while simultaneously detecting multiple markers 80 using the marker detection LiDAR 60.

[0045] Specifically, when the agricultural work vehicle 10 traveling across the ridges 5 approaches the headland 9, the marker detection LiDAR 60 detects multiple markers 80 erected at the longitudinal ends 5f of each ridge 5, and based on this detection result, the control device 70 obtains the end (end position) of the ridge 5 being worked on and causes the agricultural work vehicle 10 to exit the ridge 5 and move toward the headland 9.

[0046] Here, if the turning center of the agricultural work vehicle 10 is shifted to the left or right, the position of the center line VCL of the agricultural work vehicle 10 after turning with respect to the headland 9 will be different when turning right and when turning left. For this reason, in this embodiment, when turning the agricultural work vehicle 10 on the headland 9, the control device 70 is configured so that the agricultural work vehicle 10 turns only in the direction in which the marker detection LiDAR 60 is arranged, i.e., to the right.

[0047] Therefore, when exiting to the headland 9 is completed, the control device 70 turns the agricultural work vehicle 10 to the right (clockwise) while the marker detection LiDAR 60 detects the multiple markers 80, but because the turning direction is limited to the direction in which the marker detection LiDAR 60 is arranged, the multiple markers 80 do not enter the blind spot of the marker detection LiDAR 60, so it is possible to continue to detect the multiple markers 80 even during turning. In this way, because the marker detection LiDAR 60 continues to detect the multiple markers 80 from exiting to the headland 9 to the completion of turning, it is possible to continue to calculate the attitude of the agricultural work vehicle 10 with respect to the ridge 5, thereby making it possible to improve the turning angle accuracy.

[0048] Then, when the agricultural work vehicle 10 traveling forward or backward on the headland 9 after completing the turn approaches the next ridge 5, the marker detection LiDAR 60 detects multiple markers 80 erected near the next ridge 5, and the control device 70 makes the agricultural work vehicle 10 turn right again. In this case as well, the marker detection LiDAR 60 continues to detect the multiple markers 80, and after completing the turn, the control device 70 obtains the end (start position) of the next ridge 5 based on this detection result, and makes the agricultural work vehicle 10 enter the ridge 5 from the headland 9.

[0049] As described above, in the autonomous driving system 1 of this embodiment, it is possible to increase the turning angle accuracy using a single marker detection LiDAR 60 without using a directional sensor, etc., so that the agricultural vehicle 10 can autonomously exit or enter the ridges 5 with high accuracy while suppressing increases in costs.

[0050] <(2) About autonomous driving along ridges> In order to allow the agricultural work vehicle 10 to travel autonomously along the ridge 5, it is necessary to acquire a reference center line RCL of the ridge 5. Below, a method for estimating the center line RCL of the ridge 5 in the autonomous traveling system 1 of this embodiment will be described. However, in order to facilitate understanding of the present invention, a conventional method for estimating the center line RCL of the ridge 5 will first be described.

[0051] Fig. 18 is a diagram for explaining a conventional method for estimating the center line RCL of a ridge 5. In the conventional estimation method, as shown in Fig. 18, one of the left and right slopes 5b, 5c of the ridge 5 between the traveling surfaces 7 on which a pair of left and right traveling parts 120 are traveling, that is, the ridge 5 straddled by the agricultural work vehicle, is generally detected by a distance sensor 150 provided on the agricultural work vehicle, and the center line RCL of the ridge 5 is estimated based on the detection result.

[0052] However, such conventional estimation methods have the problem that they are not robust to changes in the shape of the ridge 5, since it becomes difficult to detect the distance from the slope 5b when the shape of the ridge 5 is distorted (for example, as shown in Figure 18, when the bottom 5d of the slope 5b of the ridge 5 is distorted).

[0053] Moreover, in the conventional estimation method, the center line RCL of the ridge 5 is estimated based on information on the slope 5b on one side, and therefore there is a risk of being affected by detection errors. Therefore, it is conceivable to increase the number of distance sensors 150 in order to obtain information on the slopes 5b, 5c on both sides of the ridge 5, but this would result in a problem of increased costs.

[0054] Therefore, in the autonomous driving system 1 according to this embodiment, the agricultural work vehicle 10 is controlled based on information about the traveling surfaces 7 on both sides of the ridge 5 straddled by the agricultural work vehicle 10. Specifically, the control device 70 is configured to estimate the center line RCL of the ridge 5 straddled by the agricultural work vehicle 10 based on information about the two traveling surfaces 7 detected by the ridge tracing LiDAR 50, and to cause the agricultural work vehicle 10 to travel along the ridge 5 based on the error between the estimated center line RCL of the ridge 5 and the center line VCL of the agricultural work vehicle 10.

[0055] FIG. 9 is a diagram for explaining the detection area DA of the ridge tracing LiDAR 50, in which FIG. 9(a) is a side view and FIG. 9(b) is a plan view. In FIG. 9, the arrow X indicates the front side in the longitudinal direction of the vehicle, the arrow Y indicates the left side in the width direction of the vehicle, and the arrow Z indicates the upper side in the vertical direction. As described above, the ridge tracing LiDAR 50 is attached to the center part in the width direction of the vehicle at the front end of the mounting part 30 so that the detection plane DP and the ground G form an angle of 30°, so that the distance to the ridge 5 and the travel surface 7 in front of the agricultural work vehicle 10 is detected as shown in FIG. 9(a). Therefore, as shown in FIG. 9(b), the ridge tracing LiDAR 50 includes the ridges 5I, 5II, and 5III in front of the agricultural work vehicle 10, and the travel surfaces 7I and 7II further forward than the ridges 5I, 5II, and 5III in the detection area DA.

[0056] 10 to 14 are diagrams for explaining a method for estimating the center line RCL of the ridge 5 in the autonomous driving system 1. When estimating the center line RCL of the ridge 5, the ridge tracing LiDAR 50 detects the ridge 5 straddled by the agricultural work vehicle 10 and the travel surfaces 7I and 7II on both sides of the ridge 5, as shown in FIG. 10. More specifically, of the multiple points P detected by the ridge tracing LiDAR 50, the control device 70 extracts a point cloud consisting of multiple points on the left travel surface 7I, a point cloud consisting of multiple points on the left slope 5b of the ridge 5, a point cloud consisting of multiple points on the top surface 5a of the ridge 5, a point cloud consisting of multiple points on the right slope 5c of the ridge 5, and a point cloud consisting of multiple points on the right travel surface 7II, as shown in the dashed frame in FIG. 10.

[0057] Next, the control device 70 extracts a flat point cloud from these point clouds. More specifically, the control device 70 obtains the normal of the surface on which each point P exists from the information around the point P, and extracts only the point cloud on the flat surface whose normal is approximately vertical. By such processing, as shown in FIG. 11, the point cloud on the left slope 5b of the ridge 5 and the point cloud on the right slope 5c of the ridge 5, which have inclined normals, are excluded. As a result, as shown by the dashed frame in FIG. 11, the point cloud on the left running surface 7I, the point cloud on the top surface 5a of the ridge 5, and the point cloud on the right running surface 7II are extracted by the control device 70.

[0058] Next, the control device 70 extracts the points included in a predetermined height range from the points on the left running surface 7I, the points on the top surface 5a of the ridge 5, and the points on the right running surface 7II. Here, the "predetermined height range" may be, for example, the height near the bottom end of the ridge 5. By this processing, the points on the top surface 5a of the ridge 5 are excluded, and only the points on the left running surface 7I and the points on the right running surface 7II are clustered by the control device 70, as shown by the dashed frame in FIG. 12.

[0059] Next, the control device 70 estimates the midpoint between point PI that is closest to the ridge 5 among the group of points on the left-hand traveling surface 7I and point PII that is closest to the ridge 5 among the group of points on the right-hand traveling surface 7II as the center line RCL of the ridge 5 (more accurately, a point on the center line RCL), as shown by the dashed frame in Figure 13. Then, the control device 70 causes the agricultural work vehicle 10 to travel along the ridge 5 based on the error between the estimated center line RCL of the ridge 5 and the center line VCL of the agricultural work vehicle 10.

[0060] For example, if the center line VCL of the agricultural work vehicle 10 is to the left of the estimated center line RCL of the ridge 5, the control device 70 drives the travel motor unit 27 of the left traveling section 20 at a faster rotation speed than the travel motor unit 27 of the right traveling section 20 in order to steer the agricultural work vehicle 10 to the right. Furthermore, if the deviation between the center line VCL of the agricultural work vehicle 10 and the estimated center line RCL of the ridge 5 is greater than a first predetermined amount, the control device 70 corrects the traveling direction of the agricultural work vehicle 10 while slowing down the speed of the agricultural work vehicle 10. Furthermore, if the deviation between the center line VCL of the agricultural work vehicle 10 and the estimated center line RCL of the ridge 5 is greater than a second predetermined amount (>first predetermined amount), the control device 70 corrects the traveling direction of the agricultural work vehicle 10 by performing a pivot turn.

[0061] As described above, according to this embodiment, the control device 70 estimates the center line RCL of the ridge 5 based on information about the running surfaces 7I, 7II on either side of the ridge 5, which are subject to less change in environment, shape, etc. than the ridge 5 itself, which is prone to significant changes in environment, shape, etc. due to the growth of planted crops, etc. Therefore, as shown in Figure 18 above, even if the bottom 5d of the ridge 5, etc., is locally collapsed, the center line RCL of the ridge 5 can be accurately estimated.

[0062] Furthermore, since the center line RCL of the ridge 5 is estimated from a point cloud that meets the condition of being within a specified flat height range on the running surfaces 7I, 7II on either side of the ridge 5, even if there are some unevenness in the running surfaces 7I, 7II, the running surfaces 7I, 7II can be determined under the same conditions on both sides of the ridge 5, so that the center line RCL of the ridge 5 can be estimated with even greater accuracy while reflecting the shape of the ridge 5.

[0063] Here, we will explain the case where the bottom of the ridge 5 is locally collapsed. In the example shown in Figure 14, the left bottom of the original ridge 5 is point 5d1, so the midpoint between this point 5d1 and the right bottom of the ridge 5 (point 5d2) is the center line RCL of the ridge 5. If the center line RCL of the ridge 5 is estimated based on point 5d1', which is closest to the ridge 5 among the points on the left running surface 7I, the distance ΔE between points 5d1 and 5d1' will directly become an error, resulting in a relatively large error ΔE.

[0064] In this regard, in the present embodiment, the control device 70 estimates the midpoint between point 5d1' which is closest to the ridge 5 in the group of points on the left running surface 7I and point 5d2 which is closest to the ridge 5 in the group of points on the right running surface 7II to be a point on the center line RCL' of the ridge 5.Therefore, even if the bottom of the ridge 5 is locally collapsed, the error (ΔE / 2) is halved, thereby further improving the estimation accuracy of the center line RCL' of the ridge 5.

[0065] Fig. 15 is a diagram for explaining the relationship between the arrangement position of the LiDAR 50 for ridge tracing and the detectable point cloud. In Fig. 15, the area that is equal to or greater than the height VH at which two virtual extension planes VP1 and VP2 extending upward from the slopes 5b and 5c on both sides of the ridge 5 intersect and is sandwiched between the two virtual extension planes VP1 and VP2 is set as an area RI, the area that is less than the height VH at which the virtual extension planes VP1 and VP2 intersect and is sandwiched between the virtual extension planes VP1 and VP2 is set as an area RIII, the area to the left of the areas RI and RIII is set as an area RII, and the area to the right of the areas RI and RIII is set as an area RIV.

[0066] As shown in region RIII of FIG. 15, if the installation height of the ridge tracing LiDAR 50C in the agricultural work vehicle 10 is too low, the ridge tracing LiDAR 50C can only detect the top surface 5a of the ridge 5, and it becomes difficult to detect the running surfaces 7I and 7II on both sides of the ridge 5. Also, as shown in region RII of FIG. 15, if the installation position of the ridge tracing LiDAR 50B in the vehicle width direction of the agricultural work vehicle 10 is too far to the left, the toe 5e of the ridge 5 blocks the field of view of the ridge tracing LiDAR 50B, and the point 5d2' on the right running surface 7II is detected as the inner end point 5d2, resulting in an error ΔE as shown in FIG. 15 between the actual center line RCL of the ridge 5. The same applies to region RIV of FIG. 15. In these cases, it becomes difficult to accurately estimate the center line RCL of the ridge 5, making it difficult to autonomously drive the agricultural work vehicle 10 along the ridge 5 with high precision, and in the worst case scenario, it is possible that the agricultural work vehicle 10 may come into contact with the ridge 5.

[0067] Therefore, as shown in Fig. 12, the LiDAR 50 for ridge tracing is preferably installed on the agricultural work vehicle 10 at a height equal to or higher than the height VH at which two imaginary extension planes VP1, VP2, which are obtained by extending upward the slopes 5b, 5c on both sides of the ridge 5 straddled by the agricultural work vehicle 10, intersect, and at a vehicle width direction position that fits within the area sandwiched between the two imaginary extension planes VP1, VP2, that is, within the area RI. In this way, if the LiDAR 50 for ridge tracing is arranged so as to fit within the area RI, it becomes possible for the LiDAR 50 for ridge tracing to always detect the traveling surfaces 7I, 7II on both sides of the ridge 5. In other words, if it fits within the area RI, even if the position of the LiDAR 50 for ridge tracing is shifted slightly to the left and right like the LiDAR 50A for ridge tracing due to the inclination of the agricultural work vehicle 10, it becomes possible to always detect the traveling surfaces 7I, 7II on both sides of the ridge 5.

[0068] In this regard, in this embodiment, as described above, the ridge tracing LiDAR 50 is attached to the center of the width of the body at the front end of the mounting unit 30, and the installation height is variable by changing the height of the mounting unit 30 using the mounting lifting mechanism 36, so that the ridge tracing LiDAR 50 can always be contained within the area RI, and this makes it possible to always detect the running surfaces 7I, 7II on either side of the ridge 5.

[0069] Fig. 16 is a diagram for explaining the relationship between the detection direction of the ridge tracing LiDAR 50 facing directly downward and the detection direction facing diagonally downward forward. Note that the error ΔE0 in Fig. 16 is the deviation of the center line VCL of the agricultural work vehicle 10 from the center line RCL of the ridge 5 when the detection direction of the ridge tracing LiDAR 50 is facing directly downward, and the error ΔE1 in Fig. 16 is the deviation of the center line VCL of the agricultural work vehicle 10 from the center line RCL of the ridge 5 when the detection direction of the ridge tracing LiDAR 50 is facing diagonally downward forward.

[0070] Generally, when the center line VCL of the agricultural work vehicle 10 straddling the ridge 5II is inclined with respect to the center line RCL of the ridge 5II, the distance between the two center lines is small directly below the agricultural work vehicle 10 and increases the further away from the agricultural work vehicle 10. Therefore, when the ridge tracing LiDAR 50 is installed so that the running surfaces 7I, 7II directly below the agricultural work vehicle 10 are included in the detection area, it is difficult for the attitude (direction) of the agricultural work vehicle 10 to be reflected in the error ΔE0 between the center line RCL of the ridge 5 and the center of the agricultural work vehicle 10.

[0071] 16, when the posture of the agricultural work vehicle 10 is taken into consideration, even if the center line VCL of the agricultural work vehicle 10 is shifted to the left by an error ΔE1 from the center line RCL of the ridge 5, it may be detected that the center line VCL of the agricultural work vehicle 10 is shifted to the right by an error ΔE0 from the center line RCL of the ridge 5 directly below the agricultural work vehicle 10. In such a case, if the agricultural work vehicle is steered to the left so as to make the error ΔE0 zero, it is also conceivable that the right rear end of the agricultural work vehicle 10 may come into contact with the slope of the right ridge III.

[0072] In this regard, in this embodiment, as described above, the LiDAR 50 for ridge tracing is installed on the agricultural work vehicle 10 facing diagonally downward and forward so that the driving surface 7 in front of the agricultural work vehicle 10 becomes the detection area. Therefore, it is possible to reflect the posture of the agricultural work vehicle 10 in the error ΔE1 between the estimated center line RCL of the ridge 5 and the center line VCL of the agricultural work vehicle 10, and therefore it is possible to reliably cause the agricultural work vehicle 10 to autonomously travel along the ridge 5.

[0073] (Other embodiments) The present invention is not limited to the embodiments, and can be embodied in various other forms without departing from the spirit or main characteristics thereof.

[0074] In the above embodiment, the crawler pitch of the agricultural work vehicle 10 is set so as to perform weeding work while straddling one ridge 5, but this is not limited thereto, and the crawler pitch of the agricultural work vehicle 10 may also be set so as to perform weeding work while straddling two or more ridges 5 (groups of ridges).

[0075] In the above embodiment, normals are obtained from information around each point to extract a point group on a flat surface, but this is not limiting. For example, as shown in FIG. 17, a predetermined height range (H1, H2 in the example of FIG. 17) may be set, and the point group included in this predetermined height range may be estimated as the point group on a flat surface. In FIG. 17, PI is the point closest to the ridge 5 on the left running surface 7I included in the range of height H1, PII is the point PII closest to the ridge 5 on the right running surface 7II included in the range of height H1, and PI' is the point closest to the ridge 5 on the right running surface 7II included in the range of height H2 (

[0076] In this case too, an error ΔE may occur between A1 and A2 depending on how the height range is set, but the error is halved by estimating the center line RCL of the ridge 5 based on information about the two running surfaces 7I, 7II, thereby improving the estimation accuracy of the center line RCL of the ridge 5.

[0077] As such, the above-described embodiment is merely illustrative in all respects and should not be construed as limiting. Furthermore, all modifications and variations within the scope of the claims are within the scope of the present invention. [Industrial Applicability]

[0078] According to the present invention, an agricultural vehicle can be made to autonomously drive along ridges with high precision without being affected by changes in the shape of the ridges, making it extremely useful when applied to an autonomous driving system for agricultural vehicles. [Explanation of symbols]

[0079] 1. Autonomous Driving System 3. Field 5 ridges 5b Slope 5c Slope​ 7 Running surface 10. Agricultural vehicles 20 Running part 50 LiDAR (distance sensor) for ridge tracking 70 Control device (estimation means) (control means) E error RCL Centerline RI area VCL-centric VP1 Virtual extension surface VP2 Virtual Extension Plane P point

Claims

1. An autonomous driving system for a work vehicle, which autonomously drives a work vehicle along a group of ridges that is made up of a plurality of ridges and has a driving surface formed between adjacent ridges in a farm field, the work vehicle performing work while straddling the group of ridges, A distance sensor installed in the work vehicle; an estimation means for estimating a center line of the group of ridges being spanned by the work vehicle based on information detected by the distance sensor; and a control means for causing the work vehicle to travel along the group of ridges based on an error between the center line of the group of ridges estimated by the estimation means and a center line of the work vehicle, An autonomous driving system for a work vehicle, characterized in that it is configured to estimate a point on the center line of the group of ridges by extracting only the point cloud on flat surfaces extending horizontally from a point cloud consisting of a plurality of points on the driving surface located on both sides of the ridge, a point cloud consisting of a plurality of points on each slope of the inclined portion of the ridge, and a point cloud consisting of a plurality of points on the top surface of the ridge.

2. The autonomous driving system for a work vehicle as described in claim 1, characterized in that the estimation means is configured to exclude the point clouds on the inclined slopes on both sides of the ridge group from the point clouds on the ridge group and on each of the two running surfaces on either side of the ridge group, and to exclude the point cloud on the top surface of the ridge group by extracting the point cloud that falls within a predetermined height range that is the height near the lower end of the ridge group, and to estimate that the midpoint between the point closest to the ridge group in the point cloud on one of the running surfaces and the point closest to the ridge group in the point cloud on the other running surface is the point on the center line of the ridge group.

Images