JP2022179331A - Unmanned forklift - Google Patents

Abstract

To detect a deviation amount between a position and an attitude of an object pallet with high accuracy by suppressing the reduction in working speed and the increase in cost without any limitation on the kind of a pallet to be handled.SOLUTION: An image acquisition unit acquires a photographed image from an imaging device S1. A pallet type identification unit has a learning model about a combination of images of plural types of pallets and the types of the pallets and identifies a type of an object pallet P1 by inputting the photographed image of the object pallet P1 acquired by the image acquisition unit in the learning model. A pallet position and shape acquisition unit acquires position and shape data of the object pallet P1 from a distance measurement device S2 for measuring a distance to the object pallet P1. A pallet deviation amount detection unit stores position and shape data of the pallet in advance and compares the stored position and shape data corresponding to the type of the identified object pallet P1 and position and shape data of the object pallet P1.SELECTED DRAWING: Figure 3

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an unmanned forklift that is self-propelled and performs cargo handling work.

In an unmanned forklift, there is a technique for determining whether or not a fork can be inserted into an insertion port (hole 20 in Patent Document 1) of a pallet (see, for example, the first embodiment of Patent Document 1). In this technology, the unmanned forklift 11 is equipped with a laser range finder 18 that can move up and down together with the forks 14 . The laser range finder 18 irradiates and scans the opening end face 19a of the pallet 19 placed at the pickup position with a laser beam to obtain image data of the edge line of the hole 20. FIG. Then, the pattern controller 21 compares the obtained image data of the line of the hole 20 with the image data of the line of the hole 20 captured in advance under a plurality of conditions in which the distance and inclination of the end surface of the pallet 19 with respect to the unmanned forklift 11 are changed. Thus, it is determined whether or not the fork 14 can be inserted into the hole 20 of the pallet 19 .

In addition, in an unmanned forklift, there is a technique for identifying the type of pallet and adjusting the fork interval according to the interval between the identified pallet insertion openings (holes 20 in Patent Document 1) (for example, Patent Document 1, No. 2 embodiment). In this technique, a plurality of types of pallets 19 with different intervals of holes 20 are used, and the unmanned forklift 11 is provided with a fork shifter capable of adjusting the interval between the pair of forks 14 . The pattern controller 21 specifies the width of the pallet 19 from the image data of the edge line of the hole 20 obtained by the laser range finder 18, thereby identifying the type of the pallet 19. FIG. Then, the controller 17 controls the fork shifters to adjust the spacing between the pair of forks 14 in accordance with the spacing of the holes 20 corresponding to the identified width of the pallet 19 .

Furthermore, there is a technique for identifying the type of pallet in an unmanned forklift (see Patent Document 2, for example). In this technique, a pattern 01 is provided on a pallet 05 and an unmanned forklift 5 is equipped with a pair of cameras 3 and 4. Cameras 3 and 4 photograph pattern 01 on pallet 05 . Then, the position determination means determines the position of the pallet 05 from the photographed pattern, and the collation means collates the photographed pattern with the reference pattern to identify the type of the pallet 05 .

Japanese Patent Application Laid-Open No. 2017-19596 Japanese Patent Application Laid-Open No. 9-218014

Identification of the type of pallet in the unmanned forklift of Patent Document 1 is performed by detecting an image of the edge line of the hole in the open end face of the pallet by distance measurement using a laser. So, for example, if there is a load or another pallet right next to the pallet, the
1647479821055_0
block the holes of the pallet, or foreign matter enters the holes of the pallet, there is a risk that the type of the pallet cannot be identified or that the type of the pallet is erroneously identified.

Identification of the type of pallet in the unmanned forklift of Patent Document 2 requires affixing a pattern to all pallets to be handled by the unmanned forklift, which increases costs. In addition, when many types of pallets with different shapes are targeted, a common place for pasting patterns is required for all of the pallets, which limits the types of pallets that can be handled.

On the other hand, in the case of detecting the amount of positional deviation of a plurality of types of pallets only with a distance detector such as a laser scanner, the distance detector can be used for all types of stored position/shape data of the plurality of types of pallets. It is necessary to identify the type of the corresponding pallet by comparing and collating the obtained measured shape data.

Therefore, due to the characteristics of the distance detector that measures only distance data, when the number of types of target pallets increases, items other than pallets placed on the floor (for example, cardboard boxes) may be incorrectly identified as a specific type of pallet. Increased chances of recognition.

In addition, the computational load of comparing and collating with shape data of a plurality of types of pallets is large, so the computation takes time and the work speed decreases, and the cost increases if an expensive computer with high processing power is used.

The present invention does not limit the types of pallets that can be handled, suppresses a decrease in work speed and an increase in costs, and detects with high accuracy the amount of deviation in the position and posture of a target pallet from its normal position, thereby performing the required cargo handling work. To provide an unmanned forklift capable of reliably performing

The unmanned forklift according to the present invention, in order to solve the above problems,
An unmanned forklift equipped with a pair of forks that automatically performs cargo handling work using pallets,
an image acquisition unit that acquires a photographed image from a photographing device that photographs the front of the unmanned forklift;
For a plurality of types of the palette, having a learned model that has undergone machine learning on combinations of the images of the palette and the types of the palette, and the captured image of the target palette acquired by the image acquisition unit. a pallet type identification unit that identifies the type of the target pallet by inputting it into the learning model;
a pallet position/shape acquisition unit that acquires position/shape data of the target pallet from a distance measuring device that measures the distance from the target pallet;
Position/shape data of each of the plurality of types of pallets is stored in advance, and the stored position/shape data corresponding to the type of the target pallet identified by the pallet type identification unit and the pallet position. a pallet deviation amount detection unit that detects the amount of deviation of the position and orientation of the target pallet from the normal position by comparing the position and shape data of the target pallet acquired by the shape acquisition unit;
(A) work including an operation of inserting the fork into the insertion opening of the target pallet using the detected deviation amount;
(B) a stacking operation of stacking the pallet and the load held by inserting the fork into the insertion slot on top of the load of the target pallet according to the deviation amount of the target pallet;
A travel control unit that controls travel for performing at least one of the work of
Prepare.

According to such a configuration, even if objects to be transported by the unmanned forklift are a plurality of types of pallets having different shapes, the types of pallets that can be handled are not limited. By inputting the photographed image into a learning model that has already undergone machine learning on combinations of types, the palette type identification unit identifies the type of the target palette with high accuracy. Then, the pallet shift amount detection unit stores stored position/shape data corresponding to the type of the target pallet identified by the pallet type identification unit, and the position/shape data of the target pallet obtained by the pallet position/shape acquisition unit. Shape data can be compared and collated.

As a result, the deviation amount of the position and attitude of the target pallet from the normal position of the pallet can be detected with high accuracy, and an article other than the pallet placed on the floor (for example, a cardboard box, etc.) is erroneously recognized as a specific type of pallet. It is possible to prevent accidents such as interference between the fork and the pallet or goods. In addition, since calculations for comparison and collation of pallet shape data are shortened and work speed is improved, there is no need to use an expensive computer with high processing power.

After detecting the amount of deviation of the target pallet with high accuracy, the unmanned forklift performs, for example, the work (A) of picking up the pallet and cargo on the floor and transporting it to another place, It is possible to perform required cargo handling work such as the work of picking up the second tier pallet on the cargo and the cargo and transporting it to another place, or the stacking work of the work (B).

Here, the position/shape data is height data including the insertion port of the pallet,
The pallet shift amount detection unit detects line segments based on the stored position/shape data corresponding to the type of the target pallet and the position/shape data of the target pallet acquired by the pallet position/shape acquisition unit. It is a preferred embodiment to superimpose and compare the line segments based on.

According to such a configuration, since the line segments are superimposed and compared to grasp the deviation amount, the calculation cost can be suppressed. Moreover, since the height data can be obtained only by scanning in the horizontal direction, there is no need to move the distance measuring device for measuring the distance to the target pallet up and down. It can shorten the detection time.

In a preferred embodiment, the apparatus further comprises a shift amount error determination section that performs an error determination when the shift amount detected by the pallet shift amount detection section exceeds a predetermined threshold value.

According to such a configuration, when the forks cannot be inserted into the target pallet due to the large amount of deviation, an error determination is made before the fork insertion operation, thereby preventing an accident in which the target pallet and the forks interfere with each other. For example, the threshold value of the amount of deviation, which is the standard for the error determination, is set when the distance measuring device measures the distance to the target pallet and the pallet position/shape acquisition unit acquires the position/shape data of the target pallet. The maximum value of the range in which the unmanned forklift can move by itself and insert the forks into the target pallet from the predetermined stop position where the is stopped.

Further, when the shape data of the target pallet acquired by the pallet position/shape acquisition unit is compared with the shape data of a pallet of the same type as the target pallet stored in advance, and the difference therebetween is greater than a predetermined threshold. In a preferred embodiment, further includes a pallet shape determination unit that determines that the acquired shape data of the target pallet has an error.

According to such a configuration, for example, when another article (for example, a cardboard box) is placed right next to the target pallet, even if the size of the target pallet is detected incorrectly due to the influence of the article, An error can be determined by comparing with the stored pallet shape corresponding to the type of the target pallet identified by the pallet type identification unit. As a result, it is possible to prevent the fork from interfering with the insertion port of the target pallet and stopping the work.

Furthermore, by performing image processing on the photographed image of the target pallet acquired by the image acquiring unit, the number of pallets within a predetermined range that is set to include the target pallet is calculated to determine whether or not there is stacking. It is a preferred embodiment to further include a pallet stacking determination unit that determines the .

According to such a configuration, by calculating only the number of pallets within a specific range in front of the unmanned forklift, it is possible to reliably determine whether or not pallets are stacked without erroneously detecting pallets other than the target pallet. . For example, if a non-target pallet that is placed on the side of the target pallet on the floor is detected and the number of pallets is calculated as 2, there is no false detection. Misjudgment to do is eliminated.

Further, it is a preferred embodiment that the pallet stacking determination unit performs machine learning in advance using respective pallet images as training data for pallets in a plurality of upper and lower tiers.

According to such a configuration, when the surrounding images of the pallet differ depending on which stage of the pallet it is, the different surrounding images are learned as teacher data, so that the pallet detection accuracy can be improved.

Furthermore, the preferred embodiment further comprises a fork width changing device that changes the interval between the forks according to the interval between the insertion openings of the target pallet according to the type of the target pallet obtained from the pallet type identification unit. be.

According to such a configuration, it is not necessary to prepare an unmanned forklift for each target pallet having different insertion opening intervals, and the fork width changing device can be operated so that the fork intervals correspond to the insertion opening intervals. A single unmanned forklift can transport multiple types of pallets.

Furthermore, a distance sensor arranged at the tip of the fork for detecting the presence or absence of an object in front;
a load sensor that detects whether the fork holds the pallet;
further comprising
Using the detection result of the presence or absence of the object by the distance sensor and the detection result of whether or not the pallet is held by the load sensor,
A stacking operation in which the second pallet and cargo are loaded on top of the first pallet cargo, and/or
In a preferred embodiment, a tier destacking operation is performed to pick up the second tier pallet and cargo on top of the first pallet cargo and transport them to another location.

According to this configuration, when the stacking work is performed, the distance sensor can detect the height of the cargo on the first pallet, and the load sensor pulls out the forks after the stacking work. Height can be detected. Further, when performing the stacking work, the distance sensor can detect the height of the insertion opening of the second pallet, and the load sensor can detect the pallet and cargo to be held in the stacking work by the fork. can detect the state held by Therefore, the distance sensor and the load sensor enable reliable stacking and/or destacking operations.

Further, the photographing device photographs the target pallet at the first stop position, the pallet type identification unit identifies the type of the target pallet, and the second stop position is closer to the target pallet than the first stop position. In a preferred embodiment, the distance measuring device measures the distance to the target pallet, and the pallet position/shape acquiring unit acquires the position/shape data of the target pallet.

According to such a configuration, the distance measuring device measures the distance from the target pallet at the second stop position closer to the target pallet than the first stop position, and the pallet position/shape acquisition unit detects the position and shape of the target pallet. Since the data is acquired, the deviation amount of the position and orientation of the target pallet can be detected more accurately by the pallet deviation amount detection unit.

As described above, the unmanned forklift of the present invention has no restrictions on the types of pallets that can be handled, can suppress a decrease in work speed and an increase in costs, and can accurately measure the amount of deviation of the position and posture of the target pallet from the normal position of the pallet. It is possible to reliably perform the required cargo handling work by detecting it.

1 is a perspective view of an unmanned forklift according to an embodiment of the present invention; FIG. Fig. 2 is a left side view of the unmanned forklift holding a pallet and a load, showing a cross section of the pallet with the forks inserted into the receptacles; 1 is a perspective view showing a pallet and cargo on the floor, and the unmanned forklift; FIG. It is a block diagram which shows the outline|summary of apparatus configuration. FIG. 4 is an enlarged cross-sectional view of a main part showing a sensor arranged at the tip of the right fork; FIG. 4 is an enlarged vertical cross-sectional view of a main part showing a sensor arranged at the tip of the right fork; FIG. 10 is an enlarged view of the essential part seen from the left side showing the surroundings of the load sensor; FIG. 4 is an enlarged cross-sectional view of a main part showing the surroundings of the load sensor; FIG. 10 is a diagram showing an example of an image in which the palette type identification unit has identified the type of the target palette; FIG. 4 is a schematic plan view showing line segments based on point cloud data of stored pallets corresponding to the type of target pallet; FIG. 8B is a schematic plan view showing a state in which the line segment in FIG. 8A and a line segment based on the point cloud data of the target pallet acquired by the pallet position/shape acquisition unit are superimposed. It is a front view of a fork width change device. It is a schematic plan view for operation|movement description which narrows a fork width with a fork width change apparatus. It is a schematic plan view for operation|movement description which widens a fork width with a fork width change apparatus. FIG. 4 is an explanatory diagram of an operation example of stacking work, showing a state in which an unmanned forklift approaching a target pallet stops a predetermined distance short of the target pallet; FIG. 10 is an explanatory diagram of an operation example of stacking work, showing a state in which the forks of the unmanned forklift are raised, the height of the cargo on the target pallet is detected by the distance sensor, and then the forks are raised a predetermined distance and stopped. there is FIG. 11 is an explanatory diagram of an operation example of stacking work, showing a state in which the unmanned forklift is advanced by a predetermined distance to position the pallet and the load to be stacked above the target pallet and the load. FIG. 10 is an explanatory diagram of an operation example of stacking work, and shows a state at the moment when the lower surface of the pallet to be stacked by lowering the fork comes into contact with the upper surface of the cargo on the target pallet. FIG. 11 is an explanatory diagram of an operation example of stacking work, showing a state in which the forks are further lowered and the load sensor detects that there is no load on the forks.

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment according to the present invention will be described below with reference to the drawings.

In the following embodiments, the direction from the base side to the tip side of the fork of the unmanned forklift is defined as the front side, and the opposite direction is defined as the rear side, and left and right are defined toward the front.

For unmanned forklifts, the front view is the front view where the forks are located. For the target pallet and cargo, the front view is the view of the target pallet facing the side where the forks are inserted.

<Unmanned forklift>
An unmanned forklift 1 according to an embodiment of the present invention shown in the perspective view of FIG. 1, the left view of FIG. 2, the perspective view of FIG. 3, and the block diagram of FIG. The unmanned forklift truck 1 has an estimation function and performs cargo handling work by self-propelled according to a destination instruction from a management device that wirelessly communicates with the unmanned forklift truck 1 . The unmanned forklift 1 includes a cargo handling device A that performs cargo handling work, a moving device B that travels and turns, a control device C that controls the cargo handling device A and the moving device B, a fork width changing device D, a processing device 10, and the like. Prepare.

(cargo handling equipment)
The cargo handling apparatus A has a mast M that moves up and down and tilts in the front-rear direction, a fork 2 that carries a pallet P and a cargo W, and a lift bracket 3 that supports the fork 2 and moves up and down along the mast M. The fork 2 consists of a right fork 2R and a left fork 2L.

The forks 2 and lift brackets 3 are moved up and down along the inner mast 4 by lift chains 7 . The mast M consists of an inner mast 4 that supports the lift bracket 3 and moves up and down, and an outer mast 5 that guides the inner mast 4 so that it can move up and down. The outer mast 5 is tilted forward and backward by a tilt cylinder 6 .

(moving device)
The moving device B has a pair of left and right front wheels, rear wheels that are driving wheels and steering wheels, and a driving device for the rear wheels.

(Control device)
The control device C controls the lifting drive device of the cargo handling device A, the drive device of the moving device B, and the drive device of the tilt cylinder 6, and has a communication device that communicates with a management device on the ground side. Further, the control device C has an interface with the processing device 10 shown in FIG. 4 and an interface with the distance sensor S3 and the load sensor S4.

(fork width changing device)
The fork width changing device D changes the space F between the right fork 2R and the left fork 2L.

(Processing device)
The processing device 10 shown in FIG. 4 has an interface with the photographing device S1 and issues a photographing command to the photographing device S1. The processing device 10 also has an interface with the distance measuring device S2, and issues a distance measurement command to the distance measuring device S2.

<Palette>
The pallets P and P1 shown in the left view of FIG. 2 and the perspective view of FIG. 3 are flat pallets, for example, and have insertion openings Q and Q for left and right forks 2L and 2R. A cargo W is placed on the pallets P and P1.

<Sensor group>
As shown in the perspective view of FIG. 1 and the block diagram showing the outline of the equipment configuration in FIG. A recognition sensor S5 and an obstacle detection sensor S6 are provided. The sensor groups S1 to S6 are arranged at positions to be described later. The control device C and the processing device 10 shown in FIG. 4 are arranged inside the main body 1A of the unmanned forklift 1 as shown in FIG.

(shooting device, distance measuring device)
As shown in FIG. 1, the photographing device S1 and the distance measuring device S2 are held by a bracket J attached to a support plate I projecting forward from the center of the front lower portion of the main body 1A of the unmanned forklift 1 in the left-right direction. That is, the photographing device S1 and the distance measuring device S2 are installed on members that are immovable with respect to the main body 1A, rather than members that are movable with respect to the main body 1A, such as the lift bracket 3 and the fork 2. FIG. The structure for holding the photographing device S1 and the distance measuring device S2 is also applied to structures shown in other drawings such as FIG.

The photographing device S1 is, for example, a monocular camera, and photographs the front of the unmanned forklift 1 at a predetermined stop position where the unmanned forklift 1 is stopped. The distance measuring device S2 is, for example, a 2D-LiDAR (Light Detection And Ranging) or TOF (Time Of Flight) camera, and detects the target pallet P1 (for example, FIG. 2) on the floor N at a predetermined stop position for stopping the unmanned forklift 1. Make a distance measurement. If the distance measuring device S2 is a 2D-LiDAR, it irradiates a laser beam in a pulsed manner while changing its direction in the horizontal direction, detects the scattered light that is reflected and returned, and detects the scattered light that is reflected back by the object. Measures the distance, direction, etc. from time to an object.

As shown in FIG. 4, the image captured by the photographing device S1 is sent to the processing device 10, and the data measured by the distance measuring device S2 is sent to the processing device 10. FIG.

(distance sensor)
As shown in the perspective view of FIG. 1, the enlarged transverse cross-sectional view of the essential part of FIG. 5A, and the enlarged vertical cross-sectional view of the essential part of FIG. Located at the tip. The distance sensor S3 may be arranged at the tip of the left fork 2L. The distance sensor S3 is, for example, a distance setting type photoelectric sensor, and detects the presence or absence of an object in front.

As shown in FIG. 4, a signal indicating the presence or absence of an object detected by the distance sensor S3 (for example, an ON/OFF signal) is sent to the controller C. As shown in FIG.

(Obstacle detection sensor)
As shown in the perspective view of FIG. 1 and the enlarged cross-sectional view of the essential part of FIG. The obstacle detection sensor S6 is, for example, a photoelectric sensor, and detects obstacles approaching the tip of the fork 2 .

A signal indicating the presence or absence of an obstacle (for example, an ON/OFF signal) detected by the obstacle detection sensor S6 is sent to the controller C. FIG. When the obstacle detection sensor S6 detects an obstacle, the controller C brings the unmanned forklift 1 to an emergency stop, for example.

(Zone sensor)
As shown in the view from the left in FIG. 2, the enlarged view of the essential part from the left in FIG. 6A, and the enlarged cross-sectional view of the essential part in FIG. It is arranged at the roots of the left and right forks 2L and 2R in the folded state. The load sensor S4 is, for example, a proximity switch, and detects whether the fork 2 holds the pallet P or not.

As shown in FIG. 4, a signal (for example, an ON/OFF signal) indicating whether or not the fork 2 is holding the pallet P detected by the load sensor S4 is sent to the controller C. As shown in FIG.

The structure around the load sensor S4 at the base of the left fork 2L will be described with reference to FIGS. 6A and 6B. A seesaw plate 22 supported by a horizontal support shaft R is stopped by a stop plate 23 and elastically biased by an extension coil spring 24 . When the fork 2 does not hold the pallet P, the seesaw plate 22 rests at the solid line position in FIG. 6A. When the fork 2 holds the pallet P as shown in FIG. 2, the seesaw plate 22 swings so that the operation piece 22A descends and the detection piece 22B rises as indicated by the phantom lines in FIG. 6A. As a result, since the presence sensor S4 detects the detection piece 22B, the state in which the fork 2 holds the pallet P can be detected.

(Self-position recognition sensor)
As shown in the perspective view of FIG. 1, the self-position recognition sensor S5 is arranged facing upward on the upper surface of the frame K that protrudes upward from the main body 1A. That is, the self-position recognition sensor S5 is installed on a non-movable member with respect to the main body 1A. The self-position recognition sensor S5 performs, for example, 3D imaging in all directions in the horizontal direction and in a vertical field of view of about 30°. The self-position recognition sensor S5 is, for example, a 3D-LiDAR, and is used for a laser SLAM (Simultaneously Localization And Mapping) self-position estimation method. The self-position estimation of the unmanned forklift 1 may be an image SLAM-type self-position estimation method or the like.

In the unmanned forklift 1 controlled using the above sensor groups S1 to S6, the signal cables and power supply cables of the sensors S3, S4, and S6 installed on members movable with respect to the main body 1A are the cable carrier shown in FIG. 8 to the controller C and the power supply in the main body 1A.

<Image acquisition unit>
As shown in FIG. 4, the processing device 10 has an image acquisition section 11 . The image acquisition unit 11 acquires a captured image from the imaging device S1. The photographing device S1 photographs, for example, the target palette P1 shown in FIG. 2 or 3 . The image acquisition unit 11 acquires the captured image of the target palette P1 from the imaging device S1.

<Pallet type identification unit>
As shown in FIG. 4, the processing device 10 has a pallet type identification section 12 . The palette type identification unit 12 has a learning model that has been trained by performing machine learning on combinations of the images of each palette P and the type of the palette P for a plurality of types of palettes P. FIG. The palette type identification unit 12 identifies the type of the target palette P1 by inputting the captured image of the target palette P1 acquired by the image acquisition unit 11 into the learning model.

FIG. 7 shows an example of an image in which the palette type identification unit 12 has identified the type of the target palette. In this example, the pallet type identification unit 12 identifies the type "small" of the target pallet P1 on the floor N and the type "small" of the target pallet P2 on the load of the target pallet P1. The pallet type identification unit 12 can identify types of pallets such as "small", "medium", and "large", for example, as well as (0.XX) and (0.ΔΔ) in FIG. The certainty (probability) of the identified pallet type can be displayed on the image.

<Pallet position/shape acquisition unit>
As shown in FIG. 4 , the processing device 10 has a pallet position/shape acquisition unit 13 . The pallet position/shape acquisition unit 13, for example, the unmanned forklift 1 shown in FIG. 2 or FIG. Acquire the position/shape data of the pallet P1. The position/shape data of the target pallet P1 is, for example, point cloud data of the height including the insertion opening Q of the target pallet P1.

<Pallet shift amount detector>
As shown in FIG. 4, the processing device 10 has a pallet shift amount detection unit 14 . The pallet shift amount detection unit 14 stores in advance the position/shape data of each pallet P for a plurality of types of pallets P. FIG. The position/shape data of each pallet P stored in advance by the pallet shift amount detection unit 14 is obtained by measuring the distance to each pallet P by the distance measuring device S2 at the predetermined stop position of the unmanned forklift 1, and determining the pallet position. - The position/shape data of each pallet P acquired by the shape acquisition unit 13 . The position/shape data of each pallet P is point cloud data of the height including the insertion opening Q of the pallet P, for example.

The pallet shift amount detection unit 14 detects the stored position/shape data corresponding to the type of the target pallet P1 identified by the pallet type identification unit 12, that is, the position/shape data in each pallet P stored in advance as described above. By comparing the position/shape data of the pallet P corresponding to the type of the target pallet P1 and the position/shape data of the target pallet P1 acquired by the pallet position/shape acquisition unit 13, the position/shape data of the target pallet P1 corresponding to the normal position is determined. Detect the amount of deviation between position and orientation.

L0 in the schematic plan view of FIG. 8A is, for example, a pallet on the floor N, based on the point cloud data of the height including the stored pallet P insertion opening Q corresponding to the target pallet P1 in FIG. 2 or FIG. is a line segment. As described above, the pallet displacement amount detection unit 14 stores the data of the line segment L0 as shown in FIG. 8A for each of a plurality of types of pallets P.

As shown in the schematic plan view of FIG. 8B, the pallet shift amount detection unit 14 detects the line segment L0 and the height point cloud data including the insertion opening Q of the target pallet P1 acquired by the pallet position/shape acquisition unit 13. and the line segment L1 based on are overlapped and compared.

According to such a pallet shift amount detection unit 14, since the line segments L0 and L1 are superimposed and compared to grasp the shift amount, the calculation cost can be suppressed. In addition, since the height point cloud data can be obtained only by scanning in the horizontal direction, there is no need to vertically move the distance measuring device S2 for measuring the distance to the target pallet P1. It is possible to shorten the time required to detect the amount of posture deviation.

<Photographing position of photographing device, distance measurement position of distance measuring device>
The position where the unmanned forklift 1 stops when the photographing device S1 photographs the target pallet P1 and the pallet type identification unit 12 identifies the type of the target pallet P1 is defined as a first stop position, and the distance measuring device S2 determines the target pallet P1. is measured and the position/shape data of the target pallet P1 is acquired by the pallet position/shape acquisition unit 13, the position at which the unmanned forklift 1 stops is defined as a second stop position.

For example, the position where the unmanned forklift truck 1 shown in FIGS. Alternatively, the second stop position may be closer to the target pallet P1 than the first stop position. By setting the second stop position closer to the target pallet P1 than the first stop position, the pallet deviation detection unit 14 detects the deviation amount of the position and posture of the target pallet P1 with higher accuracy. can be done.

<Error judgment part>
As shown in FIG. 4, the processing device 10 has a deviation amount error determination unit 15 . The deviation amount error determination unit 15 determines an error when the amount of deviation of the position and orientation of the target pallet P1 from the normal position detected by the pallet deviation amount detection unit 14 exceeds a predetermined threshold value.

According to such a deviation amount error determination unit 15, when the fork 2 cannot be inserted into the target pallet P1 because the deviation amount is large, error determination is performed before the fork insertion operation, so that the target pallet P1 and the fork 2 Interference accidents can be prevented. For example, the threshold value of the amount of deviation, which is the reference for the error determination, is obtained by measuring the distance to the target pallet P1 by the distance measuring device S2 and acquiring the position/shape data of the target pallet P1 by the pallet position/shape acquisition unit 13. The maximum value of the range in which the unmanned forklift 1 can travel by itself from a predetermined stop position where the unmanned forklift 1 is stopped and the forks 2 can be inserted into the target pallet P1.

<Pallet shape determination unit>
As shown in FIG. 4 , the processing device 10 has a pallet shape determining section 16 . The pallet shape determination unit 16 compares the shape data of the target pallet P1 acquired by the pallet position/shape acquisition unit 13 with the shape data of the same type of pallet as the target pallet P1 stored in advance, and determines that the difference between them is a predetermined value. is larger than the threshold value, it is determined that there is an error in the acquired shape data of the target pallet P1.

According to such a pallet shape determination unit 16, for example, when another article (for example, a cardboard box) is placed right next to the target pallet P1, the size of the target pallet P1 is mistaken due to the influence of the article. Even if the error is detected by the pallet type identification unit 12, the error can be determined by comparing with the stored pallet shape corresponding to the type of the target pallet P1 identified by the pallet type identification unit 12. FIG. As a result, it is possible to prevent an accident in which the fork 2 interferes with the insertion port Q of the target pallet P1 and the work is stopped.

<Pallet stacking discrimination part>
As shown in FIG. 4 , the processing device 10 has a pallet stacking discrimination section 17 . The pallet stacking determination unit 17 performs image processing as shown in FIG. 7 on the photographed images of the target pallets (for example, P1 and P2 in FIG. 7) acquired by the image acquisition unit 11, so that if there are the target pallets P1 and P2, The presence or absence of stacking is determined by calculating the number of pallets within the set specific range.

For example, if the number of pallets within the specific range is two as shown in FIG. 7, stacking the second pallet and cargo transported from another place on top of the first pallet and cargo. It can be determined that the work is impossible, and if the number of pallets within the specific range is one, it can be determined that the stacking work is possible. Further, when the number of pallets within the specific range is one, it is possible to determine that the stacking work of picking up the second pallet and cargo and transporting them to another location is not possible, and the number of pallets within the specific range is not possible. If the number is 2, it can be determined that the above-mentioned destacking work is possible. Further, if it is set in advance that stacking work is not possible when the number of pallets within the specific range is two or more, the stacking work is performed when the number of pallets within the specific range is two or more. You can decide that it is not possible.

According to such a pallet stacking determination unit 17, by calculating only the number of pallets within a specific range in front of the unmanned forklift 1, pallets other than the target pallet are not erroneously detected. The presence or absence can be determined with certainty. For example, since an erroneous detection that the number of pallets is calculated to be two by detecting a non-target pallet placed on the side of the target pallet on the floor N does not occur, stacking is possible even though it can be stacked. Misjudgment that it is not possible is eliminated.

For example, in the example of the images in FIG. 7, the surrounding image of the pallet P1 on the floor and the surrounding image of the upper pallet P2 are different. Therefore, for example, when using a trained learning model in which a large number of images including pallets on the floor are prepared and machine learning is performed using the pallets on the floor as teacher data, the upper pallet P2 may not be identified.

Therefore, in a preferred embodiment, the pallet stacking determination unit 17 is machine-learned in advance using each pallet image as training data for a plurality of pallets in the upper and lower tiers. According to the pallet stacking determination unit 17, when the peripheral images of the pallets differ depending on which tier the pallet is in, each different peripheral image is learned as training data, so the pallet detection accuracy is improved. can.

<Running control part>
As shown in FIG. 4 , the control device C has a travel control section 18 . The travel control unit 18 controls travel of the unmanned forklift 1 in order to perform at least one of the following works (A) and (B). The travel control unit 18 may be provided in the management device on the ground side.

(A) Work including an operation of inserting the forks 2 into the insertion openings of the target pallet using the displacement amount detected by the pallet displacement amount detection unit 14 .
(B) A stacking operation in which the pallet P and the cargo W held by inserting the forks 2 into the insertion openings Q are loaded onto the cargo of the target pallet according to the deviation amount of the target pallet.

For example, as shown in FIG. 3, the unmanned forklift 1 picks up the target pallet P1 and the load W on the floor N and transports them to another location. Further, the unmanned forklift 1 stacks the pallet P brought from another place on the load W of the target pallet P1 on the floor N as shown in FIG. Further, the unmanned forklift 1 picks up the target pallet P2 and the load W on the target pallet P1 on the floor N as shown in FIG. 7 and transports them to another location.

<Example of structure of fork width changing device>
As shown in the front view of FIG. 9, the right fork 2R and the left fork 2L are guided by linear guides 25 and 26 so as to be movable in the left-right direction. The actuators of the fork width changing device D that changes the space F between the right fork 2R and the left fork 2L are electric cylinders 27 and 28 with position sensors, for example. That is, by driving the electric cylinder 27 to move the piston 27A forward and backward, the right fork 2R moves in the horizontal direction, and by driving the electric cylinder 28 to move the piston 28A forward and backward, the left fork 2L moves in the horizontal direction. do. The electric cylinders 27 and 28 with position sensors may be hydraulic cylinders with position sensors or the like.

As shown in the schematic plan view of FIG. 10, when the interval F between the forks is larger than the width G between the insertion openings Q and Q of the target pallet P1, the interval F between the forks is narrowed as indicated by the arrow at the tip of the fork. to operate the fork width changing device D. As shown in the schematic plan view of FIG. 11, when the fork interval F is smaller than the width G between the insertion openings Q and Q of the target pallet P1, the fork interval F is widened as indicated by the arrow at the tip of the fork. to operate the fork width changing device D.

Based on the type of the target pallet P1 obtained from the pallet type identification unit 12, the fork width changing device D adjusts the spacing F between the left and right forks 2L and 2R in accordance with the spacing G between the insertion openings Q and Q of the target pallet P1. can be changed. Therefore, it is not necessary to prepare an unmanned forklift truck 1 for each target pallet having a different interval G between the insertion ports Q, Q. A plurality of types of pallets can be transported by one unmanned forklift 1 by operating the width changing device D.

<Work using the detection results of the distance sensor and the load sensor>
The unmanned forklift 1 uses the detection result of the distance sensor S3 that detects the presence or absence of an object in front and the detection result of the load sensor S4 that detects whether the forks 2 are holding the pallet P, for example, as follows. work like this.

That is, as shown in FIG. 2, the unmanned forklift 1 performs a stacking operation of loading the second pallet P and the load W on the load W of the first target pallet P1. The unmanned forklift truck 1 also picks up the second pallet P2 and the cargo on the first pallet P1 shown in FIG.

When performing the stacking work, the distance sensor S3 can detect the height of the load W on the first pallet P1, and the load sensor S4 can detect the height at which the forks 2 are pulled out after the stacking work. Further, when performing the stacking work, the distance sensor S3 can detect the height of the insertion opening of the pallet P2 on the second stack, and the load sensor S4 can detect the pallet P2 and the load to be held in the stacking work. The state held by the fork 2 can be detected. Therefore, the distance sensor S3 and the load sensor S4 enable reliable stacking and/or destacking operations.

(Operation example of stacking work)
The unmanned forklift 1 is located at a predetermined distance E from the target pallet P1 shown in FIG. The photographed image of the target palette P1 acquired by the image acquisition unit 11 is input to the learning model of the palette type identification unit 12, and the palette type identification unit 12 identifies the type of the target palette P1.

The distance measuring device S2 (FIG. 1) measures the distance to the target pallet P1, and the pallet position/shape acquisition unit 13 acquires the position/shape data of the target pallet P1. The pallet displacement amount detection unit 14 stores stored position/shape data corresponding to the type of the target pallet P1 identified by the pallet type identification unit 12, and the position/shape data of the target pallet P1 acquired by the pallet position/shape acquisition unit 13. By comparing the shape data, the amount of deviation of the position and orientation of the target pallet P1 from the normal position is detected.

Further, the pallet stacking determination unit 17 performs image processing on the photographed image of the target pallet P1 acquired by the image acquisition unit 11, thereby calculating the number of pallets within a predetermined specific range where the target pallet P1 exists. This determines whether or not there is stacking. That is, since the number of pallets within the specific range is one as shown in FIG. 2, the pallet stacking determination unit 17 determines that stacking is possible.

The unmanned forklift 1 is advanced by lowering the height of the fork 2 to approach the target pallet P1, and the unmanned forklift 1 is stopped at a position where the target pallet P1 and the pallet P are separated from each other by a predetermined distance U, for example, as shown in FIG. 12A. At this position, as indicated by an arrow V in FIG. 12A, the distance sensor S3 (FIG. 1) that detects the presence or absence of an object detects the target pallet P1.

Next, the fork 2 is raised as indicated by the arrow T1 in FIG. 12A, and the position where the distance sensor S3 at the tip of the fork 2 no longer detects the load W on the target pallet P1 is the load W on the target pallet P1 shown in FIG. 12B. is the height H of From the position where the distance sensor S3 no longer detects the load W, the fork 2 is raised by a predetermined distance as shown in FIG. 12B.

Next, the unmanned forklift truck 1 is advanced as indicated by the arrow T2 in FIG. 12B, and the position and posture of the target pallet P1 is shifted from the normal position detected by the pallet shift amount detection unit 14. The unmanned forklift 1 is stopped at the position shown in FIG.

Next, the fork 2 is lowered as indicated by the arrow T3 in FIG. 12C, and the pallet P and the load W are stacked on the upper surface of the load W on the target pallet P1 as shown in FIG. 12D. 12E, where the forks 2 are further lowered as indicated by the arrow T4 in FIG. 12D, the forks 2 do not hold the pallet P, that is, the presence sensor S4 detects the state where the pallet P is not held. Therefore, the fork 2 stops descending.

Next, the unmanned forklift 1 is retracted from the state of FIG. 12E as indicated by an arrow T5 in FIG.

(Operation example of stacking work)
For example, an example will be described in which the second pallet P2 and the cargo on the first pallet P1 shown in FIG. 7 are picked up and transported to another location.

When the second pallet P2 is stacked in accordance with the first pallet P1, the second pallet P2 can be picked up based on the position and attitude of the first pallet P1.

In that case, after the unmanned forklift 1 is stopped in front of the stacked pallets P1 and P2 and the cargo shown in FIG. 7, the forks 2 are raised. The lift of the fork 2 is stopped when the distance sensor S3 at the tip of the fork 2 detects the object in front and stops detecting the object in front. In this state, the left and right forks 2L and 2R are positioned in front of the insertion openings Q and Q of the second pallet P2. Let Since the fork 2 holds the load while the load sensor S4 detects the pallet P2, the lift of the fork 2 is stopped after the fork 2 is lifted by a predetermined distance, and the unmanned forklift 1 is moved backward.

When performing the above-described tier destacking work, for example, in the example of the image in FIG. It may be the same as the left-right position of the insertion openings Q, Q of the pallets P1, P2. In such a case, the distance sensor S3 at the tip of the fork 2 may fail to detect an object in front at the position of the gap between the left and right cardboard boxes. When the distance sensor S3 at the tip of the fork 2 no longer detects an object in front, the fork 2 stops rising and the left and right forks 2L and 2R are positioned in front of the insertion openings Q and Q of the second pallet P2. You will mistakenly judge that As a result, there arises a problem that it becomes impossible to perform the stacking work properly.

As one of the measures for solving such a problem, when the pallet type identification unit 12 identifies the types of the target pallets P1 and P2 at the first stop position, the lower surface of the pallet P2 on the second stage is roughly identified. It is conceivable to estimate the height based on the image data. For example, in an image in which the palette type identification unit 12 identifies the types of the target palettes P1 and P2 (for example, FIG. 7), the number of pixels in the width of the palette P1 on the first stage and the width of the palette P1 on the first stage and the number of pixels on the second stage A ratio to the number of pixels in the vertical interval of the eye palette P2 is obtained. Also, the actual dimensions of the width and height of the pallet P1 are obtained based on the prestored shape data of the same type of pallet as the target pallet P1. From the ratio and the actual dimensions, the approximate height of the lower surface of the second pallet P2 can be estimated.

Then, after recognizing that the insertion openings Q, Q of the second-tier pallet P2 are above the estimated approximate height of the lower surface of the second-tier pallet P2, the tier dismantling operation is performed. That is, after the unmanned forklift 1 is stopped in front of the stacked pallets P1 and P2 and the cargo, the forks 2 are lifted, and the lift of the forks 2 is stopped when the distance sensor S3 no longer detects an object in front. The position is set above the estimated approximate height of the lower surface of the second stage pallet P2. By operating in this manner, the above problem can be eliminated.

(Another operation example of the step dismantling work)
A 2D-LiDAR, which is the distance measuring device S2, is installed on the lift bracket 3 that moves up and down together with the fork 2, and detects the position/shape data and height of the second pallet P2 while the fork 2 moves up and down. The 2D-LiDAR is installed on the lift bracket 3 at an intermediate position between the roots of the forks 2L and 2R. Alternatively, by changing the distance measuring device S2 installed on the main body 1A of the unmanned forklift 1 to 3D-LiDAR as shown in Fig. 1, the position/shape data and height of the second pallet P2 can also be detected. . In these cases, the distance sensor S3 at the tip of the fork 2 is used to check for interference with the pallets P1 and P2.

For example, for the stacked pallets P1 and P2 shown in FIG. 7, after the unmanned forklift 1 detects the position/shape data and height of the second pallet P2, the fork 2 is moved to the insertion slot Q of the pallet P2. , the unmanned forklift 1 is advanced by a predetermined distance so that the left and right forks 2L, 2R are aligned with the insertion openings Q, Q of the second stage pallet P2. Next, the forks 2 are lifted, and since the forks 2 hold the cargo in a state where the load sensor S4 detects the pallet P2, the lift of the forks 2 is stopped after the forks 2 are lifted by a predetermined distance, and the unmanned forklift is operated. 1 backwards.

In the above embodiments of the present invention, the case where the target pallet is on the floor N has been described, but the target pallet may be on the bed of the truck. An example of work when the target pallet is on the bed of a truck is shown below.

(1) Devanning of the first pallet and cargo on the bed of the truck In the same operation as the devanning of the second pallet P2 and its cargo in the above "Another example of operation of destacking work", the bed of the truck Devanning the upper tier of pallets and cargo.

(2) Devanning the upper and lower pallets on the truck bed and the upper pallet and cargo.
In the same manner as the second pallet P2 and its cargo in the above-mentioned "another example of operation of unloading work", the upper and lower two pallets on the bed of the truck and the upper pallet of the cargo and Devanning the cargo.

If the upper and lower pallets on the truck bed and the cargo are stacked so that the upper pallet is aligned with the lower pallet, devanning the upper pallet and cargo on the truck bed is performed as follows. good too.

That is, the position/shape data and the height of the lower pallet on the loading platform are detected in the same manner as the second pallet P2 and its load in the "another operation example of the tier unpacking work". After the unmanned forklift 1 is stopped in front of the upper and lower pallets and the load, when the forks 2 are raised, the distance sensor S3 at the tip of the fork 2 detects an object in front of the front. stop the fork 2 from rising when the object is no longer detected. By moving the unmanned forklift 1 forward by a predetermined distance, lifting the forks 2 by a predetermined distance, stopping the rise of the forks 2 by means of a load sensor S4, and retreating the unmanned forklift 1, the upper pallet on the bed of the truck and the cargo are devanned. conduct.

(3) Vanning (stacking) of pallets and cargo onto a single pallet and cargo on the bed of a truck
2D-LiDAR, which is a distance measuring device S2, is installed on the lift bracket 3 that moves up and down together with the fork 2, and detects the position/shape data and height of a single pallet on the bed of the truck while the fork 2 moves up and down. do. Alternatively, by changing the distance measuring device S2 installed on the main body 1A of the unmanned forklift 1 to 3D-LiDAR as shown in Fig. 1, the position/shape data and height of one pallet on the bed of the truck can also be detected. It can be so.

Then, in the above "stacking work operation example", the target pallet P1 is replaced with a single pallet on the truck bed, and the pallet and cargo are vanned onto the single pallet and cargo on the truck bed. (Stacking).

In the work examples (1) to (3) above, the learning model used to identify the type of pallet is the first pallet (lower pallet) or the second pallet (upper pallet) on the bed of the truck. This is a trained model that has been machine-learned using the images of the palette) as training data.

<Effects of the unmanned forklift according to the embodiment of the present invention>
Even if objects to be transported by the unmanned forklift 1 are a plurality of types of pallets with different shapes, there is no limit to the types of pallets that can be handled, and machine learning is performed on combinations of the images of the respective pallets and the types of the pallets. The photographed image of the target pallet acquired by the image acquisition unit 11 is input to the learning model of the pallet type identification unit 12 that has already performed the above learning, and the pallet type identification unit 12 identifies the type of the target pallet. By inputting the photographed image, the pallet type identification unit 12 identifies the type of the target pallet with high accuracy. Then, the pallet displacement amount detection unit 14 detects the stored position/shape data corresponding to the type of the target pallet identified by the pallet type identification unit 12 and the target pallet obtained by the pallet position/shape acquisition unit 13. Position/shape data can be compared and collated.

As a result, the deviation amount of the position and attitude of the target pallet from the normal position of the pallet can be detected with high accuracy, and an article other than the pallet placed on the floor (for example, a cardboard box, etc.) is erroneously recognized as a specific type of pallet. By doing so, accidents such as interference between the forks 2 and pallets or articles can be prevented. In addition, since calculations for comparison and collation of pallet shape data are shortened and work speed is improved, there is no need to use an expensive computer with high processing power.

After detecting the deviation amount of the target pallet with high accuracy, the unmanned forklift 1 uses the deviation amount detected by the pallet deviation amount detection unit 14, for example, to the insertion port of the target pallet. Work including the operation of inserting the fork 2 includes work of picking up the pallet and cargo on the floor and transporting it to another location, and work of picking up the second pallet and cargo on top of the first pallet’s cargo and transporting it to another location. It can be performed. In addition, the unmanned forklift 1 performs a stacking operation, etc., in which the pallet and cargo held by inserting the fork 2 into the insertion port are loaded onto the cargo of the target pallet according to the deviation amount of the target pallet. can be done.

All of the descriptions of the above embodiments are examples and are not intended to be limiting. Various modifications and changes may be made without departing from the scope of the invention.

1 unmanned forklift 1A main body 2 fork 2L left fork 2R right fork 3 lift bracket 4 inner mast 5 outer mast 6 tilt cylinder 7 lift chain 8 cable bear 10 processing device 11 image acquisition unit 12 pallet type identification unit 13 pallet position/shape acquisition unit 14 pallet Deviation amount detection unit 15 Deviation amount error determination unit 16 Pallet shape determination unit 17 Pallet stacking determination unit 18 Travel control units 19 and 20 Mounting plate 21 Support plate 22 Seesaw plate 22A Operation piece 22B Detection piece 23 Stop plate 24 Extension coil spring 25, 26 Linear guides 27, 28 Electric cylinder with position sensor 27A, 28A Piston A Load handling device B Moving device C Control device D Fork width changing device E Predetermined distance F Spacing between forks G Width between insertion ports H Height of cargo I Support plate J Bracket K Frame L0 Line segment L1 based on stored point cloud data corresponding to the type of target pallet Line segment M Mast N Floor P based on the point cloud data of the target pallet acquired by the pallet position/shape acquisition unit Pallet P1, P2 Target pallet Q Slot R Spindle S1 Photographing device S2 Distance measurement device S3 Distance sensor S4 Inventory sensor S5 Self-position recognition sensor S6 Obstacle detection sensors T1 to T5 Operation U Predetermined distance V Object detected by distance sensor Detection W Cargo

Claims (9)

An unmanned forklift equipped with a pair of forks that automatically performs cargo handling work using pallets,
an image acquisition unit that acquires a photographed image from a photographing device that photographs the front of the unmanned forklift;
For a plurality of types of the palette, having a learned model that has undergone machine learning on combinations of the images of the palette and the types of the palette, and the captured image of the target palette acquired by the image acquisition unit. a pallet type identification unit that identifies the type of the target pallet by inputting it into the learning model;
a pallet position/shape acquisition unit that acquires position/shape data of the target pallet from a distance measuring device that measures the distance from the target pallet;
Position/shape data of each of the plurality of types of pallets is stored in advance, and the stored position/shape data corresponding to the type of the target pallet identified by the pallet type identification unit and the pallet position. a pallet deviation amount detection unit that detects the amount of deviation of the position and orientation of the target pallet from the normal position by comparing the position and shape data of the target pallet acquired by the shape acquisition unit;
(A) work including an operation of inserting the fork into the insertion opening of the target pallet using the detected deviation amount;
(B) a stacking operation of stacking the pallet and the load held by inserting the fork into the insertion port on top of the load of the target pallet according to the deviation amount of the target pallet;
A travel control unit that controls travel for performing at least one of the work of
comprising
driverless forklift. The position/shape data is height data including the insertion port of the pallet,
The pallet shift amount detection unit detects line segments based on the stored position/shape data corresponding to the type of the target pallet and the position/shape data of the target pallet acquired by the pallet position/shape acquisition unit. superimpose and compare with the line segment based on,
The unmanned forklift according to claim 1. further comprising a deviation amount error determination unit that determines an error when the deviation amount detected by the pallet deviation amount detection unit exceeds a predetermined threshold,
The unmanned forklift according to claim 1 or 2. The shape data of the target pallet acquired by the pallet position/shape acquisition unit is compared with the pre-stored shape data of a pallet of the same type as the target pallet. , further comprising a pallet shape determination unit that determines that the acquired shape data of the target pallet has an error;
An unmanned forklift according to any one of claims 1 to 3. By performing image processing on the photographed image of the target pallet acquired by the image acquisition unit, the presence or absence of stacking is determined by calculating the number of pallets within a specific range preset when the target pallet exists. Further comprising a pallet stacking discrimination unit,
An unmanned forklift according to any one of claims 1 to 4. The pallet stacking determination unit is machine-learned in advance using each pallet image as teacher data for multiple pallets in the upper and lower tiers.
The unmanned forklift according to claim 5. Further comprising a fork width changing device that changes the interval between the forks according to the width between the insertion openings of the target pallet according to the type of the target pallet obtained from the pallet type identification unit,
An unmanned forklift according to any one of claims 1 to 6. a distance sensor arranged at the tip of the fork for detecting the presence or absence of an object in front;
a load sensor that detects whether the fork holds the pallet;
further comprising
Using the detection result of the presence or absence of the object by the distance sensor and the detection result of whether or not the pallet is held by the load sensor,
A stacking operation in which the second pallet and cargo are loaded on top of the first pallet cargo, and/or
Carrying out the tier unpacking work of picking up the 2nd tier pallet and cargo on the 1st tier pallet and transporting it to another place,
An unmanned forklift according to any one of claims 1 to 7. At a first stop position, the photographing device photographs the target pallet, the pallet type identification unit identifies the type of the target pallet, and at a second stop position closer to the target pallet than the first stop position, the A distance measuring device measures the distance to the target pallet, and the pallet position/shape acquisition unit acquires the position/shape data of the target pallet;
An unmanned forklift according to any one of claims 1 to 8.

Images