JP2022188455A - Palette carrying system - Google Patents

Abstract

To provide a palette carrying system that reduces an occasion in which a position of a palette becomes undetectable.SOLUTION: A palette carrying system executes: a data collecting step of obtaining plural pieces of detection data each related to a reflection point that is regarded as a point where detection light is reflected; a direction classifying step including a corner point detecting step of detecting a corner point located at a position where an arrangement direction in which point groups where at least reflection points gather as a group changes from the plural pieces of detection data, a grouping step of grouping the point groups into plural groups divided by the corner point, and a classifying step of classifying a horizontal group 322 aligned in a widthwise direction of a palette or a vertical group 321 aligned in a depthwise direction orthogonal to the widthwise direction and a vertical direction regarding a direction in which the point groups of the group extend on the basis of approximate position information; and a palette position deriving step of deriving a center position of a pair of leg portions 12 using the plural pieces of detection data belonging to at least one of the vertical group or the horizontal group.SELECTED DRAWING: Figure 9

Description

TECHNICAL FIELD The present invention relates to a pallet transport system used when transporting cargo handling pallets by a transport vehicle capable of unmanned operation.

For example, in a manufacturing facility such as an ironworks, there is a method of transporting a load together with a pallet by allowing a transport vehicle to enter the space below a pallet for cargo handling and lifting the pallet. An example of this method is described in US Pat.

Regarding a method of guiding an unmanned transport vehicle to a space below a pallet, Patent Document 1 discloses that a laser beam is emitted from a sensor attached to the vehicle body to detect the relative position and angle of the pallet with respect to the vehicle body of the transport vehicle. By irradiating the pillars of both legs possessed by the pallet The direction of both legs and the center line of both legs are calculated. In addition, the column part consists of a rigid-frame structure or a truss structure using shaped steel (such as H-shaped steel).

Japanese Patent No. 4427360

However, in the above method, depending on the positional relationship of the transport vehicle with respect to the pallet, the laser light may not hit either the left or right column of the pallet, or the reflected light of the laser light may not be detected. In this case, the position of the pallet cannot be detected.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a pallet transport system that reduces the occurrence of situations in which the position of a pallet cannot be detected.

A pallet transport system of the present invention moves a pallet having a loading platform and a pair of legs for supporting the loading platform from below at intervals in the width direction while the pallet is placed below the loading platform and lifted from the ground. In the pallet transport system using a transport vehicle that allows the transport vehicle to move, the transport vehicle includes an irradiation unit that irradiates the pair of legs with detection light, and a light receiving unit that receives light reflected by the pair of legs from the detection light. and a general position grasping step of grasping general position information about the approximate position of the pallet with respect to the transport vehicle; a data collection step of acquiring a plurality of detection data relating to a reflection point treated as a point at which the detection light is reflected by receiving the reflected light; and a group of at least the reflection points gathered from the plurality of detection data A corner point detection step of detecting a corner point that is the reflection point at a position where the direction in which the point group is arranged changes, a grouping step of dividing the point group into a plurality of groups separated by the corner points, and the approximate position A classification step of classifying into horizontal groups aligned in the width direction of the pallet or vertical groups aligned in the depth direction orthogonal to the width direction and the vertical direction with respect to the direction in which the point cloud of the group extends based on the information. A direction classification step and a pallet position derivation step of deriving the center positions of the pair of legs using the plurality of detection data belonging to at least one of the vertical group and the horizontal group are performed.

According to this configuration, a plurality of pieces of detection data are divided into groups by corner points, and the groups are divided into vertical groups and horizontal groups according to the direction in which the point groups forming each group are arranged based on the general positions grasped by the general position grasping means. Classifying into groups and guiding the center position of a pair of legs based on a plurality of detection data belonging to at least one of a vertical group and a horizontal group. Pallet position can be detected even when light cannot be detected.

Further, in the direction classification step, when the direction in which the point groups of each group are arranged is substantially the same as the direction toward each group from the origin where the irradiation unit irradiates the detection light, the corresponding group is classified. It is also possible to include a group exclusion step for exclusion from classification targets in the process.

According to such a configuration, the direction classification step includes the group exclusion step of excluding groups from the classification target when the direction in which the group of points is arranged is substantially the same as the direction from the origin to the group. , the group extending parallel to the detection light can be excluded, thus increasing the accuracy of detecting the pallet position.

Further, in the group exclusion step, as the direction from the origin to each group, the direction from the origin to the midpoint between the corner points located at both ends of the direction in which the point groups of each group are arranged is used. can also

According to this configuration, the direction from the origin to each group is regarded as the direction from the origin of irradiation of the detection light to the midpoint between the corner points. Even if few, falsely detected groups can be excluded.

Further, the direction classification step is performed when one corner point detected by the corner point detection step is detected as another corner point when the reflection point closer to the two adjacent reflection points is detected as another corner point. , the one corner point may be regarded as an abnormal point and excluded from the objects in the grouping step.

According to such a configuration, among the detected corner points, if the closer one of the two reflection points adjacent to the corner point is the other corner point, the reflection point aligned in the same direction from the other corner point Since there is no reflection point, it is likely to be erroneous detection data caused by reflections from dust in the air or the like. Therefore, excluding this detection data increases the detection accuracy of the pallet position.

Further, the general position information may include information regarding the orientation of the pallet, and the classification step may classify into either the horizontal group or the vertical group based on the information regarding the orientation of the pallet.

According to this configuration, the direction in which the point clouds of the group are arranged is classified into a vertical group or a horizontal group based on the information about the orientation of the pallet grasped by the general position grasping means. Easy to accurately classify vertically or horizontally.

Further, the pallet has a plurality of pairs of legs, and the plurality of pairs of legs are provided side by side along the depth direction of the pallet. For at least one of the horizontal groups, an integrated group is derived by integrating one group and another group positioned substantially on an extension line in a direction in which the one group is arranged, and using the plurality of detection data belonging to the integrated group. can also be used to guide the center position of the pair of legs.

According to this configuration, the position of the pallet is derived based on the detection data of the integrated group obtained by integrating a plurality of groups. can be derived, making it easier to accurately derive the position of the pallet.

According to the present invention, it is possible to obtain a pallet conveying system that reduces the occurrence of a situation in which detection of the position of a pallet becomes impossible.

It is a left side view showing an example of a pallet used in a pallet conveying system concerning one embodiment of the present invention. It is a front view which shows an example of the said pallet with a conveyance vehicle (two-dot chain line display). It is a schematic diagram for demonstrating the data collection process in the same pallet conveying system. It is a schematic diagram which shows the outline of the detection data in the same pallet conveying system. It is a flowchart which shows the flow of the same pallet conveying system. It is a flowchart which shows the flow of the direction classification process in the same pallet conveying system. It is a schematic diagram which shows the outline|summary of the group exclusion process in the same pallet conveying system. It is a schematic diagram for explaining the state in which the direction classification process of the same pallet conveying system is completed. It is a schematic diagram for demonstrating the pallet position derivation|leading-out process of the same pallet conveying system.

DESCRIPTION OF THE PREFERRED EMBODIMENTS A pallet transport system of the present invention will be described below with reference to the drawings by taking up one embodiment.

The pallet transport system of the present embodiment is a system using a transport vehicle 2 that moves the pallet 1 while floating it from the ground in a state that the pallet 1 is placed under a loading platform 11 of the pallet 1 . When controlling the transport vehicle 2 remotely, a control unit (not shown) is provided at a position away from the transport vehicle 2 .

First, the pallet 1, which is an object to which the pallet conveying system of the present embodiment is applied, will be described. As shown in FIGS. 1 and 2, the pallet 1 has a loading platform 11 on which cargo can be loaded, and a pair of legs 12 supporting the loading platform 11 from below with a gap in the width direction (horizontal direction in FIG. 2). , has As shown in FIG. 2, the pair of leg portions 12 in this embodiment are opposed to each other with a gap left and right. Further, the pallet 1 of the present embodiment has a plurality of pairs of legs 12 spaced apart in the depth direction of the loading platform 11 (horizontal direction in FIG. 1). Each of the pair of leg portions 12 (each leg portion 12) has a plurality of column portions 121 extending in the vertical direction and a leg bottom portion 122 connecting the plurality of column portions 121 downward. In this embodiment, each pillar 121 extends in the vertical direction (perpendicular to the ground of the pallet storage area Y), and the loading platform 11 and each leg 12 form a Rahmen structure. However, it is not limited to this, and each column portion 121 may extend in an oblique direction. In this case, it becomes a rigid-frame structure or a truss structure. As shown in FIG. 2 , the pallet 1 of this embodiment has a so-called “gate type” shape, and a space 11S is formed below the loading platform 11 .

The transport vehicle 2 is generally called a "carrier pallet vehicle", and is formed to have a vehicle height that allows it to be arranged so as to enter a space 11S that exists below the loading platform 11 of the pallet 1 (two-dot chain line in FIG. 2). ). By extending the vehicle height, the pallet 1 can be moved while floating from the ground. The transport vehicle 2 of the present embodiment has a cab at least at the front end and can be manned. However, when used in the pallet transport system of the present embodiment, it is mainly unmanned. However, the pallet transport system of the present embodiment may be applied as an auxiliary during manned travel. Further, the transport vehicle 2 is not limited to having a driver's cab only at the front end, and may be configured to have a cab at both the front end and the rear end. That is, the driver's cab of the transport vehicle 2 can be placed at any position as long as the transport vehicle 2 can be operated during manned travel. The transport vehicle 2 of the present embodiment has a storage unit (not shown) that stores map data including information about the position at which the pallet 1 is arranged, and grasps general position information about the approximate position of the transport vehicle 2 with respect to the pallet 1. and general position grasping means. Examples of the rough position grasping means include, but are not limited to, means including a global positioning system (GNSS). For example, map data stored in the control unit of the transport vehicle 2 and an image obtained by an in-vehicle camera. Alternatively, the rough position can be grasped based on the map data and the route data (travel route history data) traveled by the transport vehicle 2 in the pallet storage Y. It can also be configured to The transport vehicle 2 also includes an irradiation unit 21 that irradiates the detection light toward the pallet 1 (specifically, each leg 12 of the pallet 1) facing each other, and a reflection unit 21 that reflects the irradiated detection light on the pallet 1. and a light receiving portion 22 for receiving light. As the irradiation unit 21 and the light receiving unit 22, various sensors (for example, LiDAR, ToF camera, etc.) that perform detection by emitting and receiving light rays can be used. A two-dot chain line shown in FIG. 3 corresponds to the detected light and the reflected light. The irradiation unit 21 of the present embodiment performs scanning by rotating around an axis extending vertically with respect to the vehicle body of the transport vehicle 2 (for example, rotating in the direction R). Further, as shown in FIG. 4 , it is possible to obtain data on the reflection point 3 where the detection light emitted by the irradiation unit 21 is reflected from the reflected light received by the light receiving unit 22 . In addition, the scanning method of the irradiation unit 21 is not limited to the above-described method, and various methods can be adopted.

The pair of legs 12 of the pallet 1 (specifically, the pillars 121 of the legs 12) each have a facing surface facing the carrier 2 when the carrier 2 is arranged on the front side of the pallet 1. have. The column portion 121 in each leg portion 12 of this embodiment is formed of H-shaped steel. A plane, which is the outer surface of the web H2 of the H-section steel, is the opposing surface (see FIG. 4). Note that the orientation of the H-shaped steel in FIG. 4 is an example, and for example, the illustrated state may be rotated 90 degrees about the vertical direction. Moreover, in the pallet 1, the column portion 121 in each leg portion 12 was formed of H-shaped steel. However, the material is not limited to this, and various materials can be used as long as the material has a facing surface facing the carrier 2 . For this reason, the column portion 121 may be, for example, a rectangular steel pipe, a channel material, or an angle material. Furthermore, a circular steel pipe may be used for the pillar 121, a flat plate may be attached to the side surface of the part that receives the detection light emitted from the irradiation unit 21 of the transport vehicle 2, and the surface of this flat plate may be the opposing surface. .

Next, a plurality of processes performed in the pallet transport system of this embodiment will be described with reference to FIG. First, the transport vehicle 2 grasps the current general position information of the transport vehicle 2 by means of the general position grasping means, matches the map data with the rough position information, moves, and The vehicle stops in front of the pair of legs 12. - 特許庁The movement of the transport vehicle 2 until it stops may be performed autonomously by a control unit provided on the vehicle, or may be performed by receiving an instruction, for example, wirelessly from a control unit provided outside the vehicle.

In the pallet transport system of the present embodiment, before the transport vehicle 2 enters below the loading platform 11 of the pallet 1 (before the front end of the transport vehicle 2 touches the front end of the pair of legs 12), the main Then, a target area setting step S1, a data collection step S2, a direction classification step S3, a pallet position derivation step S4, and a transport vehicle movement step are performed in order. In addition to these steps, other auxiliary steps can also be performed. These processes are autonomously carried out by a control section provided on the vehicle. Alternatively, it is performed by a control unit provided outside the vehicle, and the operation on the vehicle is performed by receiving an instruction, for example, by radio, from the control unit outside the vehicle.

In the target area setting step S1, a step of setting a target area in which the pair of legs 12 of the pallet 1 is estimated to exist based on the information on the current position of the transport vehicle 2 and the information on the position of the pallet 1 grasped by the general position grasping means. is. In the target area setting step S1, information on the position of the pallet 1 is acquired from the map data of the pallet storage area Y, and the information on the position of the pallet 1 is collated with the general position grasped by the general position grasping means to determine the pair of pallets 1. A target area is set for each leg 12 (in this embodiment, the left and right legs 12 as seen from the transport vehicle 2) where the leg 12 is estimated to exist. In addition to the above method, the target area can also be set by image recognition using a camera provided on the carrier 2 .

In the data collection step S2, the target area in which the pair of legs 12 of the pallet 1 is estimated to exist, which is set in the target area setting step S1, is irradiated with the detection light from the irradiation unit 21 of the transport vehicle 2, and the target area is transported. A plurality of detection data (a group of detection data) relating to the reflection point 3, which is the point at which the detection light is reflected, is obtained by scanning the light receiving unit 22 of the vehicle 2 to receive the reflected light. The detection data acquired in the data collection step S2 may contain noise. This noise is eliminated in later steps. Also, the scanning intervals (angular intervals) in the data collection step S2 are uniform. In addition, in FIG. 3, since the detected light beams are schematically indicated by two-dot chain lines, the two-dot chain lines are not evenly spaced. In FIGS. 4, 7, and 8, the point group (reflection point 3) of the detection data is schematically indicated by a circle. Also, although a large number of reflection points 3 are actually detected, the minimum number is shown for convenience of explanation. Laser light is used as the detection light. However, the detection light is not limited to laser light, and may be other types of light.

As shown in FIG. 6, in the direction classification step S3, from a plurality of detection data, a corner inspection is performed to detect a corner point 31, which is the reflection point 3 at a position where the direction in which the group of points in which the reflection points 3 are gathered changes. Each step of output step S31, grouping step S32 of dividing the point group into a plurality of groups 32 separated by corner points 31, classification step S33 of classifying the direction in which the groups 32 extend, and corner point processing step S34 are executed. . Further, the direction classification step S3 includes a step of detecting an abnormal point to be excluded from the classification target by regarding detection data determined to be noise among the plurality of detection data as an abnormal point. In this embodiment, the step of detecting an abnormal point is appropriately performed in the direction classification step S3 (specifically, the first group exclusion step S332, the second group exclusion step S334, and the abnormal corner point detection step S341).

In the corner point detection step S31, a corner point 31, which is the reflection point 3, is detected at a position where the direction in which the group of points in which the reflection points 3 are gathered changes. That is, in the corner point detection step S31, when the reflection points 3 are continuously connected in the order in which the detection light is irradiated during scanning, the corner reflection points 3 are detected as the corner points 31. FIG. This corner point 31 is a reflection point corresponding to the corner formed on the flange of the H-beam and the corner where the web and the flange intersect. However, at the time of detection, other reflection points 3 are also included, and are excluded in subsequent steps. Further, in the scanning of one leg 12, the first reflection point 3 and the last reflection point 3 in the scanning order are also detected as corner points 31 for the convenience of subsequent processing. Scanning of one leg 12 can be determined based on the target area set in the target area setting step S1, or based on the distance between the detected reflection points 3, or the like. In the corner point detection step S31 of the present embodiment, for each reflection point 3, a front point group and a rear point group are obtained within a predetermined distance from the reflection point 3 in the order in which the detection light is irradiated during scanning. Next, for each of the previous point group and the latter point group, the position of the center of gravity of the point group is determined, and the angle formed by the reflection point 3 and the direction of each center of gravity (the angle formed by the three points with reference to the reflection point 3 angle). Then, the corner point 31 is determined as the reflection point 3 whose angle is equal to or less than the threshold value and whose angle is the smallest in the order in which the detection light is irradiated during scanning. Further, the method of detecting the corner point 31 is not limited to the above method, and various processing methods for detecting the corner point 31 can be adopted. For example, in parallel with the above method, when the distance between two adjacent reflection points 3 is greater than or equal to a threshold, both of the two points can be detected as corner points 31 . According to such a method, when two adjacent points are the reflection points 3 where the detection light is reflected by another leg 12 , the two points can be detected as another group 32 . In addition, the threshold value of the above method can be, for example, half the length of the width of the flange H1 of the H-section steel. By setting the thresholds in this way, it becomes possible to detect the reflection points 3 at the end of the flange H1 and the reflection points 3 at the middle portion of the width of the web H2 as separate groups 32. FIG.

The grouping step S<b>32 is a step of dividing the point group into groups 32 separated by corner points 31 . In the grouping step S<b>32 , a group of points sandwiched between the corner points 31 is classified as one group 32 for the points that are not classified as the corner points 31 in the point group. That is, at the time of the grouping step S32, the corner points 31 do not constitute the group 32 and are used in the classification step S33. Note that the corner points 31 can eventually be incorporated into groups 32 . Specifically, among the point groups, the point groups that are not identified as the corner points 31 are classified as one group 32 that includes points from one corner point 31 to the next corner point 31 in the scanning order. do. Grouping can also be performed according to the coordinates of the point cloud. In other words, a group of points positioned close to a straight line connecting one corner point 31 to another corner point 31 can be classified as one group 32 . In the present embodiment, "nearly positioned" refers to being positioned on a straight line or being shifted from the straight line within a margin of error. Note that the range recognized as the error range can be appropriately set depending on the performance of the light receiving section 22, the required detection accuracy, and the like. Grouping can also be performed so that the corner point 31 is included in the group 32 .

After the grouping of the point cloud is completed, the classification step S33 is performed. In the classification step S33, it is classified whether the direction in which the point group of each group 32 extends is the horizontal direction along the width direction of the pallet 1 or the vertical direction along the depth direction of the pallet 1. Also, in the classification step S33, a group exclusion step, which is a step of detecting abnormal points, is executed. In the classification step S33 of this embodiment, the steps described later are sequentially performed for each group 32, but this is not the only case, and each step is applied to all groups 32 before proceeding to the next step. You can also

In the classification step S33, first, the direction in which the group of points of the group 32 extends is determined (step S331). In this embodiment, the direction in which the group of points of the group 32 extends is the direction in which a straight line connecting a pair of corner points 31 (corner points 31 positioned at both ends in the direction in which the group of points of the group 32 extends) sandwiching the group 32 extends. It is determined as the direction in which the group of points of the group 32 extends. It should be noted that the direction in which the point group of the group 32 extends can be determined not only by such a method but also by other methods. For example, if there are a plurality of reflection points 3 included in the group 32, the direction in which the straight line that minimizes the sum of the distances to each reflection point 3 extends can be determined as the direction in which the point group of the group 32 extends. However, as in the corner point detection step S31, the direction can also be determined using the position of the center of gravity. Further, in this embodiment, the direction in which the point group of the group 32 extends is determined based on the angle of the sensor of the irradiation unit 21 with respect to the horizontal axis 21A. , extending in a direction shifted by an angle φ with respect to the horizontal axis 21A of the sensor of the irradiation unit 21 (in the figure, a virtual line 21B parallel to the horizontal axis of the sensor).

After the direction in which the point group of the group 32 extends is determined, as a determination of whether or not the point group of the group 32 is the group 32 formed by the point group detected by noise during detection, the first group An exclusion step S332 is performed. In the first group exclusion step S332, when the direction in which the point group of the group 32 extends is substantially the same as the direction toward the group 32 from the origin 211 in which the irradiation unit 21 irradiates the detection light, the point group that constitutes the corresponding group 32 is classified as an abnormal point, and the group 32 is excluded from classification targets in the classification step S33. A straight line extending parallel to the direction in which the detection light is emitted from the origin 211 cannot be originally detected because the detection light does not reflect. (For example, when the detection light is reflected on two or more surfaces of the leg 12, this abnormality is likely to occur due to the shape of the H-section steel.). In the first group exclusion step S332 of the present embodiment, a straight line connecting the origin 211 and the midpoint 33 between the pair of corner points 31 sandwiching the group 32 is used as the direction from the origin 211 toward the group 32 . Specifically, the angle φM between the straight line extending from the origin 211 toward the midpoint 33 and the horizontal axis 21A of the sensor is substantially the same as the angle φ indicating the direction in which the point group of the group 32 shown in FIG. 7 extends. In this case, it is determined that the direction in which the group of points of the group 32 extends is substantially the same as the direction from the origin 211 where the irradiation unit 21 irradiates the detection light toward the group 32 . By using a straight line connecting the middle point 33 of the pair of corner points 31 sandwiching the group 32 and the origin 211 as the direction from the origin 211 to the group 32, the detection data (reflection Even if the number of points 3) is small, a group 32 composed of abnormal points can be detected. Note that the case where the direction in which the point group of the group 32 extends and the direction from the origin 211 to the group 32 are substantially the same is not limited to the case where the direction in which the point group of the group 32 extends and the direction from the origin 211 to the group 32 are exactly the same. , including the case where it deviates within the range of error. The range that is accepted as the error range can be appropriately set depending on the performance of the light receiving section 22, the required detection accuracy, and the like.

For the groups 32 not excluded in the first group exclusion step S332, the direction in which the group of points of the groups 32 extends is aligned in the horizontal group 322 aligned in the width direction of the pallet 1, or aligned in the depth direction orthogonal to the width direction and the vertical direction. A direction identification step S333 for classifying into the vertical group 321 is executed. Classification of the direction in which the point cloud of the group 32 extends is performed based on the general position information. Specifically, the control unit predicts the degree of inclination and direction of the pallet 1 with respect to the transport vehicle 2 from the positional relationship between the transport vehicle 2 and the pallet 1 included in the general position information. , predict the width direction and the depth direction of the pallet 1 . If the tilt of the direction in which the point cloud of the group 32 extends with respect to the predicted width direction is within the threshold (extends in substantially the same direction), the point cloud of the group 32 is a horizontal group 322 arranged in the width direction. I judge. If the tilt of the direction in which the point group of the group 32 extends with respect to the predicted depth direction is within a threshold value (extending in substantially the same direction), the point group of the group 32 is a vertical group 321 arranged in the depth direction. I judge. Note that the threshold range can be set based on the range that can occur as a prediction error.

After the direction identification step S333 is completed, the process proceeds to the second group exclusion step S334. In the second group exclusion step S334, the group of points that constitute the group 32 that was not classified into either the vertical group 321 or the horizontal group 322 in the direction identification step S333 is determined as an abnormal point, and the group 32 is subjected to the direction classification step S3. It is a step of determining that it should be excluded from the target. By excluding the groups 32 classified into neither the vertical group 321 nor the horizontal group 322 in the direction identification step S333, the groups 32 including detection data caused by noise can be excluded.

Although the main steps of the direction classification step S3 are completed by the above steps, the direction classification step S3 can additionally include a corner point processing step S34 as an auxiliary step. For each corner point 31, the corner point processing step includes an abnormal corner point detection step S341 for determining whether or not the corner point 31 is an abnormal point, and a corner point 31 determined not to be an abnormal point to be placed in the nearest group. and a group addition step S342, which is the step of adding to .32. Only one of the abnormal corner point detection step S341 and the group addition step S342 may be included in the direction classification step S3.

In the abnormal corner point detection step S341, when the reflection point 3 located closest to the corner point 31 is detected as another corner point 31, the corner point 31 is regarded as an abnormal point. Through this process, it is possible to determine the reflection point 3, which is detected by being reflected by dust or the like and is floating with respect to other point groups, as an abnormal point.

In the group addition step S342, the corner point 31 determined as not being an abnormal point is added to the closest group 32. FIG. In the present embodiment, in the group addition step S342, for two groups 32 positioned adjacent to the corner point 31, the reflection point 3 closest to the corner point 31 is detected from among the point groups forming each group 32. , the reflection point 3 closest to the corner point 31 of each group 32 determines the group 32 closer to the corner point 31 as the closest group 32 . That is, in the group addition step S342, the corner point 31 is added to the group 32 including the reflection point 3 closest to the corner point 31. FIG. By this process, all reflection points 3 except anomalous points can be included in any group 32 .

The direction classification step S3 is completed by the above steps. In this embodiment, since the corner point processing step S34 is executed, as shown in FIG. All the reflection points 3 other than 1 belong to one of the groups 32 .

After the above-described processing is completed for the detection data regarding each leg 12 set in the target area setting step S1, the pallet position deriving step S4 for deriving the pallet position is performed. In the pallet position derivation step S4, an integration step of integrating a plurality of groups 32 extending in the same direction, and a position calculation step of computing the pallet position with respect to the transport vehicle 2 based on the integrated group in which the plurality of groups 32 are integrated in the integration step. and run Also, in the pallet position deriving step S4 of the present embodiment, a case where the transport vehicle 2 (specifically, the origin 211) is positioned in the middle of the pair of legs 12 in the width direction will be described. That is, a case where the transport vehicle 2 grasped by the general position grasping means is positioned in the middle of the pair of legs 12 of the pallet 1 in the width direction will be described.

As shown in FIG. 9 , in the integration process, for vertical groups 321 that are groups 32 of points extending along the depth direction of the pallet 1 , an integrated vertical group 41 obtained by integrating a plurality of vertical groups 321 and width of the pallet 1 For horizontal groups 322, which are groups 32 in which point groups extend along a direction, an integrated horizontal group 42 obtained by integrating a plurality of horizontal groups 322 is determined.

The indexing of the integrated vertical group 41 will be described. When determining the integrated vertical group 41, one vertical group 321 detected for one leg 12 is used as a reference, and another vertical group on a substantially straight line in the direction in which the point group of the one vertical group 321 extends 321 is integrated with one vertical group 321 to form an integrated vertical group 41 . In the selection of one leg 12, for example, the leg 12 closest to the transport vehicle 2 can be targeted, but the selection is not limited to this. In selecting the vertical group 321 of the reference position, the vertical group 321 having the longest distance between both ends of the point group of the vertical group 321 can be selected as the vertical group 321 to be the reference. In the case of this embodiment, the origin 211 is located in the middle of the pair of legs 12 in the width direction, so the vertical group 32 of the point group reflected on the inner surface of the legs 12 in the width direction is used as the reference. becomes a vertical group 321 . Furthermore, among the vertical groups 321, the length of the vertical group 321 having the longest distance between both ends is equal to or less than the threshold with respect to the design length of the leg 12 in the depth direction (the width of the flange H1 of the H-shaped steel). If so, it can be treated as failure in integration of the vertical group 321 . The threshold range can be set, for example, such that the length of the longitudinal group 321 with the longest distance between both ends is 60% or more of the design length of the leg 12 in the depth direction. By processing in this manner, the vertical group 321 detected for the leg 12 can be reliably used as a reference, so the reliability of the derived position of the pallet 1 increases.

Once the reference vertical group 321 is determined, an integrated vertical group 41 is derived by integrating the other vertical groups 321 positioned substantially on the extended line of the point group forming the reference vertical group 321 . Approximately on the extension line is not limited to the case where the other vertical group 321 is positioned on the extension line of the direction in which the point group of the vertical group 321 serving as the reference extends, but the extension line and the other vertical group 321 are within the range of error. Including cases where there is a deviation. The range that is accepted as the error range can be appropriately set depending on the performance of the light receiving section 22, the required detection accuracy, and the like. Also, if there is an error, the integrated vertical group 41 is determined in consideration of the error.

The indexing of the integrated horizontal group 42 will be described. When determining the integrated horizontal group 42, one horizontal group 322 detected for one leg 12 of the pair of legs 12 detected on the frontmost side is taken as a reference, and the points of one horizontal group 322 are calculated. The other horizontal groups 322 on a substantially straight line in the direction in which the groups extend are integrated with the one horizontal group 322 to form an integrated horizontal group 42 . In selecting the horizontal group 322 of the reference position, the horizontal group 322 having the longest distance between both ends of the point group of the horizontal group 322 can be selected as the horizontal group 322 to be the reference. Further, among the horizontal groups 322, the length of the horizontal group 322 having the longest distance between both ends is equal to or less than the threshold with respect to the design width direction length of the leg portion 12 (the width of the web H2 of the H-section steel). If it is not, it can be treated as failure in integration of horizontal group 322 . By processing in this way, the horizontal group 322 detected for the leg 12 can be reliably used as a reference, so the reliability of the derived position of the pallet 1 is increased.

Once the reference horizontal group 322 is determined, an integrated horizontal group 42 is derived by integrating the other horizontal groups 322 positioned substantially on an extension line in the direction in which the point group forming the reference horizontal group 322 extends. Approximately on an extension line is not limited to the case where the other horizontal group 322 is positioned on the extension line in the direction in which the point group of the horizontal group 322 serving as the reference extends, and the extension line and the other horizontal group 322 are within the range of error. Including cases where there is a deviation. The range that is accepted as the error range can be appropriately set depending on the performance of the light receiving section 22, the required detection accuracy, and the like. Also, if there is an error, the integrated horizontal group 42 is determined in consideration of the error.

The above procedure completes the integration process. After the integration process is completed, the process shifts to the position calculation process of computing the pallet position with respect to the transport vehicle 2 based on the integration group. In addition, if neither the integrated vertical group 41 nor the integrated horizontal group 42 can be determined in the integration process, the process does not proceed to the position calculation process, and an error is issued as failure to derive the pallet position. End the process. After the process ends due to an error, the transport vehicle 2 is brought closer to the pallet 1 again, and then the target area setting process S1, the data collection process S2, the direction classification process S3, the pallet position derivation process S4, and the transport vehicle movement process are performed in order. implement.

In the position calculation process, in the integration process, the integrated vertical group 41 and the integrated horizontal group 42 could be determined, only the integrated vertical group 41 could be determined, and only the integrated horizontal group 42 could be determined. The method of calculating the position of the pallet 1 differs between the cases. Each case will be described below.

In the integration process, if the integrated vertical group 41 and the integrated horizontal group 42 can be determined, based on the integrated vertical group 41, the position of the pallet 1 in the width direction with respect to the transport vehicle 2 is calculated, and the integrated horizontal group 42 is calculated. , the position of the pallet 1 in the depth direction with respect to the transport vehicle 2 is calculated. Specifically, the position of the pallet 1 in the width direction with respect to the transport vehicle 2 is calculated based on the central line C in the width direction of the pallet 1 obtained based on the integrated vertical group 41 and design values of the dimensions of the pallet 1. . That is, by calculating the distance from the carrier 2 to the center line C, the position of the pallet 1 in the width direction with respect to the carrier 2 is calculated. The center line C in the width direction of the pallet 1 is a straight line extending parallel to the integrated vertical group 41 which is virtually provided at a predetermined position away from the integrated vertical group 41 based on the design values of the pallet 1 dimensions. Desired. Since the central line C of this embodiment is a straight line extending along the width of the flange H1 on the inner side in the width direction of the H-shaped steel, the center line C is the width direction of the pallet 1 from the inner surface of the leg portion 12. A center line C is an imaginary straight line extending parallel to the integrated vertical group 41 at a position spaced inward in the width direction from the integrated vertical group 41 by the distance to the center. The position of the pallet 1 in the depth direction with respect to the transport vehicle 2 is obtained based on the design values of the dimensions of the integrated horizontal group 42 and the pallet 1 . In this embodiment, the integrated horizontal group 42 is a straight line extending along the width of the H-shaped steel web H2, so the distance in the depth direction from the transport vehicle 2 to the pallet 1 is the integrated horizontal group 42 and the transport vehicle 2 is calculated by subtracting the distance between the front end of the pallet 1 and the web H2 of the legs 12 from the distance of .

If only the integrated vertical group 41 can be determined in the integration process, the calculation is performed based on the design values of the integrated vertical group 41 and the dimensions of the pallet 1 (previously stored in the control unit). The position of the pallet 1 in the width direction with respect to the transport vehicle 2 is calculated in the same manner as when the integrated vertical group 41 and the integrated horizontal group 42 can be determined. Further, the position of the pallet 1 in the depth direction with respect to the transport vehicle 2 is obtained based on the frontmost point in the point group of the integrated vertical group 41 and the design value of the pallet 1 dimension. Specifically, the distance between the transport vehicle 2 and the pallet 1 in the depth direction is defined by a virtual line that passes through the frontmost reflection point 3 of the integrated vertical group 41 and extends in the horizontal direction so as to be orthogonal to the integrated vertical group 41 and the transport vehicle. 2 minus the distance between the front end of the leg 12 and the front end of the pallet 1.

When only the integrated horizontal group 42 can be determined in the integration process, the calculation is performed based on the design values of the integrated horizontal group 42 and the dimensions of the pallet 1 (previously stored in the control unit). The position of the pallet 1 in the depth direction with respect to the transport vehicle 2 is calculated in the same manner as when the integrated vertical group 41 and the integrated horizontal group 42 can be determined. In addition, regarding the position of the pallet 1 in the width direction with respect to the transport vehicle 2, in the point group of the integrated horizontal group 42, the center points of the reflection points 3 located at both ends (outer ends of the horizontal group 322) A straight line extending horizontally perpendicular to the group 42 is used as the center line C, and the position of the pallet 1 in the width direction with respect to the transport vehicle 2 in the integrated vertical group 41 and integrated horizontal group 42 is calculated in the same manner as described above. This is because the detection light does not hit the inner end of the web H2 due to the relationship between the arrangement of the H-section steel and the position of the transport vehicle 2 in this embodiment. Specifically, by calculating the distance from the transport vehicle 2 to the virtual center line C, the position of the pallet 1 in the width direction with respect to the transport vehicle 2 is computed.

Through the steps described above, the position of the pallet 1 relative to the transport vehicle 2 is determined. As described above, the case where the integrated vertical group 41 and the integrated horizontal group 42 could be determined, the case where only the integrated vertical group 41 could be determined, and the case where only the integrated horizontal group 42 could be determined. Since the method of calculating the position of the pallet 1 is different, when the integrated vertical group 41 and the integrated horizontal group 42 are determined, the calculation is performed based on a plurality of information, so highly reliable information can be obtained and integrated. Information on the position of the pallet 1 can be appropriately obtained even when only one of the vertical group 41 and the integrated horizontal group 42 is determined.

The carriage moving step is a step of moving the carriage along the obtained center line C (center position of the pair of legs 12). Specifically, it is a step of moving the carriage 2 toward the pallet position derived in the above step. Since the positions of the pallet 1 in the width direction and the depth direction with respect to the carriage 2 are obtained from the above steps, the carriage 2 can be properly slid under the pallet 1 based on the obtained information. Specifically, since it has information about the center line C and information about the distance to the pallet 1, the center line C is set so that the transport vehicle 2 does not hit the pallet 1 (specifically, the legs 12). It can be hidden under the pallet according to the

According to the pallet transport system as described above, a plurality of detection data are separated into groups by the corner points 31, and the direction in which the point groups constituting each group 32 are arranged based on the approximate positions grasped by the rough position grasping means is arranged. The group 32 is classified into a vertical group 321 and a horizontal group 322 by , and the center position (center line C) of the pair of legs 12 is derived based on a plurality of detection data belonging to at least one of the vertical group 321 and the horizontal group 322. , the position of the pallet 1 can be detected even if the detection light cannot be applied to a part of the legs 12 of the pallet 1 or the reflected light cannot be detected.

In the above embodiment, the case where the transport vehicle 2 (the origin 211 of the irradiation unit 21) is positioned in the middle of the pair of legs 12 in the width direction has been described. This system can also be applied when the vehicle 2 is positioned on one side in the width direction with respect to both of the pair of legs 12 . As described above, when the transport vehicle 2 is located far away from the pallet 1, for example, the reception of radio waves such as GNSS may fail, or the accuracy of GNSS may be low. In such a case, the integrated vertical group 41 can be obtained by calculating the integrated vertical group 41 as a straight line extending along the outer surface of the leg portion 12 in the width direction. 41 and the design values of the dimensions of the pallet 1, the position of the pallet 1 with respect to the carrier 2 can be calculated.

In addition, when the carrier 2 is located on one side in the width direction with respect to both of the pair of legs 12 and the integrated vertical group 41 cannot be determined, the pallet for the carrier 2 is moved based on the horizontal group 322. The position of 1 can be computed. Specifically, the distance in the depth direction between the transport vehicle 2 and the pallet 1 can be calculated from the distance to the integrated horizontal group 42 . In addition, among the horizontal groups 322 detected for each of the pair of legs 12, the end point (reflection point 3) on one side in the width direction of one leg 12 and the one side in the width direction of the other leg 12 By referring to the design values of the dimensions of the pallet 1, the center line C, which is the center position of the pallet 1 in the width direction, can be derived. By calculating the distance between the derived center line C and the carrier 2, the position of the pallet 1 with respect to the carrier 2 can be calculated.

DESCRIPTION OF SYMBOLS 1... Pallet, 11... Cargo bed, 12... Leg, 121... Column, 122... Leg bottom, 2... Carrier, 21... Irradiation part, 211... Origin, 22... Light receiving part, 3... Reflection point, 31... Corner Point, 32... Group, 321... Vertical group, 322... Horizontal group, 33... Middle point, 41... Integrated vertical group, 42... Integrated horizontal group, center line C

Claims (6)

A pallet conveying system using a transport vehicle that moves the pallet while floating it from the ground while the pallet has a loading platform and a pair of legs that support the loading platform from below at intervals in the width direction, and the pallet is placed under the loading platform. ,
The transport vehicle has an irradiation unit that irradiates the pair of legs with detection light, and a light receiving unit that receives light reflected by the pair of legs from the detection light,
a general position grasping step of grasping general position information about the approximate position of the pallet with respect to the transport vehicle;
By irradiating at least the pair of legs with the detection light from the irradiation unit and receiving the reflected light with the light receiving unit, a plurality of detection data regarding a reflection point treated as a point where the detection light is reflected is acquired. a data collection step for
A corner point detection step of detecting, from the plurality of detection data, a corner point that is the reflection point located at a position where the direction in which the group of points in which the group of reflection points are gathered changes, and dividing the group of points by the corner point. and a horizontal group aligned in the width direction of the pallet or orthogonal to the width direction and the vertical direction with respect to the direction in which the point group of the group extends based on the general position information. a direction classification step including each classification step of classifying into vertical groups arranged in the depth direction;
a pallet position derivation step of deriving the center position of the pair of legs using the plurality of detection data belonging to at least one of the vertical group and the horizontal group;
A pallet transport system that implements In the direction classification step, when the direction in which the point groups of each group are arranged is substantially the same as the direction toward each group from the origin where the irradiation unit irradiates the detection light, the corresponding group is selected in the classification step. 2. The pallet transport system according to claim 1, further comprising a group exclusion process for excluding items from classification targets. 3. In the group exclusion step, as the direction from the origin to each of the groups, a direction from the origin to a midpoint between the corner points located at both ends of a direction in which the point groups of the groups are arranged is used. 2. The pallet conveying system according to 2. The direction classification step is performed when one corner point detected by the corner point detection step is detected as another corner point when the reflection point closer to the two adjacent reflection points is detected as the other corner point. 4. The pallet conveying system according to any one of claims 1 to 3, wherein one corner point is regarded as an abnormal point and excluded from targets in the grouping step. The general position information includes information about the orientation of the pallet,
5. The pallet conveying system according to any one of claims 1 to 4, wherein the sorting step sorts the pallets into either the horizontal group or the vertical group based on information about the orientation of the pallets. The pallet comprises a plurality of the pair of legs,
The plurality of pairs of legs are provided side by side along the depth direction of the pallet,
The pallet position derivation step includes:
For at least one of the vertical group and the horizontal group, deriving an integrated group by integrating one group and another group located substantially on an extension line in the direction in which the one group is arranged,
6. The pallet conveying system according to any one of claims 1 to 5, wherein the plurality of pieces of detection data belonging to the integrated group are used to derive the central positions of the pair of legs.

Images