Conveying system, conveying method, and conveying program

Abstract

The present invention provides a transport system, transport method, and transport program that can reduce the likelihood of workers performing tasks around an automated guided vehicle (AGV) feeling anxious about the AGV's movement. [Solution] The transport system, transport method, and transport program use a controller connected to sensors that generate distance point data for multiple distance points in an area where one or more automated guided vehicles (AGVs) and workers and AGVs may be present. The controller acquires distance point data from the sensors, generates difference data showing the difference between the distance point data and reference data, identifies the position of workers in the area based on the difference data, and controls the movement of the AGVs based on their positions.

Description

Technical Field

[0001] The present disclosure relates to a conveyance system, a conveyance method, and a conveyance program.

Background Art

[0002] Patent Document 1 discloses a technology related to a mobile body traveling system including a first mobile body capable of traveling in pursuit of a target object to be pursued and a second mobile body capable of autonomous traveling based on its own position. When the second mobile body cannot detect a moving body with the mounted detection unit, it autonomously travels aiming at the first mobile body based on the current position of the first mobile body and the own position of the second mobile body.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] According to the technology described in Patent Document 1, a detection unit is provided on the unmanned carrier itself for the purpose of pursuing a target object such as an operator. When the operator hides in a blind spot of other obstacles, the operator may not be accurately detected. Due to such a situation, there is a problem that an operator working around the unmanned carrier feels uneasy about the movement of the unmanned carrier.

[0005] The present disclosure has been made in view of the above problems. The object is to provide a conveyance system, a conveyance method, and a conveyance program capable of reducing the possibility that an operator working around an unmanned carrier feels uneasy about the movement of the unmanned carrier.

Means for Solving the Problems

[0006] The transport system, transport method, and transport program described herein utilize one or more automated guided vehicles (AGVs) and a controller connected to sensors that generate distance point data for multiple distance points in an area where workers and AGVs may be present. The controller acquires distance point data from the sensors, generates difference data indicating the difference between the distance point data and reference data, identifies the position of workers within the area based on the difference data, and controls the movement of the AGVs based on their positions.

[0007] The controller may classify the distance measurement points related to the difference into one or more clusters based on the difference data, extract a specific cluster associated with the worker from among the clusters based on the characteristics of each cluster (height, width, or size), and then determine the location based on the specific cluster.

[0008] The controller may be composed of multiple pixels and generate a distance image in which the intensity of each pixel is associated with the distance to a distance measuring point located in the direction corresponding to the pixel.

[0009] The controller may generate a distance image by excluding distance measurement points located outside the rectangular area containing the distance measurement points within the area.

[0010] The controller may also perform image recognition on distance images to extract workers.

[0011] The controller may classify the distance measurement points related to the difference into one or more clusters based on the distance image.

[0012] The controller may calculate the rectangular region containing the cluster based on the distance image.

[0013] The controller may determine the location based on the cluster with the largest volume of rectangular parallelepiped region among the clusters.

[0014] The controller may calculate the proportion of the area of ​​each cluster in the total area on the depth image corresponding to multiple clusters, and determine the location based on the cluster with the largest proportion.

[0015] The controller may integrate multiple clusters relating to multiple rectangular parallelepiped regions when the distance between multiple rectangular parallelepiped regions is less than or equal to a predetermined distance.

[0016] The controller may integrate multiple clusters relating to multiple rectangular parallelepiped regions when the ratio of the volume of the overlapping portion of the multiple rectangular parallelepiped regions to the total volume occupied by the multiple rectangular parallelepiped regions is greater than or equal to a predetermined threshold.

[0017] The reference data may also be distance measurement point data for an area where no automated guided vehicles or workers are present.

[0018] Multiple sensors may be positioned at different locations and observe the distance measurement point from different directions.

[0019] The controller may also be one that causes the automated guided vehicle to follow the worker's position.

[0020] The controller may also be used to move the automated guided vehicle away from the worker's position. [Effects of the Invention]

[0021] According to this disclosure, it is possible to provide a transport system, a transport method, and a transport program that can reduce the likelihood of workers performing tasks around an automated guided vehicle feeling anxious about the movement of the automated guided vehicle. [Brief explanation of the drawing]

[0022] [Figure 1] This is a block diagram showing the configuration of the transport system related to this disclosure. [Figure 2] This flowchart shows the processing procedure of the transport system related to this disclosure. [Figure 3] This is a diagram showing an example of an area where an operator and an automated guided vehicle may exist. [Figure 4A] This is a diagram showing an example of the distribution of ranging points. [Figure 4B] This is a diagram showing an example of the distribution of ranging points in reference data. [Figure 5] This is a diagram showing an example of the distribution of ranging points related to differences and clusters. [Figure 6] This is a diagram showing an example of feature amounts for each height of clusters related to an operator.

Embodiments for Carrying out the Invention

[0023] Hereinafter, several exemplary embodiments will be described with reference to the drawings. In the drawings, the same reference numerals are assigned to common parts, and duplicate descriptions are omitted.

[0024] [Configuration of the Conveying System] FIG. 1 is a block diagram showing the configuration of a conveying system according to the present disclosure. The conveying system 1 includes automated guided vehicles AM1 and AM2, sensors SC1 to SC3, and a controller 10. The conveying system 1 may include a communication unit 20.

[0025] For example, the controller 10 is connected so as to be able to communicate wirelessly with the automated guided vehicles AM1 and AM2. The controller 10 may be connected so as to be able to communicate with a communication device (not shown) mounted on each of the automated guided vehicles AM1 and AM2 via the communication unit 20. Further, the controller 10 may be connected to the sensors SC1 to SC3 so as to be able to communicate wirelessly or by wire.

[0026] Figure 3 shows an example of an area where workers and automated guided vehicles (AGVs) may be present. Figure 3 shows a side view of the area (workspace) where workers and AGVs may be present, viewed in a direction parallel to the floor level (FL). One or more workers (PS) may be present in the area where AGVs AM1 and AM2 travel. This assumes a situation where AGVs AM1 and AM2 and workers (PS) stay in the same area and work together.

[0027] For example, a workbench TB may exist within the area, and the automated guided vehicles AM1 and AM2 may place the cargo LD they have transported onto the workbench TB, where a worker PS may begin working on the cargo LD. Alternatively, after the worker PS has completed work on the cargo LD on the workbench TB, the automated guided vehicles AM1 and AM2 may transport the completed cargo LD to another destination. The work performed collaboratively by the automated guided vehicles AM1 and AM2 and the worker PS is not limited to the examples given here. For example, the workbench TB may not be necessary for the work to be performed.

[0028] The transport system 1 may include one or more automated guided vehicles (AGVs). Figure 3 shows the transport system 1 equipped with two AGVs AM1 and AM2, but is not limited to this configuration. The transport system 1 may also include one or more sensors. Figure 3 shows the transport system 1 equipped with three sensors SC1 to SC3, but is not limited to this configuration.

[0029] Automated guided vehicles (AGVs) AM1 and AM2 transport cargo LD within a predetermined area, for example. The movement of AGVs AM1 and AM2 is controlled by controller 10. Alternatively, AGVs AM1 and AM2 may determine their travel path based on information about the worker's position transmitted from controller 10.

[0030] For example, when the automated guided vehicles AM1 and AM2 travel on the floor surface FL, they may travel along a route that keeps them away from the worker PS. More specifically, the travel routes of the automated guided vehicles AM1 and AM2 may be set to maintain a distance of a predetermined distance or more from the position of the worker PS.

[0031] Furthermore, when the worker PS moves, the automated guided vehicles AM1 and AM2 may travel along a route that follows the worker PS. More specifically, the travel routes of the automated guided vehicles AM1 and AM2 may be set to maintain a distance of less than a predetermined distance from the position of the worker PS.

[0032] The travel routes set for the automated guided vehicles AM1 and AM2 may be selected by a transport management system (not shown) from among routes connecting the starting point to the destination point for transporting the cargo LD. Alternatively, the selected travel route may be modified to set a route that moves away from the operator PS, or a route that follows the operator PS. Various technologies can be applied to setting the travel route.

[0033] Sensors SC1 to SC3 generate distance point data for multiple distance points in an area where worker PS and automated guided vehicles AM1 and AM2 may be present. For example, sensors SC1 to SC3 may be LiDAR (Light Detection and Ranging), optical cameras, or ultrasonic sensors. For example, sensors SC1 to SC3 may be installed on the ceiling WL. Alternatively, sensors SC1 to SC3 may be fixed to the floor FL, or to furniture installed on the floor FL.

[0034] In addition, multiple sensors SC1 to SC3 may be placed in different positions, observing a single distance measurement point from different directions. The presence of multiple sensors SC1 to SC3 observing a single distance measurement point from different directions helps to prevent the distance measurement point from being obscured by the sensors' blind spots due to workers PS or automated guided vehicles AM1 and AM2 moving within the area.

[0035] Distance point data refers to data containing positional information about distance points located on the surface of objects within an area. Distance point data may also include data on the distance to the distance point and the direction of the distance point as seen from sensors SC1 to SC3. Alternatively, distance point data may also include three-dimensional positional information of the distance point as viewed in a coordinate system fixed to the floor surface FL.

[0036] The distance measurement point data acquired by sensors SC1 to SC3 may be transmitted to the communication unit 20 wirelessly or via a wired connection and processed by the controller 10.

[0037] Controller 10 is a general-purpose computer equipped with a CPU (Central Processing Unit), memory, and input / output units. A computer program (transportation program) for controlling the transport system 1 is installed in Controller 10. By executing the computer program, Controller 10 functions as one of the multiple information processing circuits (11, 13, 15, 17, 19) provided by the transport system 1.

[0038] This disclosure provides an example of implementing multiple information processing circuits (11, 13, 15, 17, 19) using software. However, it is also possible to configure the information processing circuits (11, 13, 15, 17, 19) by preparing dedicated hardware for each of the information processing operations described below. Alternatively, the multiple information processing circuits (11, 13, 15, 17, 19) may be configured using separate hardware.

[0039] As shown in Figure 1, the controller 10 comprises multiple information processing circuits (11, 13, 15, 17, 19), including a difference data acquisition unit 11, a cluster classification unit 13, a distance image generation unit 15, a location identification unit 17, and an execution instruction unit 19.

[0040] The difference data acquisition unit 11 acquires distance point data from sensors SC1 to SC3 and generates difference data showing the difference between the distance point data and reference data. Here, "reference data" refers to, for example, distance point data for an area where automated guided vehicles AM1 and AM2 and worker PS are not present. For example, "reference data" may be data that has been registered in advance.

[0041] The relationship between distance measurement point data and reference data will be explained using Figures 4A and 4B. Figure 4A shows an example of the distribution of distance measurement points. Figure 4B shows an example of the distribution of distance measurement points in the reference data.

[0042] In Figures 4A and 4B, the distribution of distance measurement points is schematically shown by black circles. Multiple distance measurement points are located on the surfaces of the floor FL, the automated guided vehicle AM1, the workbench TB, the cargo LD, and the worker PS, and represent their shapes. In Figure 4A, the automated guided vehicle AM1, cargo LD, and worker PS are present, whereas in Figure 4B, the automated guided vehicle AM1, cargo LD, and worker PS are absent. Therefore, in Figure 4B, only distance measurement points related to the floor FL and workbench TB are present.

[0043] As shown in Figure 4B, the reference data includes only the distance measurement points related to objects that do not move within the area. Therefore, by comparing the distance measurement point data related to objects that move within the area, as shown in Figure 4A, with the reference data, as shown in Figure 4A, and obtaining the difference, it is possible to detect objects that move within the area.

[0044] Figure 5 shows an example of the distribution and clusters of distance measurement points related to the difference. In Figure 5, the distribution of distance measurement points related to the difference is obtained by subtracting the reference data shown in Figure 4B from the distance measurement point data shown in Figure 4A. It also shows clusters CL1 to CL3 obtained when the distance measurement points related to the difference are classified into clusters.

[0045] As shown in Figure 5, the difference data acquisition unit 11 subtracts pre-registered reference data from the distance point data acquired from sensors SC1 to SC3 to generate difference data that shows the difference between the distance point data and the reference data. In Figure 5, the distance points related to the difference data are shown as distance points located on the surface of the automated guided vehicle AM1, distance points located on the surface of the cargo LD, and distance points located on the surface of the worker PS.

[0046] The cluster classification unit 13 classifies the distance measurement points related to the difference into one or more clusters based on the difference data. For example, as shown in Figure 5, the cluster classification unit 13 may classify distance measurement points separated by a distance less than a predetermined threshold as the same cluster. Alternatively, the cluster classification unit 13 may classify the distance measurement points related to the difference into one or more clusters based on a distance image described later.

[0047] Figure 5 shows how the distance measurement points located on the surface of the automated guided vehicle (AGV) AM1 are classified as cluster CL1. It also shows how the distance measurement points located on the surface of the cargo LD are classified as cluster CL2. Furthermore, it shows how the distance measurement points located on the surface of the worker PS are classified as cluster CL3.

[0048] The cluster classification unit 13 may calculate a rectangular prism region that encloses each cluster. Here, the length, width, and height of the rectangular prism region may be calculated to be equal to the length, width, and height of the cluster that encloses it, respectively. In other words, the rectangular prism region calculated by the cluster classification unit 13 may be sized to just fit the cluster.

[0049] The cluster classification unit 13 may calculate the rectangular region based on the distance measurement point data, or it may calculate the rectangular region based on the difference data. Alternatively, the cluster classification unit 13 may calculate the rectangular region based on the distance image described later.

[0050] In addition, the cluster classification unit 13 may integrate multiple clusters relating to multiple rectangular parallelepiped regions when the distance between multiple rectangular parallelepiped regions is less than or equal to a predetermined distance. This integrates smaller clusters and reduces the computational load related to the processing of distance measurement points. The integration of multiple rectangular parallelepiped regions may be performed by selecting two rectangular parallelepiped regions from among the multiple regions and sequentially deciding whether or not to integrate them. Furthermore, the distance between rectangular parallelepiped regions may be the distance between adjacent vertices or the distance between the centroids of the rectangular parallelepiped regions.

[0051] Furthermore, the cluster classification unit 13 may integrate multiple clusters relating to multiple rectangular parallelepiped regions if the ratio of the volume of the overlapping portion of the multiple rectangular parallelepiped regions to the total volume occupied by the multiple rectangular parallelepiped regions is greater than or equal to a predetermined threshold. In other words, if multiple rectangular parallelepiped regions overlap, the overlapping multiple rectangular parallelepiped regions may be integrated if the degree of overlap exceeds a predetermined standard.

[0052] Furthermore, the cluster classification unit 13 may extract specific clusters associated with workers from among the clusters obtained through classification, based on the characteristics of each cluster height.

[0053] The extraction of specific clusters will be explained using Figure 6. Figure 6 shows an example of feature quantities for each height of clusters related to workers. For example, the cluster classification unit 13 may divide the distance measurement points that constitute the cluster by height and calculate feature quantities for the width and size of the cluster related to that division based on the distance measurement points included in that division. The cluster classification unit 13 may then determine whether the pattern of feature quantities for each division matches the pattern associated with the worker. For example, when using LiDAR as sensors SC1 to SC3, the value of the Z component (distance from the distance measurement point to the floor surface FL) in the coordinate data of the distance measurement point may be calculated as the height.

[0054] As an example, the cluster classification unit 13 may calculate the horizontal spread of multiple distance measurement points for each category as a feature. For example, if the target cluster is associated with a worker, the spread of distance measurement points corresponding to the neck area is smaller than the spread of distance measurement points corresponding to the head or torso area. Also, the spread of distance measurement points corresponding to the feet area is smaller than the spread of distance measurement points corresponding to the head or torso area.

[0055] Therefore, as shown on the right side of Figure 6, the feature quantity is small at positions close to the floor (FL), and increases at body height. Furthermore, the feature quantity tends to decrease at neck height and increase at head height. Thus, the cluster classification unit 13 may pre-register patterns associated with workers and extract clusters that match those patterns as specific clusters.

[0056] The distance image generation unit 15 generates a distance image composed of multiple pixels, where the intensity of each pixel is associated with the distance to the distance measurement point located in the direction corresponding to the pixel. For example, the distance image generation unit 15 selects one sensor of interest and generates a distance image based on the distance from the sensor of interest to the distance measurement point and the direction of the distance measurement point as seen from the sensor of interest.

[0057] According to the depth image, distance measurement points that are close in direction from the sensor are placed in adjacent pixels on the depth image, and the intensity of pixels related to distance measurement points that are at a similar distance from the sensor is set to a similar level. Therefore, if there is a group of pixels with similar intensity on the depth image, it can be seen that the distance measurement points represented by the pixels constituting that group are close together. Thus, by using the depth image, clustering of distance measurement points can be performed with a low computational load.

[0058] Therefore, if a cluster of pixels with similar intensity exists on the depth image, the cluster classification unit 13 may classify the distance measurement points represented by the pixels constituting that cluster into the same cluster.

[0059] In addition, the distance image generation unit 15 may generate the distance image by excluding distance measurement points located outside the rectangular parallelepiped region calculated by the cluster classification unit 13, in order to reduce the computational load when generating the distance image.

[0060] The location identification unit 17 identifies the worker's location within the area based on the differential data.

[0061] For example, the location identification unit 17 may identify the worker's location based on the extracted specific cluster. Specifically, the location identification unit 17 may use the location of the specific cluster as the worker's location. The location of the specific cluster may be, for example, the centroid of the specific cluster, or the location where a straight line drawn vertically from the center of the specific cluster intersects the floor surface FL.

[0062] Alternatively, the location identification unit 17 may identify the worker's location based on the cluster with the largest rectangular parallelepiped volume among the clusters obtained by the cluster classification unit 13. The location identification unit 17 may also determine the worker's location as the location of the cluster with the largest rectangular parallelepiped volume.

[0063] Furthermore, the position identification unit 17 may calculate the ratio of the area of ​​each cluster to the total area on the depth image corresponding to multiple clusters, and identify the worker's position based on the cluster with the largest ratio. Alternatively, the position identification unit 17 may determine the worker's position as the location of the cluster with the largest area ratio on the depth image.

[0064] Furthermore, the position identification unit 17 may extract workers by performing image recognition on the depth image. More specifically, it performs image recognition on the depth image to extract the region corresponding to the worker in the depth image. Here, various techniques can be applied to extract the region corresponding to the worker using image recognition. For example, object detection and region segmentation can be used as object recognition methods.

[0065] The execution instruction unit 19 controls the movement of the automated guided vehicles AM1 and AM2 based on the location of the identified worker. For example, the execution instruction unit 19 may set a travel path that moves away from the worker PS. More specifically, the travel paths of the automated guided vehicles AM1 and AM2 may be set to maintain a distance of a predetermined distance or more from the location of the worker PS.

[0066] Furthermore, the execution instruction unit 19 may set a travel path that follows the worker PS. More specifically, the travel paths of the automated guided vehicles AM1 and AM2 may be set so as to maintain a distance of less than a predetermined distance from the position of the worker PS.

[0067] The execution instruction unit 19 instructs the automated guided vehicles AM1 and AM2 to travel along the set travel path. Upon receiving the instruction, the automated guided vehicles travel along the set travel path. In doing so, the automated guided vehicles travel in a manner that satisfies the set travel conditions.

[0068] In addition, the execution instruction unit 19 may control the automated guided vehicles AM1 and AM2 to transfer the cargo LD when they reach a predetermined destination. Alternatively, the execution instruction unit 19 may control a conveyor or transfer device (not shown) to transfer the cargo LD between the automated guided vehicles AM1 and AM2.

[0069] [Processing procedure for the transport system] Figure 2 is a flowchart showing the processing procedure of the transport system according to this disclosure. The processing shown in Figure 2 may be performed repeatedly at a predetermined interval.

[0070] In step S101, the difference data acquisition unit 11 acquires distance measurement point data from sensors SC1 to SC3.

[0071] In step S103, the difference data acquisition unit 11 generates difference data that shows the difference between the distance measurement point data and the reference data.

[0072] In step S105, the cluster classification unit 13 classifies the distance measurement points related to the difference into one or more clusters.

[0073] In step S107, the location identification unit 17 identifies the worker's location within the area based on the difference data.

[0074] In step S109, the execution instruction unit 19 sets the travel path of the automated guided vehicle based on the worker's position.

[0075] In step S111, the execution instruction unit 19 instructs the automated guided vehicle to travel along the designated route.

[0076] [Effects of the Embodiment] As described in detail above, the transport system, transport method, and transport program relating to this disclosure use one or more automated guided vehicles (AGVs) and a controller connected to sensors that generate distance point data for multiple distance points in an area where workers and AGVs may be present. The controller acquires distance point data from the sensors, generates difference data showing the difference between the distance point data and reference data, identifies the position of workers in the area based on the difference data, and controls the movement of the AGVs based on their positions.

[0077] This reduces the likelihood of workers feeling anxious about the movement of automated guided vehicles (AGVs) while working around them. It also allows AGVs and workers to stay in the same area and collaborate, improving labor productivity. Furthermore, the AGV's movement speed can be increased as long as it does not approach workers, thereby increasing work efficiency.

[0078] The controller may classify the distance measurement points related to the difference into one or more clusters based on the difference data, extract a specific cluster associated with the worker from among the clusters based on the characteristics of each cluster (height, width, or size), and then determine the location based on the specific cluster. This can improve the accuracy of worker identification. In addition, it can reliably increase the movement speed of the automated guided vehicle, thereby improving work efficiency.

[0079] The controller may be composed of multiple pixels and generate a distance image in which the intensity of each pixel is associated with the distance to a distance measurement point located in the direction corresponding to the pixel. This makes it possible to determine the worker's position based on the distance image, instead of directly processing the distance measurement point data.

[0080] The controller may generate a distance image by excluding distance measurement points located outside the rectangular area containing the distance measurement points within the area. This reduces the computational load associated with processing distance measurement points when generating the distance image.

[0081] The controller may perform image recognition on the depth image to extract workers. By applying various techniques known in the field of image recognition to the depth image, worker extraction can be performed with greater accuracy. Furthermore, the computational load can be reduced compared to directly processing the distance point data to determine the worker's position.

[0082] The controller may classify the distance measurement points related to the difference into one or more clusters based on the distance image. This reduces the computational load compared to directly processing the distance measurement point data and classifying it into clusters.

[0083] The controller may calculate a rectangular region containing the cluster based on the distance image. This allows for limiting the range measurement points to be processed.

[0084] The controller may determine the location based on the cluster with the largest volume of rectangular region among the clusters. This allows for easy identification of the worker's location with low computational cost.

[0085] The controller may calculate the percentage of the area of ​​each cluster in the total area on a depth image corresponding to multiple clusters, and determine the location based on the cluster with the largest percentage. This allows for easy identification of the worker's location with low computational cost.

[0086] The controller may integrate multiple clusters relating to multiple rectangular parallelepiped regions when the distance between multiple rectangular parallelepiped regions is less than or equal to a predetermined distance. This integrates smaller clusters and reduces the computational load associated with processing distance measurement points.

[0087] The controller may integrate multiple clusters relating to multiple rectangular parallelepiped regions when the ratio of the volume of the overlapping portion of the multiple rectangular parallelepiped regions to the total volume occupied by the multiple rectangular parallelepiped regions is greater than or equal to a predetermined threshold. This allows for the integration of overlapping clusters to a certain extent, thereby reducing the computational load related to the processing of distance measurement points.

[0088] The reference data may be distance measurement point data for an area where no automated guided vehicles or workers are present. This allows for the detection of objects moving within the area by comparing the distance measurement point data with the reference data.

[0089] Multiple sensors may be positioned at different locations, and the distance measurement points may be observed from different directions. This helps to suppress blind spots caused by moving objects within the area. As a result, the worker's position can be determined more reliably.

[0090] The controller may also cause the automated guided vehicle (AGV) to follow the worker's position. This allows the AGV and the worker to stay in the same area and work together, thereby improving labor productivity.

[0091] The controller may also be designed to keep the automated guided vehicle (AGV) away from the worker's position. This can prevent the AGV from approaching the worker's position and reduce the likelihood of the worker feeling uneasy about the AGV's movement.

[0092] Each of the functions described in the embodiments above may be implemented by one or more processing circuits. These processing circuits may include programmed processors, electrical circuits, and other devices such as application-specific integrated circuits (ASICs), or circuit components arranged to perform the described functions.

[0093] According to this disclosure, it is possible to reduce the likelihood of workers feeling anxious about the movement of automated guided vehicles (AGVs) while working around them. As a result, the working environment for workers can be improved, and labor productivity can be increased. Therefore, for example, it can contribute to United Nations Sustainable Development Goal (SDG) 8, "Promote inclusive and sustainable economic growth and full and productive employment and decent work for all."

[0094] Although several embodiments have been described, it is possible to modify or transform the embodiments based on the above disclosure. All components of the above embodiments, and all features described in the claims, may be taken individually and combined, provided that they do not conflict with each other. [Explanation of symbols]

[0095] 1. Conveying System 10 Controllers 11. Differential Data Acquisition Unit 13. Cluster Classification Section 15 Distance Image Generation Unit 17 Location identification part 19 Execution Instruction Unit 20 Communications Department AM1, AM2 Automated Guided Vehicles CL1~CL3 cluster FL floor surface LD luggage PS worker SC1~SC3 Sensors TB workbench WL Ceiling

Claims

1. One or more automated guided vehicles, A sensor that generates distance point data for multiple distance measurement points in an area where workers and the automated guided vehicle may be present, Controller and A transport system equipped with, The aforementioned controller, From the sensor, acquire the distance measurement point data, Difference data is generated that shows the difference between the distance measurement point data and the reference data. Based on the aforementioned difference data, the location of the worker within the area is identified. Based on the aforementioned position, control the movement of the automated guided vehicle. Conveyor system.

2. The aforementioned controller, Based on the difference data, the distance measurement points related to the difference are classified into one or more clusters, Based on the characteristics of each of the height, width, or size of the aforementioned clusters, a specific cluster associated with the worker is extracted from among the aforementioned clusters. The location is identified based on the specified cluster. The transport system according to claim 1.

3. The transport system according to claim 1 or 2, wherein the controller is composed of a plurality of pixels and generates a distance image in which the intensity of each pixel is associated with the distance to the distance measuring point located in the direction corresponding to the pixel.

4. The transport system according to claim 3, wherein the controller generates the distance image by excluding the distance measurement point located outside the rectangular parallelepiped region including the distance measurement point within the area.

5. The transport system according to claim 3, wherein the controller performs image recognition on the distance image to extract the worker.

6. The transport system according to claim 3, wherein the controller classifies the distance measurement points relating to the difference into one or more clusters based on the distance image.

7. The transport system according to claim 6, wherein the controller calculates a rectangular parallelepiped region containing the cluster based on the distance image.

8. The aforementioned controller, The position is determined based on the cluster with the largest volume of the rectangular parallelepiped region among the aforementioned clusters. The transport system according to claim 7.

9. The aforementioned controller, The ratio of the area of ​​each cluster to the total area on the distance image corresponding to multiple clusters is calculated. Based on the cluster with the largest proportion, the location is determined. The transport system according to claim 7.

10. The transport system according to claim 7, wherein the controller integrates the clusters relating to the plurality of rectangular parallelepiped regions when the ratio of the volume of the overlapping portion of the plurality of rectangular parallelepiped regions to the total volume occupied by the plurality of rectangular parallelepiped regions is greater than or equal to a predetermined threshold.

11. A transport method for controlling one or more automated guided vehicles and a controller connected to sensors that generate distance point data relating to multiple distance measurement points in an area where workers and the automated guided vehicles may be present, The aforementioned controller, From the sensor, acquire the distance measurement point data, Difference data is generated that shows the difference between the distance measurement point data and the reference data. Based on the aforementioned difference data, the location of the worker within the area is identified. Based on the aforementioned position, control the movement of the automated guided vehicle. Method of transport.

12. A transport program to be executed in a controller connected to one or more automated guided vehicles and sensors that generate distance point data relating to multiple distance measurement points in an area where workers and the automated guided vehicles may be present, The aforementioned controller, The steps include acquiring the distance measurement point data from the sensor, The steps include generating difference data that shows the difference between the distance measurement point data and the reference data, Based on the difference data, the steps include identifying the location of the worker within the area, A step of controlling the movement of the automated guided vehicle based on the position, A transport program that includes this.

Images