Working System

Abstract

To easily acquire a relative position to a specific part of a mobile body in a region.SOLUTION: In a work system, a relative position between a mobile body and a specific part is acquired on the basis of at least one of a size of a specific image included in a captured image or a position of a reference line of the specific image. For example, in the captured image, a distance between the specific part and the mobile body when the specific image corresponding to the specific part is large is shorter than when the specific image is small. For example, in the captured image, a deviation in a width direction between the center of an imaging device provided in the mobile body and a line corresponding to the reference line of the specific part is larger when a position of the reference line of the specific image, for example, the length between the reference line of the specific image and the central line of the captured image is greater than a set length than when the length is less than or equal to the set length. As described above, the relative position between the mobile body and the specific part can be easily acquired on the basis of at least one of the size of the specific image in the captured image or the position of the reference line.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a work system in which a mobile object moves within a predetermined area to perform work. [Background technology]

[0002] Patent Document 1 describes a mobile object position acquisition device that acquires the position of a mobile object based on an image captured by a camera. In the mobile object position acquisition device described in Patent Document 1, an image similar to an actually acquired image is selected from a plurality of captured images stored in advance in a storage unit, and the position of the mobile object is acquired based on the image capture position of the selected captured image. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2016-162013 A Summary of the Invention [Problem to be solved by the invention]

[0004] An object of the present invention is to easily obtain the relative position between a moving object and a specific part within an area. [Means for solving the problem]

[0005] In the work system according to the present invention, the relative position between the moving body and the specific part is acquired based on at least one of the size of the specific image included in the captured image and the position of the reference line of the specific image. For example, in the captured image, when the specific image corresponding to the specific part is large, the distance between the specific part and the moving body is shorter than when it is small. Also, for example, in the captured image, when the position of the reference line of the specific image, for example, the length between the reference line of the specific image and the center line of the captured image, is greater than a set length, the deviation in the width direction between the center of the imaging device provided on the moving body and the line corresponding to the reference line of the specific part is greater than when it is equal to or less than the set length. As described above, the relative position between the moving body and the specific part can be easily obtained based on at least one of the size of the specific image in the captured image and the position of the reference line. [Brief description of the drawings]

[0006] [Figure 1] 1 is a plan view showing a schematic area in which a moving object moves in a working system according to an embodiment of the present invention; [Diagram 2] FIG. 2 is a side view conceptually showing the moving body. [Diagram 3] FIG. 2 is a block diagram conceptually illustrating the above-mentioned operation system. [Figure 4] 5(a) and 5(b) are plan views showing various positions of the moving body within the region. [Diagram 5] 3 is a diagram conceptually showing an image captured by the imaging device of the moving body. FIG. [Figure 6] 5(a) to 5(c) are diagrams conceptually showing captured images when the moving object is located at positions P1, P2, and P3 in FIG. 4(a). [Figure 7] FIG. 5 is a diagram conceptually illustrating a captured image when the moving object is in the position shown in FIG. 4(b). [Figure 8] 5 is a flowchart showing a mobile object position acquisition program stored in a mobile object ECU of the above-mentioned working system. [Figure 9] 1A is a diagram showing the position of the moving object, and FIG. [Figure 10] FIG. 11 is a block diagram conceptually illustrating a work system according to another embodiment of the present invention. [Figure 11] 5 is a flowchart showing a part of a mobile object position acquisition program stored in a mobile object ECU of the above-mentioned working system. [Figure 12] (a) is a flowchart showing a part of the moving object position acquisition program stored in the management ECU of the working system. (b) is a map showing the relationship between the distance between the moving object and the specific part and the number of pixels. (c) is a map showing the relationship between the amount of deviation in the width direction of the moving object and the length between the center lines. [Figure 13] 10A is a flowchart showing an abnormality detection program stored in the vehicle ECU of the above-mentioned working system, and FIG. 10B is a map showing the relationship between the traveling speed of the vehicle and the rate of change in the number of pixels. [Figure 14] 5 is a flowchart showing a collision possibility acquisition program stored in a mobile object ECU of the above-mentioned working system. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0007] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A working system according to an embodiment of the present invention will be described in detail below with reference to the drawings. EXAMPLES

[0008] As shown in FIG. 1, the area R is a work area where, for example, a moving object M moves to perform work (for example, putting objects in and out of shelves, transporting objects, etc.). In the area R, for example, shelves T1, T2, etc. are arranged in a row in one direction. Also, at both ends of the rows of shelves T1, T2, etc., aisles A1, A2 are provided, respectively, extending in a direction substantially perpendicular to the shelves T1, T2, etc., and aisles B1, B2, etc. are provided between adjacent shelves T1, T2, etc. In this embodiment, in the area, the direction in which the shelves T1, T2, etc. extend is the x direction, and the direction in which the shelves T1, T2, etc. are lined up is the y direction. The aisles A1, A2 extend in the y direction, and the aisles B1, B2, etc. extend in the x direction.

[0009] In this embodiment, the portions where the aisles B1, B2... and the aisles A1, A2 meet are referred to as the exits G1, G2..., G11, G12... from the aisles B1, B2... to the aisles A1 or A2. The exits G1, G2..., G11, G12... are determined by the ends of two adjacent shelves T1, T2...

[0010] In addition, in the aisles A1 and A2, a plurality of markers Q1, Q2..., Q11, Q12..., Q21, Q22..., and Q31, Q32... are provided in correspondence with the shelves T1 and T2..., respectively. Each of the plurality of markers Q1, Q2..., Q11, Q12..., Q21, Q22..., and Q31, Q32... can have information that allows the position on the x and y coordinates in the area to be read, for example. Hereinafter, the position on the x and y coordinates in the area can be referred to as an absolute position. For example, each of the markers Q1, Q2..., Q11, Q12..., Q21, Q22..., and Q31, Q32... can have information (barcode or two-dimensional code) indicating the absolute position on the surface, or information that allows the absolute position to be obtained. Hereinafter, when there is no need to distinguish between shelves T1, T2..., aisles B1, B2..., markers Q1, Q2..., etc. or when they are referred to collectively, the subscripts 1, 2..., etc. will be omitted.

[0011] The mobile object M moves within this area R to perform work. During work, the mobile object M often moves linearly along the passage B. In other words, the mobile object M often moves between a pair of shelves T with the exit G located in front of the mobile object M. In the area R, a plurality of mobile objects M often perform work. In addition, a management device C capable of wirelessly communicating with each of the plurality of mobile objects M is provided. The management device C manages the work of the plurality of mobile objects M. The management device C may be located inside the area R or outside the area R.

[0012] The moving body M may be, for example, an autonomous moving body capable of traveling without a human. As shown in FIG. 2, the moving body M includes one front wheel 10, two rear wheels 12, a vehicle body 14, and the like. In the moving body M, the one front wheel 10 is a steering wheel, and the two rear wheels 12 are driving wheels. In addition, the moving body M is provided with a steering device 16 capable of automatically steering the front wheels 10, corresponding to the front wheels 10, and a driving / braking device 18, corresponding to the two rear wheels 12. The driving / braking device 18 includes an electric motor, an inverter, a battery, and the like. The electric motors are provided corresponding to the two rear wheels 12, respectively. The inverters are connected corresponding to the two electric motors, respectively. The battery supplies power to the two electric motors, and the like.

[0013] The rear of the vehicle body 14 serves as a loading platform, and an upright portion 26 is provided in the center of the front in the width direction. A first camera 30, a second camera 32, a moving body ECU 34, and the like are provided in the upright portion 26 as imaging devices. The first camera 30 captures an image of the area in front of the moving body M. The first camera 30 is located in the center of the moving body M in the width direction.

[0014] The second camera 32 captures an image of the marker Q. Based on the image captured by the second camera 32, information attached to the marker Q is read, and the absolute position of the moving object M is acquired.

[0015] The mobile object ECU 34 is mainly a computer and includes an execution unit, a storage unit, an input / output unit, etc. As shown in Fig. 3, the first camera 30, the second camera 32, the inertial measurement unit 40, a GPS (Global Positioning System) receiver 42 as a GNSS (Global Navigation Satellite System) receiver, a mobile object communication device 44, etc. are connected to the input / output unit of the mobile object ECU 34. The steering device 16 is also connected to the input / output unit via a drive circuit (not shown). The inverter of the drive / braking device 18 is also connected to the input / output unit.

[0016] The inertial measurement device 40 detects acceleration acting in the front-rear direction, acceleration acting in the width direction, and acceleration acting in the up-down direction as inertial forces. The inertial measurement device 40 also detects the rotational angular velocity around a front-rear axis extending in the front-rear direction, the rotational angular velocity around a width axis extending in the width direction, and the rotational angular velocity around a vertical axis extending in the up-down direction. Based on the detection values ​​by these inertial measurement device 40, the front-rear acceleration of the moving body M, the yaw rate which is the rotational angular velocity around the vertical axis, and the like can be obtained. Furthermore, by accumulating the front-rear acceleration of the moving body M, the running speed and movement amount in the front-rear direction of the moving body M can be obtained, and by accumulating the yaw rate of the moving body M, the yaw angle of the moving body M can be obtained.

[0017] The GPS receiver 42 receives and processes the GPS signal. In this embodiment, the GPS receiver 42 acquires the position of the moving body M on the earth based on the received GPS signal. In addition, the moving body ECU 34 acquires the absolute position of the moving body M based on the position of the moving body M on the earth and the position of the set point of the region R, which is known, on the earth.

[0018] The mobile communication device 44 is capable of transmitting and receiving wireless information. Each of the mobile units M is provided with identification information ID, and mobile unit information including the identification information ID is transmitted. The mobile units M also receive management information including the identification information ID representing the mobile unit M. The mobile units M perform work according to work plan information included in the management information transmitted from the management device C. Also, the steering device 16, the driving and braking device 18, etc. can be controlled based on a travel command included in the management information transmitted from the management device C.

[0019] The mobile object ECU 34 includes a mobile object position acquisition unit 48, a traveling state control unit 50, etc. The mobile object position acquisition unit 48 includes an external information-based position acquisition unit 52, a traveling state-based provisional position acquisition unit 54, an image-based provisional position acquisition unit 56, a map information storage unit 58, etc. The map information storage unit 58 stores map information of the interior of the area R. The map information includes information on the absolute positions of the shelves T and the absolute positions of the exits G. Strictly speaking, the map information storage unit 58 stores information on the absolute positions of predetermined points such as the ends of the shelves T, information on the absolute positions of predetermined points such as the center of the exits G, and the like.

[0020] The external information-based position acquisition unit 52 acquires the absolute position within the area of ​​the moving body M based on external information, which is information outside the moving body M. The external information includes at least one of information indicating the position of the moving body M on the earth acquired by the GPS receiver 42 and information read from an image of the marker Q captured by the second camera 32. The external information-based position acquisition unit 52 can acquire the absolute position of the moving body M with high accuracy.

[0021] As shown in Fig. 1, the marker Q is provided on the passages A1 and A2. Therefore, when the moving object M is moving on the passages A1 and A2, the absolute position of the moving object M can be acquired with high accuracy by the external information-based position acquisition unit 52. However, when the moving object M leaves the passages A1 and A2 and moves on the passage B, the information of the marker Q cannot be acquired. Therefore, when the GPS receiver 42 cannot receive a GPS signal, it may be difficult for the external information-based position acquisition unit 52 to acquire the absolute position of the moving object M in the area R (hereinafter, may be abbreviated as the moving object position). Even when the moving object M is moving through the passages A1 and A2, the information attached to the marker Q may not be readable due to dirt, peeling, etc. Also, in some areas, the marker Q may not be provided.

[0022] The traveling state-based provisional position acquisition unit 54 provisionally acquires the absolute position of the moving body M based on the measurement values ​​measured by the inertial measurement unit 40, and can be called a dead reckoning-based provisional position acquisition unit. For example, the absolute position of the moving body M can be provisionally acquired by estimating the direction of the moving body M based on the yaw angle of the moving body M and acquiring the amount of movement of the moving body M by accumulating the traveling speed of the moving body M.

[0023] Even when it is difficult to acquire the moving object position by the external information-based position acquisition unit 52, the moving object position can be provisionally acquired by the running state-based provisional position acquisition unit 54. However, compared to the case where the external information-based position acquisition unit 52 is used, the accuracy of acquiring the moving object position is low when the running state-based provisional position acquisition unit 54 is used.

[0024] The image-based tentative position acquisition unit 56 provisionally acquires the absolute position of the moving object M based on the image captured by the first camera 30. The moving object M often moves toward the exit G (with the front surface of the moving object M facing the exit G) along the passage B between a pair of shelves Ti, Tj. Therefore, the first camera 30 often captures an image of a portion including the exit G as a specific portion inside the region R, and a captured image S including the exit image SG as a specific image is obtained.

[0025] Fig. 5 shows an example of a captured image S captured by the first camera 30 when a moving object M is located in the center of the passage B toward the exit G. As shown in Fig. 5, an exit image SG is located in the center of the captured image S, and shelves Ti, Tj are located on both sides of the exit image SG. In addition, edges LTi, LTj, which are the contours of each of a pair of adjacent shelves Ti, Tj, are symmetrical with respect to a center line LG of the exit image SG. i is 1, 2, 3,..., and j is i+1.

[0026] The exit G is a space, but the exit image SG can be defined by the portions LTgi, LTgj extending in the vertical direction of the edges LTi, LTj of the shelves Ti, Tj, and the virtual lines LGc, LGd determined by the corners Ci, Cj, Di, Dj of the edges LTi, LTj. As a result, it is possible to acquire the size (number of pixels) of the exit image SG, and the center line LG as the reference line of the exit image SG can be determined. In this embodiment, the position of the center line LG is represented by the deviation Δβ of the center line LG from the center line LS of the captured image S. Note that the deviation of the exit G from the center line of the first camera 30 of the moving body M corresponds to the deviation of the center line LS of the captured image S from the center line LG of the exit image G, but the position of the center line LG of the exit image G in the captured image S is more appropriately represented by the deviation Δβ of the center line LG of the exit image G from the center line LS of the captured image S, so it is called that. On the other hand, the position of the center line LG of the exit image G can also be represented by the deviation of the center line LS of the captured image S from the center line LG of the exit image G.

[0027] The deviation Δβ can be expressed by the number of pixels aligned along the deviation Δβ and the direction of deviation (rightward or leftward) of the center line LG from the center line LS. The deviation Δβ can be a positive value or a negative value. In the captured image S shown in FIG. 5, the center lines LG and LS are aligned, so the deviation Δβ is 0.

[0028] The image-based tentative position acquisition unit 56 includes an image recognition unit 60 and a tentative relative position acquisition unit 62. The image recognition unit 60 has a trained neural network, and recognizes an exit image SG from the captured image S, acquires the size of the exit image SG, and acquires the position of a center line LG of the exit image SG.

[0029] From the captured image S, edges LTi, LTj, which are the contours of a pair of shelves Ti, Tj, and corners Ci, Cj, Di, and Dj of edges LTi, LTj are recognized by a method such as filtering. In addition, a virtual line LGc connecting mutually opposing corners Ci, Cj among corners Ci, Cj, Di, and Dj of edges LTi, LTj, and a virtual line LGd connecting mutually opposing corners Di, Dj are determined. As a result, an exit image SG surrounded by virtual lines LGc, LGd, and LTgi, LTgj, which are parts of edges LTi, LTj, can be specified. For example, an image recognition unit 60 having a trained neural network can be created by inputting a large number of captured images S and information associated with the captured images S (the extracted exit images SG, the number of pixels in the exit image SG, the deviation Δβ of the center line LG of the exit image SG from the center line LS of the captured image S, etc.) into a well-known machine learning algorithm such as a neural network.

[0030] The provisional relative position acquisition unit 62 provisionally acquires the relative positions in the x and y directions between the moving body M and the exit G based on the characteristics of the first camera 30, the number of pixels of the exit image SG, and the position of the center line LG of the exit image SG. Strictly speaking, the relative position between the center point of the first camera 30 of the moving body M and a line (part) on the center line of the exit G (corresponding to the reference line of the exit image SG) is acquired. The x direction is the approach / separation direction between the moving body M and the exit G, and the y direction is the width direction of the exit G. The characteristics of the first camera 30 can be expressed, for example, by the angle of view, the magnification / reduction rate, etc. Based on these characteristics of the first camera 30, the relationship between the number of pixels N occupied by the exit image SG and the distance in the x direction between the exit G and the moving body M is determined. In other words, based on the characteristics of the first camera 30, etc., the change gradient of the number of pixels N occupied by the exit image SG with respect to the change in the distance in the x direction between the exit G and the moving body M is determined. In this embodiment, the change gradient of the number of pixels N of this exit image SG is acquired and stored in advance.

[0031] 6(a)-(c), the number of pixels occupied by the exit image SG in the captured image S increases in the order of Fig. 6(a), (b), and (c). In addition, the shapes of the edges LTi and LTj of a pair of shelves Ti and Tj are symmetrical with respect to the center line LG of the exit image SG, and the center line LG of the exit image SG almost coincides with the center line LS of the captured image S. Therefore, it can be seen from the captured images S shown in Fig. 6(a)-(c) that the moving object M is moving forward in a straight line toward the exit G at almost the center of the width direction of the passage B.

[0032] The center line LS of the captured image S is on the axis of the first camera 30, but the first camera 30 is provided at the center of the moving body M. Therefore, the center line LS of the captured image S can be considered to be the center of the moving body M in the width direction. From this, when the shapes of the edges LTi, LTj of a pair of shelves Ti, Tj are symmetrical and the center line LG of the exit image SG and the center line LS of the captured image S are almost coincident, it can be seen that the center of the moving body M is located on the center of the exit G in the width direction. In this embodiment, when the moving body M is located at each of P1, P2, and P3 in FIG. 4(a), the captured images S captured by the first camera 30 respectively correspond to FIGS. 6(a)-(c).

[0033] Also, as shown in FIG. 7, in the captured image S, the center line LG of the exit image SG is offset from the center line LS of the captured image S. From this, it can be seen that the moving object M is at a position offset in the width direction from the center of the exit G as shown in FIG. 4(b). Also, as shown in FIG. 1, the width W of the exit G is known. Therefore, based on the width W, the length wg of the exit image SG in the width direction on the captured image, the deviation Δβ of the center line LG of the exit image SG from the center line LS of the captured image S (the deviation direction, the deviation amount of the center line LG from the center line LS of the captured image S), etc., the offset amount and the offset direction of the first camera 30 of the moving object M from the center of the exit G can be obtained. The deviation amount can be expressed as the length between the center line LG of the exit image SG and the center line LS of the captured image S.

[0034] Then, in the image-based provisional position acquisition unit 56, the absolute position of the moving body M within the area R is provisionally acquired based on the relative position between the moving body M and the exit G acquired based on the captured image S and the absolute position of the exit G stored in the map information storage unit 58.

[0035] Hereinafter, in this specification, the absolute position of the moving body M acquired by the external information-based position acquisition unit 52 will be referred to as the "external information-based acquired position," the absolute position of the moving body M provisionally acquired by the driving state-based provisional position acquisition unit 54 will be referred to as the "driving state-based provisional acquired position," and the absolute position of the moving body M provisionally acquired by the image-based provisional position acquisition unit 56 will be referred to as the "image-based provisional acquired position."

[0036] As described above, in this embodiment, when the absolute position of the moving object M can be acquired by the external information-based position acquisition unit 52, the external information-based acquired position is regarded as the final absolute position of the moving object M. However, when it is difficult for the external information-based position acquisition unit 52 to acquire the absolute position of the moving object M, the traveling state-based provisional position acquisition unit 54 acquires the provisional absolute position of the moving object M in parallel with the image-based provisional position acquisition unit 56 acquires the provisional absolute position of the moving object M. Then, the final absolute position of the moving object M is acquired based on both the traveling state-based provisional acquired position and the image-based provisional acquired position.

[0037] In detail, after the absolute value of the difference between the absolute position of the moving body M at the time when it becomes difficult for the external information-based position acquisition unit 52 to acquire the absolute position of the moving body M and the image-based provisional acquired position acquired thereafter becomes larger than a predetermined set value, the final absolute position of the moving body M is acquired based on the traveling-state-based provisional acquired position and the image-based provisional acquired position. The absolute position of the moving body M at the time when it becomes difficult for the external information-based position acquisition unit 52 to acquire the absolute position of the moving body M can be the absolute position of the moving body M last, in other words, acquired immediately before by the external information-based position acquisition unit 52. Also, at the present time, after the absolute value of the difference between at least one of the size of the exit image SG of the captured image S captured by the moving object M and the deviation Δβ of the center line LG from the center line LS, and at least one of the size of the exit image SG extracted from the captured image S when the absolute position was last acquired by the external information-based position acquisition unit 52 and the deviation Δβ of the center line LG of the exit image SG from the center line LS of the captured image S becomes larger than a set value, the final moving object position is acquired based on the traveling state-based provisional acquisition position and the image-based provisional acquisition position. For example, the position obtained by combining the traveling state-based provisional acquisition position (xr, yr) and the image-based provisional acquisition position (xd, yd) using a weighting coefficient α can be set as the final position (x, y) of the moving object M. The weighting coefficient α is a value of 0 or more and 1 or less. x=α*xr+(1-α)*xd y=α*yr+(1-α)*yd

[0038] The traveling condition control unit 50 controls the traveling condition of the moving body M. The traveling condition control unit 50 can control the driving and braking device 18 and the steering device 16 so that the moving body M travels according to the work plan information, for example. Furthermore, when the management information transmitted from the management device C includes a traveling command, the traveling condition control unit 50 can control the driving and braking device 18 and the steering device 16 according to the traveling command.

[0039] The management device C includes a management ECU 70, a management communication device 72, etc. The management ECU 70 manages a plurality of moving objects M performing work within an area R, and is mainly a computer including an execution unit, a storage unit, an input / output unit, etc. The management ECU 70 includes a work plan information storage unit 74, etc., and controls each of the multiple moving objects M based on the work plans for each of the multiple moving objects M stored in the work plan information storage unit 74. The management communication device 72 is connected to the management ECU 70. The management communication device 72 is capable of wirelessly communicating information, and transmits, for example, management information including work plan information.

[0040] In this embodiment, the mobile object ECU 34 executes a mobile object position acquisition program shown in the flow chart of FIG. 8 at predetermined set time intervals. In step 1 (hereinafter, simply abbreviated as S1; the same applies to the other steps), it is determined whether the moving object M is moving. For example, if the running speed v acquired based on the measurement value of the inertial measurement unit 40 is greater than a set speed, it is determined that the moving object M is moving.

[0041] In S2, it is determined whether or not a GPS signal has been received by the GPS receiver 42 within a predetermined set time, and whether or not information on the marker Q has been read based on the image captured by the second camera 32. In other words, it is determined whether or not the absolute position of the moving body M can be acquired based on the GPS signal received by the GPS receiver 42, and whether or not the absolute position of the moving body M can be acquired by reading the signal of the marker Q. If at least one of the acquisition of the absolute position based on the GPS signal and the acquisition of the absolute position based on the signal of the marker Q is possible, the determination in S2 is YES, and if neither is possible, the determination in S2 is NO. If the determinations in S1 and S2 are both YES, the external information-based position acquisition unit 52 acquires the final moving object position based on at least one of the GPS signal and the information read from the marker Q in S3.

[0042] On the other hand, if the determination in S2 is NO, it is determined in S4 whether or not the initial processing has been completed. As will be described later, it is determined whether or not the initial processing required for the image-based tentative position acquisition unit 56 to tentatively acquire the moving object position has been completed. When S4 is executed for the first time, the determination is NO.

[0043] In S5, acquisition of the moving object position by the traveling state-based provisional position acquisition unit 54 is started, and in S6, imaging is performed by the first camera 30 to obtain a captured image S. Then, in S7, it is determined whether the exit image SG is recognizable (extractable) in the captured image S. If the determination is NO, in S8, the moving object position provisionally acquired by the traveling state-based provisional position acquisition unit 54 is set as the final moving object position.

[0044] If the determination in S7 is YES, in S9, the shortest exit Gs that is the shortest exit G for the moving body M is determined based on the moving body position most recently acquired by the external information-based position acquisition unit 52, the yaw angle at that time, map information, etc. The map information includes information on the absolute position of each exit G, so the shortest exit Gs can be determined based on the moving body position most recently acquired by the external information-based position acquisition unit 52. This makes it possible to suppress a decrease in the acquisition accuracy of the moving body position as early as possible.

[0045] Then, in S10, the distance and direction between the moving object M and the shortest exit Gs are acquired. The traveling state control unit 50 controls the steering device 16, the driving and braking device 18, etc. so that the moving object M moves toward the shortest exit Gs. Next, in S11, it is determined whether it has become possible for the external information-based position acquisition unit 52 to acquire the moving object's position. In other words, it is determined whether the GPS receiver 42 has received a GPS signal well, and whether it has become possible to read the information of the marker Q well. If the determination is YES, in S12, the traveling state-based provisional position acquisition unit 54 ends provisional acquisition of the absolute position of the moving object M, and in S3, the external information-based position acquisition unit 52 acquires the moving object's position.

[0046] If the determination in S11 is NO, in S13, the exit image SG is recognized based on the captured image S first obtained by the first camera 30, and an initial value Ns of the number of pixels of the exit image SG and an initial value Δβs of the deviation Δβ of the center line LG of the exit image SG from the center line LS of the captured image S are obtained. This is the initial process. Incidentally, the determination in S2 is often YES when the moving object M is traveling in the passage B. Therefore, it is considered that the initial value Ns of the number of pixels of the exit image SG and the initial value Δβs of the deviation Δβ of the center line LG of the exit image SG from the center line LS of the captured image S are often obtained satisfactorily based on the captured image S acquired in S6.

[0047] Next, when this program is executed, the initial processing is completed, and the determination in S4 is YES. In S14, the first camera 30 captures a captured image S at the current time, and in S15, the exit image SG is recognized based on the captured image S, the number of pixels N and the center line LG of the exit image SG are determined, and the deviation Δβ of the center line LG from the center line LS of the captured image S is acquired. In S16, the absolute values ​​of the differences dN and dΔβ between the number of pixels N and deviation Δβ (which can be referred to as current values) acquired based on the captured image S at the current time and their initial values ​​Ns and Δβs are acquired. dN = |Ns-N| dΔβ=|Δβs-Δβ|

[0048] In S17, it is determined whether the absolute value of the difference dN and the absolute value of dΔβ are greater than the threshold values ​​dNth and dΔβth, respectively. If the determination is YES, in S18, the relative position of the moving body M and the exit G in the x direction is acquired based on the number of pixels N of the exit image SG, the initial value Ns, the distance between the moving body M and the exit G when the initial value is acquired, the increasing gradient of the number of pixels of the exit image SG, etc., and the absolute position of the moving body M in the x direction is provisionally acquired based on this relative position and the absolute position of the exit G in the x direction. In addition, the relative position of the moving body M in the y direction with respect to the center line of the exit G is acquired based on the deviation Δβ of the center line LG of the exit image SG from the center line LS of the captured image S, etc., and the absolute position of the moving body M in the y direction is provisionally acquired based on this relative position and the absolute position of the exit G in the y direction. Then, in S19, the final moving body position is acquired based on the moving body position provisionally acquired by the driving condition-based provisional position acquisition unit 54 and the moving body position provisionally acquired by the image-based provisional position acquisition unit 56.

[0049] In this manner, in this embodiment, the final moving body position is acquired based on the moving body position provisionally acquired by the traveling state-based provisional position acquisition unit 54 and the moving body position provisionally acquired by the image-based provisional position acquisition unit 56. Therefore, even if it is difficult to acquire the moving body position based on a GPS signal or the like, it is possible to suppress a decrease in the accuracy of acquiring the moving body position. In addition, the exit image SG is extracted from the captured image S, and the position of the moving object M can be obtained by calculation based on the size of the exit image SG and the position of the center line LG. As a result, it is possible to easily obtain the moving object position compared to the case where a similar image is selected from a plurality of captured images S. Furthermore, by using a trained neural network, there is an advantage that it is not necessary to store a large number of captured images in the memory unit of the vehicle ECU 34.

[0050] Furthermore, within this region, the position of the moving object is acquired by utilizing the fact that the moving object moves along the passage between a pair of shelves Ti, Tj toward the exit G. Furthermore, the edges of the shelves Ti, Tj are clearly distinguishable. Therefore, it is possible to further suppress the decrease in the accuracy of acquiring the moving object position.

[0051] In the above embodiment, the part that stores and executes S9 of the moving object position acquisition program of the moving object ECU 34 constitutes the identification part determination part. Also, the provisional relative position acquisition part 62 corresponds to the relative position acquisition part. The initial value Ns of the image SG corresponds to the past value, and the distance between the moving object M and the exit G when the position of the moving object M was last acquired by the external information-based position acquisition part 52 corresponds to the set distance.

[0052] Note that a plurality of sets of data representing the distance in the x direction between the moving object M and the exit G and data on the number of pixels N of the exit image SG can be stored in the storage unit of the moving object ECU 34. In this case, the distance in the x direction between the moving object M and the exit G can be obtained based on these plurality of data sets and the number of pixels N of the exit image SG acquired based on the actual captured image S.

[0053] In the above embodiment, the relative positions in the x and y directions between the moving body M and the exit G are acquired based on both the number of pixels N of the exit image SG in the captured image S and the position of the center line LG of the exit image SG, but this is not essential. It is also possible to acquire one of the relative position in the x direction and the relative position in the y direction between the moving body M and the exit G without acquiring the other.

[0054] Furthermore, it is not essential to provide the provisional relative position acquisition unit 62. For example, in addition to the captured image S, information on the relative position between the exit G and the moving body M can be input to the neural network to learn it. In this embodiment, the trained neural network outputs the relative position between the exit G and the moving body M based on the captured image S. Also, the trained neural network corresponds to the relative position acquisition unit.

[0055] Moreover, it is not essential to provide the first camera 30 at the center in the width direction of the moving body M. The first camera 30 can be provided at a position offset from the center in the width direction of the moving body M. In that case, the relative position between the first camera 30 of the moving body M and a reference point separated from the center of the exit G is acquired. Furthermore, the relative positional relationship between the center of the moving body M and the center of the exit G can be acquired based on the relative positional relationship between the first camera 30 of the moving body M and the reference point of the exit G and the offset amount between the center of the moving body M and the first camera 30.

[0056] Furthermore, the orientation of the moving object M can be obtained based on the captured image S. For example, as shown in Fig. 9(a) and (b), when the edges LTi, LTj of a pair of shelves Ti, Tj in the captured image S are asymmetric with respect to the center line LG of the exit image SG, and the center line LG of the exit image SG is shifted from the center line LS of the captured image S, it can be obtained that the moving object M is inclined with respect to the aisle B. Furthermore, based on the degree of asymmetry, the shift Δβ of the center line LG from the center line LS, the distance between the moving object M and the exit G, etc., the inclination of the moving object M with respect to the aisle B can be estimated. EXAMPLES

[0057] In the above embodiment, the absolute position of the moving object M is acquired by the moving object ECU 34, but it may also be acquired by the moving object M and the management device C.

[0058] 10, the mobile object ECU 34 is provided with a map information storage unit 58 and an image recognition unit 60, and the management ECU 70 is provided with a relationship storage unit 82 and a tentative relative position acquisition unit 84. The map information storage unit 58, the image recognition unit 60, the relationship storage unit 82, the tentative relative position acquisition unit 84, etc. configure an image-based tentative position acquisition unit 86. Note that in the first and second embodiments, components corresponding to each other are denoted with the same reference numerals and descriptions thereof will be omitted.

[0059] The relationship memory unit 82 of the management ECU 70 stores a first relationship which is the relationship between the number of pixels N of the exit image SG shown in Figure 12(b) and the distance in the x-direction between the moving body M and the exit G, and a second relationship which is the relationship between the deviation Δβ of the center line LG from the center line LS shown in Figure 12(c) and the deviation in the width direction of the moving body M.

[0060] In each of S13 and S16 in the flowchart of FIG. 8 in the above embodiment, the vehicle ECU 34 executes a routine shown in the flowchart of FIG. 11, and the management ECU 70 executes a routine shown in the flowchart of FIG. 12(a). In S21, the image recognition unit 60 acquires the number of pixels N of the exit image SG and the deviation Δβ of the center line LG of the exit image SG from the center line LS of the captured image S based on the captured image S. In S22, mobile object information including the number of pixels N, the deviation Δβ, information indicating the absolute position of the exit G, identification information, etc. is created and output to the mobile object communication device 44. The mobile object communication device 44 transmits the mobile object information.

[0061] In S31, it is determined whether or not moving body information has been received. If the determination is YES, in S32, the moving body M is identified from the identification information of the moving body information, and the number of pixels N of the exit image SG, the position of the center line LG of the exit image SG, and the absolute position of the exit G are acquired. In S33, based on the number of pixels N, the position of the center line LG, and the first relationship and the second relationship stored in the relationship storage unit 82, the relative position in the x direction between the moving body M and the exit G is acquired. In addition, based on the acquired relative position and the absolute position of the exit G, the absolute position of the moving body M is provisionally acquired. Then, in S34, management information including the image-based provisional acquisition position acquired in S33, the identification information identifying the moving body M, and the like is created and output to the management communication device 72. The management communication device 72 transmits the management information.

[0062] In S23, the moving object ECU 34 determines whether or not management information including its own identification information has been received. If the determination is YES, in S24, the image-based provisional acquisition position is acquired from the management information.

[0063] The final moving object position may also be obtained by the management device C. For example, the mobile object information including the pixel number N of the exit image SG, the deviation Δβ of the center line LG from the center line LS, the absolute position of the exit G, the provisionally acquired position based on the traveling state, etc. is transmitted from the mobile object ECU 34 to the management device C. An image-based provisional acquisition position is acquired based on the number of pixels N of the image SG, the deviation Δβ of the center line LG from the center line LS, and the absolute position of the exit G. Then, a final absolute position of the moving body M is acquired based on this acquired image-based provisional acquisition position and the traveling-state-based provisional acquisition position included in the moving body information. In addition, the final absolute position of the moving body M acquired in the management device C may be stored in the management device C.

[0064] Furthermore, the mobile body ECU 34 can obtain, based on the captured image, the presence or absence of abnormalities in the inertial measurement unit 40, or in the inertial measurement unit 40 and the parts that obtain the running state of the mobile body M based on the measurement values ​​of the inertial measurement unit 40 of the mobile body ECU 34 (hereinafter referred to as the inertial measurement unit 40, etc.). 13(b), if the traveling speed of the moving body M in the x direction (corresponding to the approaching / leaving speed of the moving body M with respect to the exit G) acquired based on the measurement value by the inertial measurement unit 40 and the rate of change of the number of pixels N occupied by the exit image SG are outside a range determined by a predetermined relationship K, it can be determined that there is an abnormality in the inertial measurement unit 40, etc. For example, the range determined by the relationship K is the range determined by the dashed line, and if the traveling speed of the moving body M and the rate of change of the number of pixels N are located in the shaded area, it can be determined that there is an abnormality in the inertial measurement unit 40, etc.

[0065] The abnormality detection program shown in the flowchart of FIG. 13(a) is executed in the vehicle ECU 34 at predetermined intervals. In S41, the traveling speed of the moving body M is detected based on the measurement value of the inertial measurement device 40, and in S42, it is determined whether the traveling speed v is greater than a set speed vth. If the determination is YES, in S43, a captured image S is acquired, and in S44, it is determined whether the increase ΔN in the pixel count N of the exit image SG is greater than a set number ΔNath. If the determination is NO, it is determined that the inertial measurement device 40, etc., is abnormal. It is also possible to configure the inertial measurement unit 40 to be determined as abnormal if the traveling speed v is lower than a set speed vth and the increase ΔN in the number of pixels N is larger than a set number ΔNcth. Furthermore, the management device C can be notified of any abnormality in the inertial measurement device 40 or the like.

[0066] In this embodiment, the traveling state acquisition device is configured by the inertial measurement unit 40 and a part of the moving body ECU 34 that acquires the traveling speed of the moving body M based on the measurement value of the inertial measurement unit 40. Also, the abnormality detection unit is configured by a part of the moving body ECU 34 that stores and executes the abnormality presence / absence detection program represented by the flowchart in Fig. 13(a).

[0067] In the above embodiment, the specific part is an exit, but the specific part may be an object. For example, if the number of pixels N of a specific image representing the specific part included in the captured image S increases suddenly, it can be determined that the moving body M and the specific part have suddenly approached each other, and that there is a possibility of a collision.

[0068] The collision possibility program shown in the flowchart of FIG. 14 is executed in the moving object ECU 34 at predetermined set time intervals. In S51, a captured image S is acquired, and in S52, it is determined whether an increase ΔN in the current pixel count N from the previous pixel count N is greater than a set number ΔNbth. If the determination is YES, in S53, it is determined that there is a possibility of collision between the moving body M and the specific part. If it is determined that there is a possibility of collision, the moving body M can be made to avoid the specific part by the control of the driving / braking device 18 and the steering device 16 by the traveling state control unit 50.

[0069] In the above embodiment, the collision possibility acquisition section is constituted by a part of the moving body ECU 34 that stores and executes the collision possibility acquisition program shown in the flowchart of FIG.

[0070] In addition, the present invention can be embodied in various forms with various modifications and improvements based on the knowledge of those skilled in the art. [Explanation of symbols]

[0071] 30: First camera 32: Second camera 34: Mobile ECU 40: Inertial measurement unit 42: GPS receiver 44: Mobile communication device 48: Mobile position acquisition unit 50: Driving state control unit 52: External information-based position acquisition unit 54: Driving state-based provisional position acquisition unit 56: Image-based provisional position acquisition unit 58: Map information storage unit 60: Image recognition unit 70: Management ECU 72: Management communication device 74: Work plan information storage unit 82: Relationship storage unit 86: Mobile position acquisition unit

[0072] (1) A work system in which a mobile object moves within a predetermined area to perform work, An operation system including a relative position acquisition unit that acquires the relative position between the moving body and a specific part corresponding to the specific image within the area, based on at least one of the size of a specific image included in an image captured by an imaging device provided on the moving body and the position of a reference line of the specific image.

[0073] The specific part may or may not be an object. For example, if the specific part is an object, the specific part may be a stationary object or a movable object. Also, if the specific part is not an object, the specific part may be a part surrounded by an object or a part that can be identified by the edge of the object or the like. Specifically, when there are multiple rows of shelves or walls within an area and a moving object moves along the shelves or walls to perform work, the specific part can be the exit of a passage between a pair of adjacent shelves or walls among the multiple rows of shelves or walls (a part determined by the edge of the adjacent shelves or walls), or the like.

[0074] The reference line may be set anywhere on the specific image. For example, the reference line may be a center line in the width direction, or a line passing through an edge of the specific image. The position of the reference line of the specific image in the captured image may be represented by a deviation from a predetermined line such as the center line of the captured image or an edge of the captured image.

[0075] The relative position includes at least one of the relative position in the approaching / separating direction between the specific part and the moving body and the relative position in the width direction between the specific part and the moving body. The width direction is the direction perpendicular to the approaching / separating direction. The relative position in the approaching / separating direction between the specific part and the moving body can be expressed by these distances. The relative position in the width direction between the specific part and the moving body can be expressed by the deviation in the width direction of the moving body with respect to the part of the specific part corresponding to the reference line.

[0076] Based on the relative positions of the moving body and the specific part, it is possible to obtain the position of the moving body, to obtain whether or not the moving body is abnormal, and to obtain whether or not there is a possibility that the moving body will collide with the specific part. For example, when the position of the identifying part is known, the position of the moving body can be obtained based on the relative position between the moving body and the identifying part and the position of the identifying part. Furthermore, as the distance between the moving body and the specific part becomes shorter, the specific image in the captured image becomes larger. However, for example, if the size of the specific image does not change in relation to the approach / departure speed of the moving body relative to the specific part acquired by the traveling state acquisition device of the moving body, it can be determined that the traveling state acquisition device of the moving body, etc., is abnormal. Furthermore, if the specific image in the captured image suddenly becomes larger, it can be determined that the moving object is suddenly approaching the specific portion, and therefore there is a high possibility that the moving object will collide with the specific portion.

[0077] (2) The work system described in (1), wherein the relative position acquisition unit includes an image recognition unit capable of recognizing the specific image from the captured image, and acquires the relative position between the moving body and the specific unit based on at least one of the size of the specific image recognized by the image recognition unit and the position of a reference line of the specific image.

[0078] The image recognition unit can be, for example, a trained neural network that has the function of being able to recognize a specific image from a captured image through training.

[0079] (3) The work system described in (1), wherein the relative position acquisition unit acquires that the distance between the moving body representing the relative position and the identification unit is shorter when the number of pixels representing the size of the specific image is greater than a set number than when the number is equal to or less than the set number.

[0080] As shown in FIG. 12(b), it is clear that when the number of pixels is greater than the set number, the distance between the moving object and the specified portion is shorter than when the number is equal to or less than the set number.

[0081] (4) A work system as described in (1) or (2), wherein the relative position acquisition unit acquires the relative position of the moving body with respect to the specific part based on a past value which is the size of the specific image in the captured image when the distance between the specific part and the moving body is a set distance, a current value which is the size of the specific image in the captured image currently captured, and characteristics of the imaging device.

[0082] The set distance can be the distance (known) at the time when acquisition of the relative position between the identifying unit and the moving body based on the captured image is started, or the distance when the relative position between the identifying unit and the moving body was acquired last time. The past value is the size of the identifying image in the captured image when the relative position between the identifying unit and the moving body is known. In the above embodiment, the past value corresponds to the initial value. The characteristics of the imaging device are determined, for example, by the angle of view, the magnification / reduction ratio, etc. The characteristics of the imaging device determine the gradient of change in the size of a specific image in a captured image relative to a change in the distance between a moving object and a specific part.

[0083] In the work system described in this section, the distance of the moving object in the approaching or leaving direction to a specific part can be obtained by calculation.

[0084] (5) A work system described in any one of (1) to (3), wherein the relative position acquisition unit includes a relationship memory unit that stores the relationship between the distance between the moving body and the specific part and the size of the specific image, and acquires the distance between the moving body and the specific part, which represents the relative position, based on the relationship stored by the relationship memory unit and the size of the specific image in the captured image.

[0085] The relationship between the distance between the moving body and the identifying unit and the size of the identified image may be stored in the moving body or in the management device.

[0086] (6) A work system described in any one of (1) to (5), wherein the relative position acquisition unit acquires that the amount of widthwise deviation of the specific part of the imaging device of the moving body from the part corresponding to the reference line of the specific image, which represents the position of the reference line of the specific image, from the center line of the captured image is greater than a set value than when the amount of deviation is equal to or less than the set value.

[0087] The position of the reference line of the specific image can be expressed by the deviation of the reference line of the specific image from the center line of the captured image. As shown in FIG. 12(c), there are cases where the reference line of the specific image is deviated to the right or to the left with respect to the center line of the captured image. Based on the direction of deviation of the reference line of the specific image from the center line of the captured image, the direction of deviation of the specific part of the moving body in the width direction with respect to the part corresponding to the reference line of the specific image can be determined. Also, it can be seen that when the deviation amount of the reference line of the specific image from the center line of the captured image (the length between the reference line of the specific image and the center line of the captured image) is larger than a set value, the deviation amount of the moving body from the line (part) corresponding to the reference line of the specific part is larger than when it is smaller than a set value.

[0088] (7) A work system described in any one of items (1) to (6), wherein the work system obtains the orientation of the moving object relative to a specific part based on the deviation of a reference line of the specific image, which represents the position of the reference line of the specific image, from the center line of the captured image and the symmetry of the captured image relative to the center line of the specific image.

[0089] For example, in a case where a pair of objects are present symmetrically on both sides of a specific part and a moving body moves through a passage between the pair of objects, if the axis of the imaging device of the moving body is parallel to the passage, the object images corresponding to the pair of objects (a shelf or a wall) in the captured image will be symmetrical with respect to the center line of the specific image. In contrast, if the axis of the imaging device of the moving body is inclined with respect to the passage, the pair of object images will be asymmetric. In addition, if the axis of the imaging device of the moving body is inclined with respect to the passage, the distance between the reference line of the specific image and the axis of the captured image will be a different length from when the axis of the imaging device is parallel to the passage, even if the position of the moving body is the same.

[0090] (8) A work system described in any one of (1) to (7), wherein the relative position acquisition unit acquires the relative position between the moving body and the specific unit at the current time when the absolute value of the difference between at least one of the size of the specific image of the captured image captured by the imaging device at the start of acquiring the moving body position and the position of a reference line of the specific image, and at least one of the size of the specific image of the captured image at the current time and the position of a reference line of the specific image, is greater than a set value.

[0091] (9) The operation system includes a management device that manages the plurality of moving objects, the mobile object includes a mobile communication device capable of wirelessly transmitting and receiving information; The management device includes a management communication device capable of wirelessly transmitting and receiving information, The work system described in any one of items (1) to (8), wherein the relative position acquisition unit acquires the relative position between the mobile body and the identification unit through communication between the mobile body communication device and the management communication device.

[0092] (10) The moving object includes an image recognition unit that extracts the specific image from the captured image and acquires at least one of a size of the specific image and a position of a reference line of the specific image; the mobile communication device wirelessly transmits mobile information including information representing at least one of a size of the specific image acquired by the image recognition unit and a position of a reference line of the specific image; the management device includes a relationship storage unit that stores at least one of a first relationship, which is a relationship between a size of the specific image and a distance between the moving body and the specific portion, and a second relationship, which is a relationship between a position of a reference line of the specific image and a widthwise deviation of the specific portion of the moving body relative to a portion corresponding to the reference line; the management communication device is capable of receiving the mobile object information, The work system described in (9) above, wherein the relative position acquisition unit is provided in the management device and acquires the relative position between the moving body and the identification unit based on at least one of the relationships stored in the relationship memory unit and at least one of the size of the specific image and the position of the reference line of the specific image contained in the moving body information received by the management device communication device.

[0093] The position of the mobile object acquired in the management device can be provided to the mobile object and stored in the management device.

[0094] (11) A work system described in any one of items (1) to (10), wherein the work system includes a moving body position acquisition unit that acquires a moving body position, which is the position of the moving body within the area, based on the relative position between the moving body and the identification unit acquired by the relative position acquisition unit.

[0095] When the position of the identifying unit is known, the position of the moving body can be obtained based on the position of the identifying unit and the relative position between the identifying unit and the moving body. The position of the moving body can be an absolute position within the moving body's domain (e.g., it can be expressed as a position on x, y coordinates), or it can be an absolute position on the Earth (it can be expressed as latitude and longitude), etc.

[0096] (12) The moving object position acquisition unit, an image-based tentative position acquisition unit that acquires a position of the moving object based on the relative position between the moving object and the identification unit acquired by the relative position acquisition unit and the position of the identification unit; a running state-based provisional position acquisition unit that provisionally acquires a position of the moving object based on a running state of the moving object; and acquiring the moving body position, which is the final position of the moving body, based on the tentative position of the moving body provisionally acquired by the driving state-based tentative position acquisition unit and the tentative position of the moving body provisionally acquired by the image-based tentative position acquisition unit.

[0097] (13) The operation system includes an external information-based position acquisition unit capable of acquiring a position of the moving object based on information from outside the moving object, The working system according to claim (11) or (12), wherein the mobile body position acquisition unit acquires the position of the mobile body when the external information-based position acquisition unit is no longer able to acquire the position of the mobile body, and includes an identification unit determination unit that determines the identification unit based on the position of the mobile body acquired just before the external information-based position acquisition unit is no longer able to acquire the position of the mobile body.

[0098] For example, in a case where there are multiple rows of shelves within an area, the spaces between adjacent shelves among the multiple rows of shelves are regarded as aisles, and one of the exits of each of the multiple aisles is regarded as a specific part, when it becomes impossible to acquire the position of a moving object based on external information, it is possible to determine the shortest exit, which is the closest exit, based on the position of the moving object acquired last before it became impossible to acquire the position of the moving object (the position of the moving object acquired just before it became impossible to acquire the position of the moving object), and the shortest exit can be regarded as the specific part. This makes it possible to quickly suppress a decrease in the estimation accuracy of the moving object position.

[0099] (14) A work system described in any one of items (1) to (13), wherein the work system includes a collision possibility acquisition unit that acquires a collision possibility between the moving body and the specific part based on a change in size of the specific image acquired by the relative position acquisition unit.

[0100] (15) The moving body includes a running state acquisition device that acquires a running state of the moving body, The work system described in any one of items (1) to (14) includes an abnormality detection unit that detects that the traveling condition acquisition device is abnormal if the approach / departure speed of the moving body relative to the specific part and the speed of change in size of the specific image acquired by the traveling condition acquisition device fall outside a range determined by a predetermined relationship.

[0101] The running state acquisition device may include a sensor for detecting the running state of the moving body and a part for processing the sensor value mainly composed of a computer. In the above embodiment, the running state acquisition device is configured by the inertial measurement device and a part of the moving body ECU, etc.

[0102] (16) A mobile object position acquisition system for acquiring a mobile object position, which is a position of a mobile object within a predetermined area, comprising: A mobile body position acquisition system including a mobile body position acquisition unit that acquires the relative position between a specific part corresponding to a specific image within the area and the mobile body based on at least one of the size of a specific image included in an image captured by an imaging device provided on the mobile body and the deviation of a reference line of the specific image from the center line of the captured image, and acquires the position of the mobile body.

[0103] The mobile object position acquisition system described in this section may employ any of the technical features described in (1) to (15).

Claims

1. A work system in which a mobile object moves within a predetermined area to perform work, a relative position acquisition unit that acquires a relative position between the moving body and a specific portion corresponding to the specific image in the region based on at least one of a size of the specific image included in an image captured by an imaging device provided on the moving body and a position of a reference line of the specific image, A work system in which the relative position acquisition unit acquires the relative position of the moving body with respect to the specific part based on a past value which is the size of the specific image in the captured image when the distance between the specific part and the moving body is a set distance, a current value which is the size of the specific image in the captured image currently captured, and characteristics of the imaging device.

2. The work system of claim 1, wherein the relative position acquisition unit includes a relationship memory unit that pre-stores the relationship between the distance between the moving body and the specific part and the size of the specific image, and acquires the distance between the moving body and the specific part, which represents the relative position between the moving body and the specific part, based on the relationship stored by the relationship memory unit and the size of the specific image in the captured image.

3. A work system in which a mobile object moves within a predetermined area to perform work, a relative position acquisition unit that acquires a relative position between the moving body and a specific portion corresponding to the specific image in the region based on at least one of a size of the specific image included in an image captured by an imaging device provided on the moving body and a position of a reference line of the specific image, A work system in which the relative position acquisition unit acquires that the amount of widthwise deviation of the specific part of the imaging device of the moving body from the part corresponding to the reference line of the specific image is greater than a set value when the amount of deviation of the reference line of the specific image, which represents the position of the reference line of the specific image, from the center line of the captured image is greater than a set value than when the deviation is equal to or less than the set value.

4. A work system as described in any one of claims 1 to 3, comprising a mobile body position acquisition unit that acquires a mobile body position, which is the position of the mobile body within the area, based on the relative position between the mobile body and the identification unit acquired by the relative position acquisition unit.

5. The moving object position acquisition unit, an image-based provisional position acquisition unit that provisionally acquires a position of the moving object based on the relative position between the moving object and a specifying unit acquired by the relative position acquisition unit and the position of the specifying unit; a running state-based provisional position acquisition unit that provisionally acquires a position of the moving object based on a running state of the moving object; The work system according to claim 4, further comprising: a driving condition-based provisional position acquisition unit that acquires a final position of the moving body based on the position of the moving body provisionally acquired by the driving condition-based provisional position acquisition unit and the position of the moving body provisionally acquired by the image-based provisional position acquisition unit.

6. A work system in which a mobile object moves within a predetermined area to perform work, a relative position acquisition unit that acquires a relative position between the moving body and a specific portion corresponding to the specific image in the region based on at least one of a size of the specific image included in an image captured by an imaging device provided on the moving body and a position of a reference line of the specific image, The work system includes a collision possibility acquisition unit that acquires the possibility of a collision between the moving body and the specific part based on a change in size of the specific image acquired by the relative position acquisition unit.

7. A work system in which a mobile object moves within a predetermined area to perform work, a relative position acquisition unit that acquires a relative position between the moving body and a specific portion corresponding to the specific image in the region based on at least one of a size of the specific image included in an image captured by an imaging device provided on the moving body and a position of a reference line of the specific image, The moving body includes a running state acquisition device that acquires a running state of the moving body, The work system includes an abnormality detection unit that detects that the driving condition acquisition device is abnormal if the approach / departure speed of the moving body relative to the specific part and the speed of change in size of the specific image acquired by the driving condition acquisition device fall outside a range determined by a predetermined relationship.

Images