Autonomous lawnmower

DE112022006889T5Pending Publication Date: 2025-01-02HONDA MOTOR CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
DE112022006889
Authority / Receiving Office
DE · DE
Patent Type
Applications
Current Assignee / Owner
Filing Date
2022-03-23
Publication Date
2025-01-02

AI Technical Summary

Technical Problem

Conventional self-propelled lawn mowers often produce uneven lawn conditions due to unsatisfactory mowing finishes, making it difficult to identify and rectify unsatisfactory areas, resulting in suboptimal mowing quality compared to professional work.

Method used

An autonomous lawn mower equipped with front and rear imaging devices that assess grass conditions before and after mowing, adjusting its operational parameters such as speed and blade rotation to ensure a high-quality finish, and instructing rework at identified areas, including varying movement direction and speed to prevent uncut regions.

Benefits of technology

The autonomous lawn mower improves mowing efficiency and finish quality by dynamically adjusting its operations based on real-time imaging feedback, ensuring thorough cutting and reducing the risk of uncut areas, while also notifying external terminals of abnormalities that require manual attention.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

Provided is an autonomous lawnmower capable of performing lawn mowing tasks to a high standard and with reliable results. This autonomous lawnmower comprises: a lawnmower body that performs predetermined mowing tasks while autonomously driving over vegetation; a rear-facing imaging device positioned in the direction of travel of the lawnmower body; a memory unit that stores the content of the predetermined mowing tasks performed by the lawnmower body; and a control unit that controls the drive of the lawnmower body and, based on an image from the rear-facing device, instructs post-processing of vegetation at a location specified by the image.
Need to check novelty before this filing date? Find Prior Art

Description

Autonomous lawnmower

[0001] The present invention relates to an autonomous lawnmower.

[0002] In recent years, there has been a demand for self-propelled lawnmowers to improve work efficiency. For example, Patent Document 1 describes a lawnmower that is a self-propelled work robot. This lawnmower is equipped with a sensor that detects obstacles ahead in the direction of travel. When the sensor detects an obstacle while the lawnmower is mowing, it stores the location of the obstacle and avoids the obstacle while completing the planned mowing operation. After completing the mowing operation, the lawnmower moves to the stored location and checks whether the obstacle has been removed. If the obstacle has been removed, the lawnmower continues mowing the remaining part of the lawn at that location.

[0003] Japanese Patent Application Publication No. 3-73004

[0004] The results of mowing work done by self-propelled lawn mowers are not necessarily as consistent as those done by professional lawn mowers due to various factors. With conventional self-propelled lawn mowers, even if there are areas that are not finished properly after mowing, it is not possible to detect these, which can result in uneven grass condition after mowing.

[0005] Therefore, an object of the present invention is to provide an autonomous lawnmower that can perform lawn mowing work with a high level of stable finish.

[0006] (1) The autonomous lawnmower of the present invention is an autonomous lawnmower (for example, the autonomous lawnmower 1 described below) that includes a lawnmower main body (for example, the lawnmower main body 2 described below) that performs predetermined lawnmowing work while autonomously traveling on vegetation, rear imaging means (for example, the rear imaging device 4) that is arranged to point rearward in the direction of travel of the lawnmower main body, a memory unit (for example, the memory unit 6 described below) that stores the details of the predetermined lawnmowing work performed by the lawnmower main body, and a control unit (for example, the control unit 5 described below) that controls the operation of the lawnmower main body and, depending on the imaging results of the rear imaging means, instructs the lawnmower main body to re-work the vegetation at the imaging location indicated by the imaging results.

[0007] (2) In the autonomous lawnmower of (1) above, the memory unit may store the image capture location behind the direction of travel during the predetermined lawnmowing operation, and the control unit may change the work operation content during the lawnmowing operation performed on the vegetation at the image capture location from the work operation content during the predetermined lawnmowing operation when the operation is restarted.

[0008] (3) In the autonomous lawnmower described in (2) above, the control unit may increase the travel speed of the lawnmower body during the re-operation compared to the travel speed of the lawnmower body during the predetermined lawnmowing operation.

[0009] (4) In the autonomous lawnmower described in (3) above, the control unit may increase the rotational speed of the blade portion of the lawnmower body (e.g., the cutter blade 241 described below) during the re-operation compared to the rotational speed of the blade portion of the lawnmower body during the predetermined lawnmowing operation.

[0010] (5) In the autonomous lawnmower described in any one of (2) to (4) above, the control unit may change the direction of movement of the lawnmower body during the re-operation from the direction of movement of the lawnmower body during the predetermined lawnmowing operation.

[0011] (6) In the autonomous lawnmower described in any of (2) to (4) above, the control unit may issue an abnormality alert in response to the imaging results, and if the abnormality alert continues for a predetermined interval or more in the direction of travel of the lawnmower body during the predetermined lawnmowing operation, when the operation is resumed, may move the lawnmower body in the same direction as the direction of travel of the lawnmower body during the predetermined lawnmowing operation.

[0012] (7) In the autonomous lawnmower described in any one of (2) to (6) above, when there are multiple re-work locations for which the control unit is instructed to re-work, the control unit may determine the order in which the multiple re-work locations are to be re-worked, depending on the image capture locations or the image capture results.

[0013] (8) In the autonomous lawnmower described in (7) above, the control unit may determine the order in which to work on the plurality of re-work areas based on a result of comparing the image capture results for each of the plurality of re-work areas.

[0014] (9) In the autonomous lawnmower described in any of (1) to (8) above, the control unit may be configured to notify a specified external terminal when it detects from the imaging results that the amount of grass discharged from the lawnmower body (for example, the amount of swarf clumps Gc1 described below) is equal to or greater than a threshold value.

[0015] According to (1) above, the autonomous lawnmower itself instructs the vegetation at the imaged location to be reworked based on the image capture results of the rear imaging means, thereby enabling the lawnmower to perform lawnmowing work with a high level of stable results.

[0016] According to (2) above, the autonomous lawnmower performs mowing operations under different operating conditions for areas that require reworking where the work results during normal operation were insufficient, thereby further improving the quality of the rework.

[0017] According to (3) above, the grass density in the area to be re-worked is low due to mowing work during normal work, so by increasing the running speed of the autonomous lawnmower during re-work, work efficiency can be improved without compromising the finish quality.

[0018] According to (4) above, as the rotation speed of the blade increases, the number of times the blade comes into contact with the grass per given area increases, so even if the autonomous lawnmower travels fast, there is no risk of leaving grass uncut, thereby further improving work efficiency.

[0019] According to (5) above, by changing the direction of travel of the autonomous lawnmower during rework from the direction of travel during normal work, the risk of leaving any grass uncut is reduced, and the quality of the finish during rework can be further improved.

[0020] According to (6) above, even if a long uncut area occurs due to an error in the wrap width, for example, the autonomous lawnmower can efficiently process the uncut area by re-working it.

[0021] According to (7) above, the autonomous lawnmower can efficiently perform reworking according to the work order, thereby reducing unnecessary power consumption.

[0022] According to (8) above, the autonomous lawnmower can efficiently perform rework in accordance with the condition of the grass at the rework location.

[0023] According to (9) above, it is possible to notify the work manager, etc., of the occurrence of an abnormality that the autonomous lawnmower itself cannot handle, thereby improving the convenience of the autonomous lawnmower and further improving the quality of the lawnmower's finished results.

[0024] 1 is a schematic side view showing an autonomous lawnmower.

[0023] FIG. 1 is a diagram showing grass being properly cut.

[0024] FIG. 2 is a diagram showing the state of grass that has been properly cut.

[0025] FIG. 3 is a diagram showing grass that has been improperly cut.

[0026] FIG. 4 is a diagram showing the state of grass where white beards have appeared due to improper cutting.

[0027] FIG. 5 is a plan view showing overflows during lawn mowing work.

[0028] FIG. 6 is a functional block diagram of a control unit.

[0029] FIG. 7 is a functional block diagram of a work operation content determination unit.

[0030] FIG. 8 is a diagram showing an example of a reward table.

[0031] FIG. 9 is a diagram showing an example of a setting value correction table.

[0032] FIG. 10 is a diagram showing an example of a work path from the start of work by an autonomous lawnmower to the end of work.

[0033] FIG. 11 is a flowchart showing an example of a control operation for determining work operation content in an autonomous lawnmower.

[0034] FIG. 12 is a flowchart showing an example of a control operation for determining re-work in an autonomous lawnmower.

[0035] FIG. 13 is a flowchart showing an example of a control operation when re-work is performed by an autonomous lawnmower.

[0036] FIG. 14 is a flowchart showing an example of a control operation when re-work is performed by an autonomous lawnmower.

[0037] FIG. 15 is a diagram showing how the autonomous lawnmower re-works in relation to the work area.

[0038] FIG. 16 is a diagram explaining overlap.

[0039] FIG. 17 is a diagram showing the travel direction of the autonomous lawnmower when re-working due to a mistake in the lap width.

[0025] (General Configuration of Autonomous Lawnmower) As shown in Figure 1, the autonomous lawnmower 1 comprises a lawnmower main body 2, a front image capture device 3, a rear image capture device 4, a control unit 5, and a memory unit 6. The autonomous lawnmower 1 comprises a pair of front wheels 11 and a pair of rear wheels 12, one on each side, and is driven by a traction motor 21 built into the lawnmower main body 2 to travel on its own initiative and mow vegetation within the work area. In addition to the above, the autonomous lawnmower 1 also comprises a steering mechanism for changing the direction of travel, a rechargeable drive battery, etc., but these are not shown in Figure 1.

[0026] The autonomous lawnmower 1 of this embodiment travels in the left direction indicated by the arrow in Figure 1 when mowing. Therefore, in the autonomous lawnmower 1 of this embodiment shown in Figure 1, the forward direction of travel is the left direction, and the rear direction of travel is the right direction. The width direction of the autonomous lawnmower 1 of this embodiment is the direction perpendicular to the plane of the paper in Figure 1.

[0027] The lawnmower main body 2 has a lawn mowing blade inside the cutter deck 24 located at the bottom. The specific structure of the blade is not particularly limited. In this embodiment, the blade consists of a pair of cutter blades 241 arranged side by side in the width direction of the lawnmower main body 2. The cutter blade 241 rotates at a predetermined rotational speed around a rotation axis that is positioned approximately perpendicular to the ground by the drive of a blade motor 22 built into the lawnmower main body 2, and cuts the grass Ga ahead in the direction of travel. When the autonomous lawnmower 1 properly performs lawn mowing work, the grass Gb after the autonomous lawnmower 1 passes is cut to a predetermined length.

[0028] The cutter deck 24 can be moved up and down relative to the lawnmower body 2 by driving a deck lifting motor 23 built into the lawnmower body 2. By moving the cutter deck 24 up and down, the height of the cutter blade 241 relative to the ground is adjusted, and the length of the grass Gb after mowing is appropriately set.

[0029] The cutter deck 24 has a discharge port 242 that opens to the rear of the lawnmower body 2. The discharge port 242 discharges grass chips Gc cut by the cutter blade 241, dispersing them rearward in the direction of travel using wind pressure generated by the rotation of the cutter blade 241. The appropriately dispersed chips Gc are reused as soil fertilizer.

[0030] The front imaging device 3 is mounted on the lawnmower main body 2 and faces forward in the direction of travel. More specifically, the front imaging device 3 is attached to the front upper part of the lawnmower main body 2 facing slightly downward so that it can capture images of the uncut grass Ga ahead in the direction of travel during lawnmowing work.

[0031] A specific example of the front imaging device 3 is a camera having an image sensor that recognizes color images of the turf Ga. The front imaging device 3 may also be a moisture camera that can measure the moisture content of the turf Ga. For example, a near-infrared spectroscopic moisture camera may be used as the moisture camera.

[0032] The forward imaging device 3 may include two or more types of cameras, such as a camera with an image sensor and a near-infrared spectroscopic moisture camera. The forward imaging device 3 captures an image of the grass Ga ahead in the direction of travel before being mowed, and sends the forward imaging results, including the condition of the grass Ga, to the control unit 5.

[0033] The rear imaging device 4 is provided on the lawnmower main body 2, facing rearward in the direction of travel. More specifically, the rear imaging device 4 is attached to the upper rear of the lawnmower main body 2 facing slightly downward so that it can capture images of the grass Gb behind the lawnmower main body 2 in the direction of travel during lawnmowing work.

[0034] A specific example of the rear image capturing device 4 is a camera having an image sensor that recognizes color images of the grass Gb. The rear image capturing device 4 captures an image of the mowed grass Gb behind the vehicle in the direction of travel, and sends the rear image capturing results including the condition of the grass Gb to the control unit 5.

[0035] Here, the state of the grass Ga included in the forward imaging result and the state of the grass Gb included in the rearward imaging result will be described.

[0036] First, the condition of the grass Ga included in the forward imaging results can include the number (density), thickness, color, moisture content, etc. of the grass Ga. These conditions of the grass Ga can be recognized by image processing the forward imaging results.

[0037] The number and thickness of the grass Ga blades relate to how easily the grass Ga can be cut. For example, if there are a large number of grass Ga blades or if the grass Ga blades are thick, the grass Ga may be difficult to cut. In this case, it may be necessary to make adjustments such as slowing down the traveling speed of the autonomous lawnmower 1 or increasing the rotation speed of the cutter blade 241. Therefore, information about the number and thickness of the grass Ga blades as information about the condition of the grass Ga can be used as information for determining the traveling speed of the autonomous lawnmower 1 and the rotation speed of the cutter blade 241 for more appropriate lawn mowing work.

[0038] The color of the grass Ga is relevant to understanding the finished state of the grass Gb after mowing. For example, when the rotation speed of the cutter blade 241 is appropriate, the grass Ga is neatly cut by the cutter blade 241, as shown in FIG. 2A. As a result, the grass Gb after mowing is neatly cut to a predetermined length, as shown in FIG. 2B, so there is no significant difference in color between the grass Ga before mowing and the grass Gb after mowing. However, when the rotation speed of the cutter blade 241 is slow, or when the cutting ability of the cutter blade 241 decreases due to deterioration, as shown in FIG. 3A, the cutter blade 241 will push down the grass Ga, making it difficult to cut the grass Ga cleanly. As a result, white beards Gb1 formed by exposed fibers remain at the cut edge of the grass Gb after mowing, as shown in FIG. 3B. This causes the color of the grass Gb to appear whitish overall, resulting in a significant difference between the color of the grass Ga and the color of the grass Gb. Therefore, the color of the grass Ga can be used, for example, as information for determining the appropriate rotation speed of the cutter blade 241.

[0039] The moisture content of grass Ga is the amount of water adhering to the grass Ga. The moisture content of grass Ga can be measured using a moisture camera. Alternatively, the moisture content of grass Ga can be measured by detecting light reflection from the grass Ga using a camera with an image sensor. Furthermore, the moisture content of grass Ga can also include rainfall information or watering information input from an external source. The adhesion of moisture to grass Ga causes an increase in the mass of the grass Ga. For example, if the moisture content is high, the grass Ga is likely to fall over. This makes it difficult to stand the grass Ga, and some grass is likely to remain uncut. Therefore, the moisture content of grass Ga can be used, for example, to determine the appropriate rotation speed of the cutter blade 241 to generate an updraft that will stand up grass Ga that has fallen over due to moisture.

[0040] The condition of the grass Gb included in the rear image capture result is the finished state of the grass Gb after mowing. The finished state of the grass Gb can include the occurrence of uncut grass, the color of the grass Gb, the amount of grass that has fallen, the length of the grass Gb (cut height, unevenness of length), etc. These finished states of the grass Gb can be recognized by image processing the rear image capture result.

[0041] The term "dropped debris" refers to the accumulation of chips Gc on the grass Gb that occurs when chips Gc are released behind the autonomous lawnmower 1 in the direction of travel, without being dispersed. As a result, as shown in Figure 4, chip lumps Gc1, where chips Gc have accumulated, are scattered on the grass Gb behind the autonomous lawnmower 1 in the direction of travel. The grass Gb below the chip lumps Gc1 is less exposed to sunlight and less moisture is distributed, which can cause the grass to wither. Therefore, if there is a large amount of grass covered with chip lumps Gc1, they need to be removed immediately.

[0042] The control unit 5 comprehensively controls the operation of the autonomous lawnmower 1. As shown in Fig. 5 , the control unit 5 includes a motor control unit 51, a work operation content setting unit 52, a work operation content determination unit 53, a re-work determination unit 54, a positioning unit 55, and a communication unit 56.

[0043] The motor control unit 51 generates operation commands for each motor based on the setting values ​​of the work operation content set in the work operation content setting unit 52. The motor control unit 51 controls the drive of each motor, such as the travel motor 21, blade motor 22, and deck lifting motor 23, by sending the generated operation commands to each motor. This allows the autonomous lawnmower 1 to perform lawn mowing work.

[0044] The work operation content setting unit 52 sets the work operation content required for the lawn mowing work operation of the autonomous lawnmower 1. The work operation content is a parameter for the operation of each part of the autonomous lawnmower 1. The work operation content includes information related to the traveling direction and traveling speed of the autonomous lawnmower 1, the rotation speed of the cutter blade 241, and the height of the cutter deck 24. The work operation content set in the work operation content setting unit 52 is rewritable, and is corrected based on information sent from the work operation content determination unit 53.

[0045] 6 , the work operation content determination unit 53 determines and outputs new work operation content for obtaining better lawn mowing results based on both the forward imaging result detected by the forward imaging device 3 and the rearward imaging result detected by the rearward imaging device 4. The work operation content determination unit 53 determines the state of the grass Gb included in the rearward imaging result detected by the rearward imaging device 4, and if the state of the grass Gb is within a predetermined range, determines new work operation content according to the state of the grass Ga included in the forward imaging result detected by the forward imaging device 3.

[0046] The work operation content determination unit 53 can determine new work operation content by, for example, reinforcement learning. Reinforcement learning learns to maximize future rewards by learning the lawn mowing work (actions) performed by the autonomous lawnmower 1. In more detail, the work operation content determination unit 53 performs lawn mowing work (actions) based on certain work operation content for the state of the grass Ga before mowing contained in the forward imaging results, and selects an action to improve the state of the grass Gb after mowing detected by the rear imaging results at that time.

[0047] In reinforcement learning, the work movement content determination unit 53 sets a positive reward value if the condition of the grass Gb after mowing improves as a result of the lawn mowing work performed based on the work movement content set in the work movement content setting unit 52 relative to the condition of the grass Ga before mowing, and sets a negative reward value if the condition of the grass Gb worsens. Specifically, a reward table shown in FIG. 7 , for example, is pre-set in the work movement content determination unit 53. Based on the reward table, the work movement content determination unit 53 evaluates, for example, the condition of the grass Ga before mowing, which is included in the forward image capture result, on a three-level scale of "normal," "slightly bad," and "bad," and evaluates the condition of the grass Gb after mowing, which is included in the rearward image capture result, on a three-level scale of "good," "medium," and "bad." The work movement content determination unit 53 calculates either a positive or negative reward value from these combinations.

[0048] As shown in FIG. 7 , the reward value is weighted. For example, if the condition of the grass Ga included in the forward imaging result is "bad" and the condition of the grass Gb included in the rear imaging result is "good," this indicates that the work action content when the lawn mowing work is performed is extremely good. Therefore, for this combination, the reward value is weighted to be larger than for other combinations. On the other hand, if the condition of the grass Ga included in the forward imaging result is "normal" and the condition of the grass Gb included in the rear imaging result is "bad," this indicates that the work action content when the lawn mowing work is performed is poor. Therefore, for this combination, the reward value is smaller than for other combinations.

[0049] Note that this method of calculating the reward is merely an example and is not limiting. As another example of a method of calculating the reward, the reward value may be calculated by digitizing the front image and the rear image, respectively, and calculating (e.g., adding) the respective numerical values. For example, the condition of the grass Ga included in the front image capture result is digitized as "normal: 0 points," "slightly bad: 3 points," or "bad: 5 points," and the condition of the grass Gb after mowing included in the rear image capture result is digitized as "good: 5 points," "medium: 3 points," or "bad: -1 point." The reward is calculated by adding the numerical values ​​of the front image and the rear image. According to this, for example, even if the condition of the grass Ga included in the front image capture result is "bad: 5 points," if the condition of the grass Gb after mowing included in the rear image capture result is "good: 5 points," the total will be 10 points, and the highest reward can be obtained. On the other hand, if the condition of the grass Ga included in the forward imaging results is "Normal: 0 points," but the condition of the grass Gb after mowing included in the rear imaging results is "Bad: -1 points," the total will be -1 point, and the reward will be the lowest.

[0050] The work movement content determination unit 53 determines work movement content as a new action for the current state by, for example, performing deep learning based on the state of the grass Ga before mowing contained in the forward image capture result, the work movement content of the lawn mowing work performed on the grass Ga, the state of the grass Gb after mowing contained in the rearward image capture result, and the reward value calculated thereby. The work movement content determination unit 53 associates the state of the grass Ga, Gb, the work movement content during the lawn mowing work, and the reward, and accumulates this as experience for determining new work movement content.

[0051] The new work movement content determined as a result of reinforcement learning by the work movement content determination unit 53 includes content that will result in a greater positive reward than the current work movement content, such as increasing or decreasing the rotation speed of the cutter blade 241, raising or lowering the height of the cutter deck 24, or increasing or slowing the traveling speed of the autonomous lawnmower 1. Specifically, the work movement content determination unit 53 has a setting value correction table, such as that shown in FIG. 8 , pre-stored. For example, if the current work movement content of the blade motor 22 is a setting value (b3) of "3," the work movement content determination unit 53 corrects the new work movement content to, for example, a setting value (b2) of "2" or a setting value (b4), etc. The new work movement content is sent to the work movement content setting unit 52, and the work movement content for the lawnmowing work recorded in the work movement content setting unit 52 is updated to a work movement content that enables more appropriate lawnmowing work to be performed.

[0052] The work operation content determination unit 53 determines the work operation content and the work operation content setting unit 52 corrects the work operation content at a predetermined control cycle after the autonomous lawnmower 1 starts mowing. The control unit 5 updates the operation commands for each motor in the motor control unit 51 based on the work operation content for the lawnmower newly rewritten in the work operation content setting unit 52, and continues mowing the lawn with the autonomous lawnmower 1.

[0053] The re-work determination unit 54 determines whether the lawn mowing work needs to be re-done based on the finished state of the lawnmowing work included in the rear image capture result from the rear image capture device 4. Specifically, the re-work determination unit 54 determines whether the lawnmowing work needs to be re-done based on the state of the grass Gb after mowing, such as the occurrence of any grass that has not been cut or has fallen off, and the color of the grass Gb. If the re-work determination unit 54 determines that the lawnmowing work needs to be re-done, it issues an abnormality notification. The abnormality notification does not necessarily have to be an explicit notification sent to an external terminal or the like, but may be information indicating that it has been determined that the lawnmowing work needs to be re-done.

[0054] The control unit 5 issues an instruction to perform rework when the rework determination unit 54 determines that rework is necessary. This instruction to perform rework includes a notification to the work manager and an instruction to operate the autonomous lawnmower 1 for rework.

[0055] Note that rework is not limited to work performed by the autonomous lawnmower 1. Rework also includes notifying an external device, such as a mobile device or personal computer owned by the work manager, of the details of the rework via the communication unit 56. This makes it possible to prompt the work manager to perform work such as removing the lump of chips Gc1 that the autonomous lawnmower 1 cannot perform, for example.

[0056] The positioning unit 55 measures the current position and direction of travel of the autonomous lawnmower 1 while it is mowing. Any specific positioning unit 55 can be used without limitation, as long as it can measure the current position of the autonomous lawnmower 1 while it is mowing. For example, the positioning unit 55 can use a Global Positioning System (GPS) to measure the current position of the autonomous lawnmower 1. The positioning unit 55 may also measure the current position of the autonomous lawnmower 1 by using a laser or the like to measure the distance and angle relative to a specific location, such as a charging point for the autonomous lawnmower 1.

[0057] The communication unit 56 performs wireless communication between the autonomous lawnmower 1 and external devices such as external terminals.

[0058] 1 , the memory unit 6 stores various data related to the operation of the autonomous lawnmower 1 when performing lawn mowing work. Specifically, the memory unit 6 stores the work content when performing lawn mowing work with the autonomous lawnmower 1. This work content includes the front image capture results captured by the front image capture device 3, the rear image capture results captured by the rear image capture device 4, information about the position of the autonomous lawnmower 1 measured by the positioning unit 55 and the image capture location, and the work operation content of the autonomous lawnmower 1.

[0059] The control unit 5 including each of the functional blocks described above is configured with a processing unit such as a CPU (Central Processing Unit). To realize the functions of each functional block, the control unit 5 may also include an auxiliary storage device such as a HDD (Hard Disk Drive) that stores various control programs such as application software and an OS (Operating System), and a main storage device such as a RAM (Random Access Memory) that stores data temporarily required for the processing unit to execute programs. The storage unit 6 may also be configured with a storage device such as a RAM (Random Access Memory).

[0060] The control unit 5 does not have to be built into the lawnmower main body 2. All or some of the functions of the control unit 5 may be provided in an external terminal such as a personal computer and connected via a network line or the like.

[0061] (Determining the Work Operation Details of the Autonomous Lawnmower) Next, we will explain the control that is used when determining the specific work operation details of the autonomous lawnmower 1 of this embodiment. First, using Figure 9, we will explain the movement path of the autonomous lawnmower 1 of this embodiment when mowing. Shown here is a work area W that is rectangular in plan view. Before mowing, the autonomous lawnmower 1 waits at the work start position in the lower left corner of the work area W.

[0062] When mowing, the autonomous lawnmower 1 of this embodiment moves in a straight line from the lower side W1 to the upper side W2 of the work area W, then turns 180 degrees at the upper side W2 and continues moving in a straight line toward the lower side W1. Once it reaches the lower side W1, the autonomous lawnmower 1 turns 180 degrees again and continues moving in a straight line toward the upper side W2. The autonomous lawnmower 1 mows the entire work area W, repeatedly moving between sides W1 and W2 from the work start position to the work end position. Information about the movement path of the autonomous lawnmower 1 is pre-recorded, for example, in the memory unit 6.

[0063] 10 is a flowchart illustrating the control operation for determining the work operation details in the autonomous lawnmower 1. The control unit 5 drives the travel motor 21, blade motor 22, and deck lifting motor 23 based on the work operation details recorded in the work operation detail setting unit 52. This causes the autonomous lawnmower 1 to perform lawn mowing work while moving along the movement path shown in FIG.

[0064] When mowing work begins, the control unit 5 acquires the front image capture result from the front image capture device 3 and the rear image capture result from the rear image capture device 4 (step S11). This acquires the state of the grass Ga before mowing and the state of the grass Gb after mowing. The control unit 5 performs reinforcement learning based on the respective states of the grass Ga and the grass Gb (step S12).

[0065] The control unit 5 calculates a reward based on the state of the grass Ga detected from the forward imaging results and the state of the grass Gb detected from the rearward imaging results through reinforcement learning in the work operation content determination unit 53, and based on the reward, determines a new work operation content for better grass mowing work compared to the current work operation content (step S13).

[0066] Here, it is preferable that the information about the condition of the grass Ga before mowing contained in the forward imaging results includes at least information about the moisture content of the grass Ga. This allows the control unit 5 to determine work operation content that takes the moisture content of the grass Ga into consideration.

[0067] Furthermore, if the condition of the grass Ga included in the forward imaging results includes information on at least one of the number of grass Ga blades and the thickness of the grass Ga, and information on the water content of the grass Ga, the control unit 5 preferably prioritizes the water content of the grass Ga over the information on the number of grass Ga blades and the thickness of the grass Ga when determining new work operation content. For example, even if the condition of the grass Ga is determined to be "normal" as shown in FIG. 7 because the number of grass Ga blades is small or the grass Ga is thin, if the water content of the grass Ga is high, the grass Ga may fall over due to the weight of the water, making it more likely that some grass will be left uncut. In this case, the condition of the grass Ga is determined to be "poor," for example. This allows the control unit 5 to determine more appropriate work operation content according to the water content of the grass Ga before mowing, which is included in the forward imaging results.

[0068] When the condition of the grass Ga included in the forward imaging results includes information on both the number of grass Ga and the thickness of the grass Ga, the control unit 5 preferably prioritizes the information on the thickness of the grass Ga over the information on the number of grass Ga when determining new work operation content. When the thickness of each grass Ga is thick, cutting is more difficult and requires more torque than when there are multiple thin grasses. For example, even if the condition of the grass Ga is judged to be "normal" as shown in FIG. 7 due to the presence of multiple thin grasses Ga, if thick grasses are mixed in, the condition of the grass Ga is judged to be, for example, "slightly poor." This allows for the determination of more appropriate work operation content according to the thickness of the grass Ga.

[0069] If the condition of the grass Gb included in the rear image capture results includes information on the amount of chip lumps Gc1 and information on the condition of the grass Gb itself (uncut grass, color), the control unit 5 preferably prioritizes the information on the amount of chip lumps Gc1 over the information on the condition of the grass Gb itself when determining new work operation content. This is because chip lumps Gc1 can cause grass to wither. For example, even if the condition of the grass Gb itself is determined to be "Good" as shown in FIG. 7 because there are no uncut grass lumps or white hairs Gb1 on the grass Gb, if the amount of chip lumps Gc1 is large, the condition of the grass Gb can be determined to be, for example, "Bad." This makes it possible to determine more appropriate work operation content according to the amount of chip lumps Gc1.

[0070] If the condition of the grass Gb included in the rearward imaging results includes information on the color of the grass Gb and information on the unevenness of the grass Gb's length, the control unit 5 preferably prioritizes information on the change in the color of the grass Gb included in the frontward imaging results and the rearward imaging results over information on the unevenness of the grass Gb's length when determining new work operation content. For example, even if the grass Gb does not have uneven length and the condition of the grass Gb is determined to be "Good" as shown in FIG. 7, if white beards Gb1 have appeared on the grass Gb and the color has changed to white compared to the grass Ga, the condition of the grass Gb may be determined to be, for example, "Bad." This makes it possible to determine more appropriate work operation content according to the occurrence of white beards Gb1 on the grass Gb.

[0071] One possible cause of white beards Gb1 appearing on the grass Gb is a decrease in sharpness due to deterioration of the cutter blade 241. Therefore, if the control unit 5 determines that the grass Gb is in a "Bad" state because white beards Gb1 have appeared on the grass Gb after mowing, it may notify the external terminal of the work manager via the communication unit 56 to prompt the replacement of the cutter blade 241.

[0072] Once the control unit 5 has determined the new work operation content, it rewrites the work operation content recorded in the work operation content setting unit 52 with the new work operation content, operates the autonomous lawnmower 1 based on the new work operation content, and continues mowing the lawn until the autonomous lawnmower 1 reaches the work end position (step S14).

[0073] (Rework by Autonomous Lawnmower) Next, a specific operation for determining whether rework is necessary by the autonomous lawnmower 1 of this embodiment will be described using the flowchart in Figure 11. Here again, the autonomous lawnmower 1 will be described as performing mowing work along a movement path for the work area W similar to that shown in Figure 9.

[0074] The autonomous lawnmower 1 performs a predetermined mowing operation on vegetation within the work area W, based on the work operation content set in the work operation content setting unit 52. While the autonomous lawnmower 1 is performing the mowing operation, the control unit 5 determines whether the operation needs to be performed again based on the rear image capture results detected by the rear image capture device 4. That is, when the mowing operation begins, the control unit 5 acquires the rear image capture results detected by the rear image capture device 4 at a predetermined control cycle (step S21).

[0075] The control unit 5 determines whether or not rework is necessary based on the acquired rear image capture result (step S22). Specifically, the control unit 5 determines whether or not rework is necessary in the rework determination unit 54 based on the finished state of the grass Gb after mowing, which is included in the rear image capture result.

[0076] In step S22, if it is determined that rework is not necessary (step S22; NO), the process returns to step S21.

[0077] If it is determined in step S22 that rework is necessary (step S22; YES), the control unit 5 issues an abnormality alert indicating that rework is necessary (step S23), and using the abnormality alert as a trigger, stores information such as the imaging location of the rear imaging results corresponding to the area determined to require rework in the memory unit 6 (step S24).

[0078] Specifically, the memory unit 6 stores the work details of the lawn mowing work at the time it is determined that rework is necessary. The work details of the lawn mowing work include information about the image capture location (the position of the autonomous lawnmower 1 measured by the positioning unit 55) included in the rear image capture results, information about the work operations performed during the lawn mowing work, and information about the rework details (the occurrence of any grass remaining uncut, the occurrence of any grass that has fallen off, etc.). The information about the image capture location is obtained from the position information measured by the positioning unit 55. Information about the traveling speed of the autonomous lawnmower 1 is obtained from the operation command output to the traveling motor 21. Information about the rotational speed of the cutter blade 241 is obtained from the operation command output to the blade motor 22. Information about the height of the cutter deck 24 is obtained from the operation command output to the deck lifting motor 23. The memory unit 6 stores information about the traveling speed of the autonomous lawnmower 1, information about the rotational speed of the cutter blade 241, and information about the height of the cutter deck 24 in association with information about the image capture location.

[0079] The control unit 5 repeats the processes from step S21 to step S24 until the autonomous lawnmower 1 reaches the work end position. If there are multiple areas that need to be re-worked during the lawnmowing operation, information about the work content at each of the re-work areas is stored in the memory unit 6.

[0080] The re-work is carried out after the autonomous lawnmower 1 has finished mowing the vegetation in the work area W. The flow of this re-work operation will be explained using the flowcharts in Figures 12A and 12B. Note that when mowing work is finished, the autonomous lawnmower 1 stops driving the cutter blade 241.

[0081] When the autonomous lawnmower 1 reaches the work end position and the lawn mowing work is completed, the control unit 5 acquires information about the work content for all areas that require re-work that is stored in the memory unit 6 (step S31).

[0082] Next, the control unit 5 checks the work content for each rework location and instructs the rework of the vegetation at the location where the rear image capture result indicates the need for rework. Specifically, the control unit 5 first determines whether or not the work content includes spillage (lumps of cuttings Gc1) (step S32). If a large amount of spillage occurs, it can cause the grass to wither and must be removed immediately.

[0083] In step S32, as shown in Figure 13, if one or more spills (lumps of chips Gc1) have occurred in the work area W after the mowing work (step S32; YES), the control unit 5 obtains information on the amount of grass in each spill (amount of clumps of chips Gc1) from the work content (step S321).

[0084] Next, the control unit 5 compares the acquired amount of grass with a preset threshold (step S322). If the comparison results in the amount of grass exceeding the threshold (step S322; YES), the control unit 5 notifies the work manager of this information via the communication unit 56 to the external terminal of the work manager and instructs the work manager to retry (step S323). In other words, the control unit 5 notifies the work manager that a stubble with an amount of grass exceeding the threshold has occurred in a specific location in the work area W, and urges the work manager to remove the swarf lump Gc1.

[0085] After notifying the worker, the control unit 5 further determines whether there are any other areas that need to be reworked other than the area where the overgrowth occurred (step S324). Also, even if the amount of grass does not exceed the threshold in step S322 (step S322; NO), the control unit 5 determines whether there are any other areas that need to be reworked (step S324). This is because if the amount of grass is a small amount of overgrowth, it is determined that there is no major problem with the growth of the grass Gb and that the grass can be used as fertilizer for the soil.

[0086] If it is determined in step S324 that there are no other areas to be reworked (step S324; NO), the rework determination process ends.

[0087] If it is determined in step S324 that there are other rework locations (step S324; YES), the control unit 5 checks whether there are multiple other rework locations (step S33).

[0088] If it is determined in step S33 that there is only one other re-work location (step S33; NO), the control unit 5 obtains location information for that re-work location from the work content and instructs the autonomous lawnmower 1 to re-work (step S34). That is, the control unit 5 operates the autonomous lawnmower 1 to head directly to the re-work location and perform the work again. This causes the autonomous lawnmower 1 to move the shortest distance from the work end position to the re-work location.

[0089] On the other hand, if it is determined in step S33 that there are multiple other areas that need to be reworked (step S33; YES), the control unit 5 determines the order in which the multiple areas will be reworked, and instructs the autonomous lawnmower 1 to perform the rework in the determined order (step S35).

[0090] The order of re-work can be determined in order of closest distance to the re-work location. For example, as shown in Figure 13, if two re-work locations OB1 and OB2 have occurred in work area W, the control unit 5 controls the operation of the autonomous lawnmower 1 to move the shortest distance to re-work locations OB2 and OB1 in that order, in order of closestness to the autonomous lawnmower 1, and re-work each of the re-work locations. This allows re-work to be performed efficiently and reduces unnecessary power consumption.

[0091] The order of rework can also be determined in descending order of the need for rework, for example, in descending order of the area of ​​grass that remains uncut. This makes it possible to make defective areas of the grass Gb in the work area W less noticeable while taking into account the remaining battery power.

[0092] Furthermore, the order of rework can be determined based on the results of comparing the condition of the grass Gb included in the rear image capture of each rework area. For example, by performing rework in order of darker grass Gb, the number of uncut grass Gb, or the density of uncut grass Gb, it is possible to perform rework efficiently according to the condition of the grass Gb at the rework area.

[0093] Once the order of re-work has been determined, the control unit 5 controls the operation of the autonomous lawnmower 1 so that the re-work is carried out in accordance with the determined order of work. In this way, the autonomous lawnmower 1 carries out re-work on each re-work location (step S36).

[0094] Here, when the autonomous lawnmower 1 performs rework, it is preferable that the control unit 5 change the work operations performed during rework compared to the work operations performed by the autonomous lawnmower 1 during normal operation (first mowing operation). Specifically, at least one of the following is changed during rework: the direction of travel and traveling speed of the autonomous lawnmower 1, the rotational speed of the cutter blade 241, and the height of the cutter deck 24. Because it is possible to perform mowing under different operating conditions for areas where the work results during normal operation were insufficient, it is possible to improve work efficiency and further improve the quality of the rework finish.

[0095] More specifically, the control unit 5 increases the travel speed of the autonomous lawnmower 1 during re-work compared to the travel speed of the autonomous lawnmower 1 during normal work. Because the density of the grass Gb in the area to be re-worked has decreased due to mowing work during normal work, the control unit 5 can improve work efficiency without compromising the finish quality by increasing the travel speed of the autonomous lawnmower 1 during re-work.

[0096] In this way, when the travel speed of the autonomous lawnmower 1 is increased during rework, it is preferable for the control unit 5 to increase the rotation speed of the cutter blade 241 during rework compared to the rotation speed of the cutter blade 241 during normal work. When the rotation speed of the cutter blade 241 is increased, the number of times the cutter blade 241 hits the grass Gb per given area increases, so even if the travel speed of the autonomous lawnmower 1 is fast, there is no risk of leaving any grass Gb uncut. This can further improve work efficiency.

[0097] Furthermore, when the autonomous lawnmower 1 performs rework, the control unit 5 preferably changes the traveling direction of the autonomous lawnmower 1 for rework compared to the traveling direction of the autonomous lawnmower 1 during normal operation. For example, as shown in FIG. 14 , if the traveling direction of the autonomous lawnmower 1 during normal operation is bottom-up, the autonomous lawnmower 1 performs rework while traveling in a direction different from the bottom-up direction relative to the rework area OB. If the autonomous lawnmower 1 performs rework while traveling in the same direction as during normal operation, the lawnmower will perform the mowing under the same conditions as during normal operation, which could result in some grass being left uncut. By changing the traveling direction of the autonomous lawnmower 1 during rework from the traveling direction during normal operation, the quality of the finished product during rework can be further improved.

[0098] 14 shows a case where, during re-work, the autonomous lawnmower 1 performs re-work while traveling in a direction perpendicular to the direction of travel during normal work. However, the direction of travel of the autonomous lawnmower 1 during re-work is not limited to a direction perpendicular to the direction of travel during normal work, and may be any intersecting direction. The direction of travel of the autonomous lawnmower 1 during re-work may also be the opposite direction to the direction of travel during normal work.

[0099] In step S23 of the flowchart shown in FIG. 11 , an abnormality notification issued when it is determined that rework is necessary may continue for more than a predetermined time interval or a predetermined distance in the direction of travel of the autonomous lawnmower 1. Such an abnormality notification occurs due to an error in the overlap width during mowing. Specifically, as shown in FIG. 15A , mowing is typically performed so that an overlapping area 100 having a predetermined overlap width is formed between two adjacent rows of work areas (working ranges) whose traveling directions of the autonomous lawnmower 1 differ by 180 degrees. However, as shown in FIG. 15B , if an error in the overlap width prevents the two adjacent rows of work areas (working ranges) from overlapping, a long, non-overlapping, uncut area 200 will be generated along the direction of travel of the autonomous lawnmower 1. In this case, the control unit 5 detects an abnormality in the direction of travel of the autonomous lawnmower 1 for more than a predetermined time interval or a predetermined distance, and therefore continues to issue an abnormality notification.

[0100] If the control unit 5 detects in step S23 that the abnormality notification has continued for more than a predetermined time interval or a predetermined distance, then when restarting work, it is preferable to perform the work while driving the autonomous lawnmower 1 in the same direction as the autonomous lawnmower 1 would travel during normal work. This makes it possible to efficiently process long uncut areas 200 that occur due to an error in the wrap width. If the abnormality notification in step S23 occurs less than the predetermined time interval or the predetermined distance, then the control unit 5, as described above, changes the direction of travel of the autonomous lawnmower 1 when restarting work from the direction of travel during normal work.

[0101] Note that the cutter blade 241 remains stopped from rotating while the autonomous lawnmower 1 moves toward the re-work area. For example, if there is one re-work area, the cutter blade 241 begins to drive when the autonomous lawnmower 1 reaches the re-work area and begins re-working at that re-work area. If there are multiple re-work areas, the cutter blade 241 begins to drive when the autonomous lawnmower 1 reaches each of the re-work areas and begins re-working. In other words, the cutter blade 241 stops driving from the end of re-working at one re-work area until the start of re-working at the next re-work area. This makes it possible to reduce unnecessary power consumption.

[0102] When reworking of all rework locations is completed, the mowing work of the autonomous lawnmower 1 in the work area W is completed.

[0103] In the above embodiment, the control unit 5 issues a command to perform rework when a drop occurs, just as it does when issuing other rework commands, after the autonomous lawnmower 1 reaches the work end position and the mowing job is completed, but this is not limited to this. The control unit 5 may also issue a command to perform rework when a drop occurs when it detects the occurrence of a drop.

[0104] In the above embodiment, the autonomous lawnmower 1 is configured to discharge the cuttings Gc rearward in the direction of travel, but this is not limited to this. The autonomous lawnmower may also discharge the cuttings Gc to the side in the direction of travel. In this case, the autonomous lawnmower may be equipped with a separate side imaging device that is oriented to the side in the direction of travel in order to measure the amount of grass discharged to the side in the direction of travel (the occurrence of grass spillage).

[0105] REFERENCE SIGNS LIST 1 Autonomous lawnmower 2 Lawnmower body 241 Cutter blade (blade section) 4 Rear imaging device (rear imaging means) 5 Control section 6 Memory section Gc1 Clump of cuttings (grass released from the lawnmower body)

Claims

1. An autonomous lawnmower comprising: a lawnmower body that performs predetermined lawnmowing work on vegetation while autonomously traveling; a rear imaging means that is provided and points rearward in the direction of travel of the lawnmower body; a memory unit that stores the details of the predetermined lawnmowing work performed by the lawnmower body; and a control unit that controls the operation of the lawnmower body and, depending on the imaging results of the rear imaging means, instructs the lawnmower body to perform re-work on the vegetation at the imaging location indicated by the imaging results.

2. The autonomous lawnmower of claim 1, wherein the memory unit stores the image capture location behind the direction of travel during the predetermined lawnmowing operation, and the control unit changes the work operation content during the lawnmowing operation performed on the vegetation at the image capture location during the re-operation from the work operation content during the predetermined lawnmowing operation.

3. An autonomous lawnmower as described in claim 2, wherein the control unit increases the travel speed of the lawnmower body during the re-operation compared to the travel speed of the lawnmower body during the predetermined lawnmowing operation.

4. An autonomous lawnmower as described in claim 3, wherein the control unit increases the rotation speed of the blade portion of the lawnmower body during the re-operation compared to the rotation speed of the blade portion of the lawnmower body during the predetermined lawnmowing operation.

5. An autonomous lawnmower as described in any one of claims 2 to 4, wherein the control unit changes the direction of movement of the lawnmower body during the re-operation from the direction of movement of the lawnmower body during the predetermined lawnmowing operation.

6. An autonomous lawnmower as described in any one of claims 2 to 4, wherein the control unit issues an abnormality alert in accordance with the imaging results, and if the abnormality alert continues for more than a predetermined interval in the direction of travel of the lawnmower body during the predetermined lawnmowing operation, when the operation is resumed, moves the lawnmower body in the same direction as the direction of travel of the lawnmower body during the predetermined lawnmowing operation.

7. An autonomous lawnmower as described in any one of claims 2 to 6, wherein, when there are multiple areas to be reworked for which rework is instructed, the control unit determines the order in which work is to be done on the multiple areas to be reworked based on the image capture location or the image capture results.

8. An autonomous lawnmower as described in claim 7, wherein the control unit determines the order of work on the multiple rework areas based on the results of comparing the image capture results for each of the multiple rework areas.

9. An autonomous lawnmower as described in any one of claims 1 to 8, wherein the control unit notifies a specified external terminal when it detects from the imaging results that the amount of grass discharged from the lawnmower body is equal to or greater than a threshold value.

Citation Information

Cited By

  • Autonomous work machine, work management system, method for evaluating a vegetation group, program and computer-readable recording medium

    DE102024102452A1