Storage and retrieval apparatus, and method for loading a storage and retrieval apparatus

Abstract

The invention relates to a storage and retrieval apparatus (1) for transporting a load (2). The storage and retrieval apparatus (1) comprises a loading region (3) designed to receive a load (2), a sensor device (4) configured to determine an actual state of the load (2), and a control unit (5) configured to compare the actual state of the load (2) with a desired state in order to determine discrepancy information, the control unit (5) being furthermore configured to determine an operating state of the storage and retrieval apparatus (1) on the basis of the discrepancy information. The invention further relates to a method for loading a storage and retrieval apparatus (1) with a load (2).

Description

[0001] Storage and retrieval machine, method for loading a storage and retrieval machine

[0002] The present invention relates to a storage and retrieval machine for transporting a load and to a method for loading a storage and retrieval machine with a load.

[0003] In the prior art, storage and retrieval machines are known that can transport a load into a racking system. To increase efficiency, the operation of such a storage and retrieval machine is largely automated. A variety of sensors are used to enable this automated operation. For example, a storage and retrieval machine has a loading area into which a load to be transported is inserted. Before a movement begins, light barriers are used to check whether any part of the load or other objects or people are present in a restricted area.

[0004] If an object or person is detected in such a restricted area, the movement of the storage and retrieval machine is immediately stopped and an error message is displayed. To rectify the error, an operator must check the restricted area and, if necessary, clear it so that the sensors no longer register an error.

[0005] For safety reasons, numerous such restricted areas are established. However, it frequently occurs that a sensor detects an object within the restricted area, thereby stopping the operation of the storage and retrieval machine unnecessarily. This results in interrupted operation of the storage and retrieval machine and requires manual intervention by an operator who must monitor the restricted area. Therefore, it is an object of the present invention to provide a storage and retrieval machine and an associated method that can reduce or eliminate unnecessary downtime and excessive manual labor.

[0006] The above problem is solved with a storage and retrieval machine having the features of claim 1 and with a method for loading a storage and retrieval machine having the features of claim 20.

[0007] According to one aspect of the present invention, a storage and retrieval machine is provided for transporting a load. The storage and retrieval machine can include a loading area configured to receive a load. Furthermore, the storage and retrieval machine can include a sensor device configured to determine the actual state of the load. The storage and retrieval machine can include a control unit configured to compare the actual state of the load with a target state in order to determine deviation information. The control unit can further be configured to determine an operating state of the storage and retrieval machine based on the deviation information.

[0008] Compared to the prior art, the present invention offers the advantage of reduced downtime for the storage and retrieval machine. In other words, unnecessary errors can be minimized, thus optimizing operation. This is achieved by having the sensor device determine the actual state and compare it to a target state. In other words, the present invention allows for the consideration of additional boundary conditions to determine whether an error is present. Furthermore, this comparison enables an assessment of whether an existing error is serious or not. More precisely, the control unit can compare the actual state with a target state. Based on this comparison, deviation information can be determined. This deviation information can then be used to determine the operating status of the storage and retrieval machine.By comparing the actual state with the target state, any deviation between the two states can be determined. However, if such a deviation is detected, an immediate operational stop cannot be triggered. Instead, deviation information can be determined based on the comparison, which can then be used to define the operating state of the storage and retrieval machine. For example, it can be detected that the actual state deviates from the target state, but that this is harmless, and therefore no error needs to be triggered. Furthermore, determining the deviation information allows for the consideration of additional parameters or boundary conditions. For instance, when an object is detected in a restricted area, the size and / or type of the object in the restricted area can be determined, which is then also included in the deviation information.For example, a piece of packaging material may protrude into the restricted area, which is harmless to the smooth operation of the storage and retrieval machine. In a prior art system, such a piece of packaging would directly trigger an error and thus a shutdown of the machine. With the storage and retrieval machine according to one aspect of the present invention, it is possible to assess whether a piece of packaging material protruding into a restricted area is detrimental to operation or not. This allows downtime of the storage and retrieval machine to be significantly reduced. Consequently, the efficiency of the storage and retrieval machine can be increased.

[0009] A storage and retrieval machine (SRM) is a device that can place a load or cargo into a storage area. A SRM cannot be an automated guided vehicle (AGV). For example, an SRM might be used in a high-bay warehouse to store and retrieve cargo. The SRM can only travel along predetermined paths, such as a rail. In other words, the SRM cannot move freely within the space. It cannot be an autonomous vehicle, but rather a SRM bound to a transport system. This significantly simplifies the SRM's control. It is, in fact, tied to a transport mechanism. Furthermore, the SRM can have a mast along which a load can be transported vertically (i.e., in a vertical direction).The mast can extend vertically, essentially corresponding to the height of the rack being serviced. Such a storage and retrieval machine (SRM) can be loaded by a conveying device, which can include a conveyor belt or similar equipment. The SRM preferably transports the load vertically. However, it is also possible for the SRM to be additionally or alternatively capable of horizontal movement. Optionally, the SRM can travel horizontally on a rail. The SRM can be designed to move the load at least 3 meters or more vertically. This allows the SRM to be used in a high-bay warehouse. A load can be a single item or a multitude of individual items. The load can be stacked on a pallet or consist of multiple pallets.The load can comprise a variety of individual goods or items, wrapped, for example, in film or similar material. It is also conceivable that the load is contained within a container, such as a box. The loading area, or loading platform, is the section of the storage and retrieval machine (SRM) that can accommodate a load. The loading area can also be referred to as a platform. Typically, the loading area is a flat surface on which the load can be placed. The loading area can be physically repositioned during operation of the SRM. The loading area can be repositioned relative to the mast. In other words, the loading area can be arranged at different positions along the mast. This allows a load to be placed or picked up at various heights within a high-bay warehouse.The loading area can also include a securing system designed to prevent cargo from falling off the loading area. Such a system can, for example, include at least one pin that can be moved between a loading position and a transport position. In the loading position, cargo can be easily placed into the loading area, whereas in the transport position, the cargo cannot easily leave the loading area. In other words, the pin can be extended in the transport position and retracted in the loading position. Thus, it can prevent the cargo from slipping or falling off during transport. Furthermore, the loading area can include at least one transport section designed to move the cargo to a desired position within the loading area.The transport device can be a belt or chain system that receives a load and moves it to the desired position in the loading area. Alternatively, the transport device can be a link conveyor. The load can be picked up into the loading area partly by the feeder and partly by the loading area itself. The storage and retrieval machine can have a sensor device, preferably located within the loading area. The sensor device can be designed to analyze the load. In other words, the sensor device can be designed to collect information about the load. The sensor device can be located directly adjacent to the loading area and / or the storage and retrieval machine. This ensures that the sensor device can easily analyze the load picked up in the loading area.The sensor device can comprise a variety of sensors. The sensor device can determine the current state of the charge. The current state of the charge can be determined at the moment when the charge is positioned in the desired location on the charging area. The control unit can be a computer-like device designed to receive, process, and output information. The received information can be input information, and the output information can be control commands or output information. The control unit can also include a comparator unit designed to compare the current state with a target state. Furthermore, a target state can be a state that the charge ideally possesses. In other words, the target state can be a state in which the charge is free of defects and damage.Furthermore, the target state can also contain charge information. For example, the target state can be indicative of a specific type of charge. By comparing the target state with the actual state, it can be detected whether the correct charge is present. The control unit can then be designed to determine deviation information based on this comparison. The deviation information can indicate whether a deviation exists or not. If a deviation exists, the deviation information can indicate the type and extent of the deviation. For example, if the external dimensions of the charge change, the extent of this change can be determined.Thus, for example, in the case of only a minor change, such as one caused by a defect in the outer packaging of the load, it can be determined, based on the deviation information, that operation is still possible. The control unit can determine the operating state of the storage and retrieval machine based on this deviation information. In other words, the control unit can be designed to evaluate the deviation information and determine the operating state of the storage and retrieval machine accordingly. This evaluation can be performed, for example, based on tolerance values. The operating state can indicate, for instance, whether operation is possible without problems, whether it is possible with limitations, or whether it is not possible at all. For example, a specific deviation information might indicate that limited operation is possible.Restricted operation can be characterized by the storage and retrieval machine operating at a lower speed and / or acceleration compared to normal operation. Therefore, it is not immediately necessary to shut down or block the entire machine and call an operator; the load can still be transported. This leads to more efficient operation of the storage and retrieval machine with fewer errors.

[0010] Preferably, the target state is a target state of the load. In other words, the control unit can receive information about the load to be transported before the load physically arrives at the storage and retrieval machine. The target state can, for example, be indicative of an external dimension of the load. Furthermore, the target state can be indicative of weight or type of load. It is also conceivable that all load information is included in the target state. This allows the control unit to comprehensively determine whether there are any deviations between the target state and the actual state measured at the storage and retrieval machine. This enables seamless monitoring of the load. Furthermore, more precise conclusions can be drawn about how to optimize the operation of the storage and retrieval machine.

[0011] Preferably, the sensor device comprises at least two sensors. This makes it particularly advantageous to determine the current state when two or more individual charges are applied to the charging area. For example, the two sensors can be arranged opposite each other on the charging area. This prevents a charge, or part of a charge, from being located in the sensor shadow of one of the sensors. Thus, the current state can be satisfactorily determined even when a large number of individual charges are arranged on the charging area.

[0012] Preferably, the sensor device comprises at least one LiDAR sensor. The LiDAR sensor (short for Light Detection and Ranging) can be a device designed to optically measure distance and / or velocity. For example, the LiDAR sensor can emit a laser beam that scans the environment or a predefined area. As a result of such scanning, at least one point cloud can be generated. In other words, an object located within the scanning area of ​​the LiDAR sensor can be scanned so that the object's outline can be displayed. In other words, the sensor's environment can be represented. In particular, a before-and-after comparison can determine whether objects within the sensor's measurement area have changed, been added, or been removed. For example, the LiDAR sensor can detect the charge.The LiDAR sensor can therefore output at least a point cloud that is indicative of the charge's outline. This allows for the creation of a large number of measurement points, enabling a more comprehensive assessment of the charge's current state. Thus, it is not just a simple light barrier that outputs only a binary signal, but a sensor capable of providing significantly more information. Preferably, the sensor device includes at least one image sensor. The image sensor can be a device for capturing two-dimensional images of light electrically or mechanically. Preferably, the image sensor is a semiconductor-based image sensor. This allows for the provision of an optical system capable of detecting the charge's current state. For example, common image-based monitoring systems are conceivable for this purpose. In this case, the control unit can analyze the image data and output a corresponding current state.The image sensor can also take further boundary conditions into account, so that these too can be considered when determining the current state. In this case, the control unit can, for example, assign objects to a specific class based on a classification of the objects depicted by the image sensor. Thus, the control unit can distinguish, for instance, whether a person, or at least part of a person, is protruding into a restricted area, or merely a piece of packaging. This allows for the determination of current states that provide a more accurate description of reality. The sensor device can, for example, include a vision system.

[0013] Preferably, the sensor device is designed to determine the current state of the load while it is being moved into the loading area. Before the storage and retrieval machine (SRM) is operated, the load is fed to the SRM by a conveyor device. More precisely, the load is pushed onto the loading area or applied in some other way. During this application, the sensor device can operate continuously, allowing the load to be analyzed and detected even as it is moving. This provides more accurate information about the dimensions of the load. For example, the sensor device can be designed to take a measurement every 0.1 to 0.5 seconds. This generates a point cloud of the load, enabling precise conclusions to be drawn about its actual state. This allows for a more comprehensive representation of the load to be created with fewer sensors.Preferably, the operating state indicates whether the storage and retrieval machine (SRM) is operational or faulty. The specific operating state can indicate whether the SRM can move the load or not. For example, if a difference between the target and actual state has been detected, the deviation information can be used to determine whether the difference prevents the SRM from operating. The result of this assessment can then be displayed as the operating state. The operating state can encompass a variety of states. If unrestricted operation is possible, the operating state can indicate that the SRM can be operated normally. However, if, for example, a person is in the loading area, the operating state can indicate that operation is not possible and an error message must be displayed.Between these two extremes, however, there can be a multitude of intermediate operating states. For example, the loading area may contain a load that protrudes slightly beyond its boundaries. If this is within acceptable limits, the operating state can indicate that the storage and retrieval machine can be operated. Furthermore, the operating state can also indicate restricted operation if, for example, a particularly light or sensitive load is placed on the loading area. This may mean that the storage and retrieval machine can only be operated at reduced speed and / or with reduced acceleration. Thus, the operating state can accommodate a wide variety of real-world scenarios.

[0014] Preferably, the sensor device is designed to monitor the condition of the load during operation of the storage and retrieval machine. In other words, the sensor device can perform a scan not only before the storage and retrieval machine is operated, but also during operation. In other words, the sensor device can be arranged on the storage and retrieval machine in such a way that it can detect the load throughout the entire operation. In contrast, prior art photoelectric sensors are not capable of this. These photoelectric sensors can only react (i.e., detect or fail to detect) if, for example, a load falls from the loading area or protrudes into a restricted area during operation. The sensor device of the present embodiment, however, makes it possible to monitor the load even while the storage and retrieval machine is in motion.This allows, for example, early detection of when the load or part of the load starts to slip (i.e., move). This enables timely responses, such as reducing the speed of the storage and retrieval machine. As a result, operation adapted to the load can be achieved without damage or an excessive number of operational interruptions.

[0015] Preferably, the control unit is designed to determine the type of load based on its current state. The type of load can, for example, be indicative of further load information. This information can include, for instance, weight, sensitivity, air resistance, dimensions, or similar characteristics. This allows the system to verify that the load is indeed the correct one, as expected, and to generate an appropriate operating state for the storage and retrieval machine. For example, with a particularly large load exhibiting high air resistance, the storage and retrieval machine's movement speed can be reduced to prevent it from falling. This offers the advantage that the storage and retrieval machine can be designed for very high movement speeds in general, but can automatically adjust accordingly for sensitive loads that cannot withstand such speeds.Determining the type of charge can be done, for example, with a classifier. This allows the control unit to recognize all possible charge types and simply compare them to identify the specific type present. This can increase the detection speed.

[0016] Preferably, the control unit is designed to perform a classification based on the deviation information in order to determine the operating status of the storage and retrieval machine. In this case, the classification can include an assessment of whether a deviation is so serious that a stop of the storage and retrieval machine must be initiated. For example, part of the load may protrude into a restricted area. This could mean that the actual state differs from the target state. However, whether this part of the load protruding into the restricted area is so significant that the entire operation of the storage and retrieval machine must be stopped can be determined by the classification. For example, the control unit can be provided with a variety of deviation information that allows continued operation or restricted operation of the storage and retrieval machine.This initial information can then be used by the control unit to classify the available deviation information. This allows for a more nuanced assessment of whether or not operation of the storage and retrieval machine is possible. In other words, a simple yes / no decision cannot be made; instead, each case must be considered individually.

[0017] Preferably, the actual state comprises a representation of the charge. In other words, the actual state can comprise a digital representation of the charge. This representation of the charge can be a simplified depiction of the charge. In other words, information that exists in reality can be omitted.

[0018] This allows for a reduction in data volume, which can increase process speed. For example, in image-based imaging, the resolution can be reduced so that the information important for determining the operating state can be obtained without requiring excessive amounts of data. The image could, for instance, be a colorless or infrared image.

[0019] Preferably, the current state comprises a point cloud that characterizes the charge. A point cloud can consist of a multitude of points in three-dimensional space. Each point can be described by three coordinates (x, y, and z). Such a point cloud can preferably be acquired by at least one LiDAR sensor. Furthermore, it is also conceivable that such a point cloud can be generated by any other suitable sensor capable of scanning the charge. The individual points of the point cloud can characterize the surface of the charge. By connecting the individual points of the point cloud, a simplified outline structure of the charge can be generated. The point cloud contains a very small amount of data, allowing for very fast processing.

[0020] Preferably, the sensor device comprises a plurality of sensors, and the control unit is preferably configured to determine the current state of the load based on the sensor output and the position of each sensor. In other words, a plurality of individual sensors, encompassed by the sensor device, can be arranged on the storage and retrieval machine to detect the load from at least two sides. This allows for a more precise determination of the current state. For example, each sensor can be assigned an area in space that it scans. This allows the operation of the large sensor to be designed efficiently, as it does not need to detect everything within its sensor area, but only a predetermined area. The control unit can be configured to evaluate these areas differently.For example, an area near the floor of the loading area might be assessed differently in the event of an occupancy than an area 1.50 m above the floor. In other words, an object protruding into the area at the bottom of the planning area might be considered more likely to be harmless compared to an object protruding from the upper edge of the load. The individual sensor areas can be fixed and not change over time. The planning area is preferably divided into a plurality of such areas. Preferably, the sensor device comprises at least three sensors. This allows the three-dimensional shape of a load to be detected. In a further embodiment, the sensor device comprises five sensors. This has led to good results in the comprehensive detection of the load. This allows the current state to be detected satisfactorily.

[0021] Preferably, the sensor device comprises a main sensor and at least one secondary sensor. The main sensor can, for example, provide sensor data used to classify the type of charge. The secondary sensor, on the other hand, can only detect the position of the charge. In other words, sensor outputs from the sensors of the sensor device can be processed differently. This can increase the efficiency of processing the sensor data. Furthermore, the secondary sensor can be designed more efficiently, since it only needs to determine the position of the charge. A more efficient design could mean that the secondary sensor has a smaller sensing range than the main sensor. For example, the main sensor could detect a large portion of the charge, whereas the secondary sensor could only detect a specific area. This can make the system more cost-effective to manufacture overall and increase the processing speed.

[0022] Preferably, the control unit is configured to perform a classification based on the sensor output of the main sensor to determine the type of charge. In this case, the main sensor can determine a larger amount of data than the secondary sensor. This larger amount of data can lead to a better classification. Furthermore, the main sensor can also be an imaging sensor that acquires image data. Preferably, the sensor device comprises a plurality of sensors, wherein at least one sensor is preferably configured to detect whether a charge is arranged in the charging area. Preferably, only one sensor is configured to detect whether a charge is arranged in the charging area. In other words, the charging area can include a sensor responsible for checking whether a charge is arranged in the charging area or is being applied to it.Thus, the sensor device can be activated by a single sensor when, for example, a charge is introduced into the charging area. This prevents the sensor device from continuously generating sensor data. Instead, the single sensor can trigger the sensor device to initiate a scan. This allows for efficient operation of the sensor device.

[0023] Preferably, the loading area is divided into a plurality of sections, with each sensor preferably being assigned to one of the plurality of sensors in a given section. The sections can be referred to as the areas described above. In other words, the loading area can be divided into similar or different areas, and each sensor can be assigned to one area. Each area can, for example, be a conical shape in space.

[0024] Preferably, each sensor is designed to assume whether a section is at least partially occupied by the load or not.

[0025] In other words, each sensor can be solely responsible for checking whether its assigned area or section is occupied by a charge or part of the charge. This allows the resolution of each sensor to be adjusted to detect the section or area to which it is assigned. The individual areas can then be combined and used to describe the current state. For example, the section or area can be larger than a known exclusion zone from the prior art. In fact, such an area can at least partially encompass the charge. Thus, not only a peripheral area of ​​the charging region can be analyzed and monitored, but a larger portion of the charging region, preferably the entire charging region. This simplifies the determination of the current state.

[0026] Preferably, the control unit includes a learning algorithm designed to determine the operating state of the storage and retrieval machine (SRM) based on deviation information. A learning algorithm can output operating states of the SRM for new deviation information based on known mappings. In other words, deviation information can be fed to the learning algorithm, which can then output an operating state of the SRM. For example, if a fault occurs, a user can assess (i.e., annotate) whether the fault is substantial and requires a stop of the SRM. For instance, if part of a package gets caught in a restricted area, this can trigger a stop of the SRM. A user who is called can then assess whether this stop was justified.For example, in this situation, where part of the packaging protrudes into a restricted area, it might be considered non-critical for the operation of the storage and retrieval machine. The user can then mark the error accordingly, which will prompt the learning algorithm to automatically check next time whether the object protruding into the restricted area is simply a part of the packaging. This is just one example of the many ways the learning algorithm can be trained during the operation of the storage and retrieval machine. In other words, the learning algorithm can have completed basic training, enabling it to determine the fundamental correlations between deviation information and the operating status. However, during operation or through targeted fine-tuning, the learning algorithm can be adapted to the individual requirements of a specific location or user.This allows the storage and retrieval machine to adapt to a wide variety of applications and to adjust itself individually to the products being handled.

[0027] Preferably, the storage and retrieval machine has an interface designed to receive user input indicating which operating state corresponds to a given deviation information. In other words, a user or operator can use the interface to provide the storage and retrieval machine, and thus the control unit, with information about which deviations allow continued operation or limited operation, and which deviation information requires an immediate stop of the storage and retrieval machine. This also makes it possible to make necessary fine adjustments to the storage and retrieval machine manually on a daily basis.

[0028] According to a further aspect of the present invention, a method for loading a storage and retrieval machine with a load is provided. The method can include providing a load. The method can include transferring the load to a loading area of ​​the storage and retrieval machine. The method can include determining the actual state of the load. The method can include comparing the actual state with a target state. The method can include determining an operating state of the storage and retrieval machine based on the comparison of the actual state with the target state. Preferably, the actual state is determined during the transfer. In other words, the actual state of the load can be determined during the transfer of the load from a feeder element to the loading area. This allows the load to be moved relative to the sensor device, enabling comprehensive detection of the load.

[0029] According to a further aspect of the present invention, a method for training a learning algorithm to determine an operating state of a storage and retrieval machine is provided. The training method comprises providing training input data, wherein the training input data is indicative of at least one actual state of the load. Furthermore, the method may comprise providing training output data, wherein the training output data is indicative of at least one operating state of the storage and retrieval machine. The method may include training the learning algorithm based on the training input and the training output data. Preferably, the above method trains a learning algorithm that is used in one of the above embodiments.

[0030] According to a further aspect, the present invention relates to the use of the storage and retrieval machine according to one of the above embodiments in a rack storage system. Furthermore, the invention provides for the use of a method according to one of the above embodiments in conjunction with a storage and retrieval machine.

[0031] Individual features and embodiments can be combined to form new embodiments. Features and advantages mentioned in connection with the features or embodiments also apply analogously to the new embodiments. Features and advantages mentioned in connection with the device also apply analogously to the methods, and vice versa.

[0032] The present invention is described in detail below with reference to the accompanying figures. Figure 1 is a schematic and perspective view of a storage and retrieval machine of an embodiment of the present invention.

[0033] Figure 2 is a schematic and perspective view of a loading area according to an embodiment of the present invention.

[0034] Figure 3 is a schematic and perspective view of a loading area with a load according to an embodiment of the present invention.

[0035] Figures 4A to 4C are schematic views of a loading area according to an embodiment of the present invention.

[0036] Figure 5 is a schematic flowchart of a method according to an embodiment of the present invention.

[0037] Figure 1 is a schematic perspective view of a storage and retrieval machine 1 according to an embodiment of the present invention. The storage and retrieval machine

[0038] The storage and retrieval machine 1 has a mast 11 on which a loading area 3 can be moved vertically in the vertical direction VR. The storage and retrieval machine has a drive element 12 which provides the necessary drive energy. Furthermore, the storage and retrieval machine 1 can be moved on a rail 13 in a horizontal direction HR. A load 2 can be picked up on the loading area 3. The storage and retrieval machine 1 is usually arranged in a rack storage system, where it can pick up loads.

[0039] 2. Store and retrieve items in appropriate shelves (not shown in Figure 1).

[0040] Figure 2 is a schematic and perspective view of a loading area 3 according to an embodiment of the present invention. The loading area 3 is designed to receive a load 2 (not shown in Figure 2). Furthermore, the storage and retrieval machine 1 has a sensor device 4. In the present embodiment, the sensor device consists of five individual sensors 41,

[0041] Sensors 41, 42, 43, 44, and 45 are arranged on the loading area 3. In the present embodiment, almost all sensors of the sensor device 4 are arranged on a frame 7. This ensures the necessary distance between the sensors and the load. In the present embodiment, sensors 41, 42, 43, 44, and 45 are LiDAR sensors. The sensor device 4 is designed to determine the current state of the load 2. Based on this, a control unit 5 can classify and control the load 2. The sensor device 4, which comprises a plurality of individual sensors 41, 42, 43, 44, and 45, can monitor a varying number of fields simultaneously. Thus, a digital map of the load 2 can be generated. The assignment of the load to a specific load type can be achieved by the main sensor 41, which is located on the rear side of the loading area 3.The main sensor 41 can detect all types of loads. To minimize or prevent a detection shadow that the main sensor 41 has when the load is arranged in a certain way on the loading area 3, the first secondary sensor 44 is installed at the front of the loading area 3. In combination with the main sensor 41, the first secondary sensor 44 can provide optimal results regarding the detection of the actual state of the load 2. In the present embodiment, the main sensor 41 is responsible for load control, including the total overhang of the load, the maximum height, and the overhang of the load at the rear of the loading area 3. The first secondary sensor 44 can be responsible for monitoring any overhang of the load at the front of the loading area 3.On the opposite side, the second auxiliary sensor 43 and the third auxiliary sensor 42 are mounted (one at the front, one at the back of the charging area) to detect the charge. The auxiliary sensors 42,...

[0042] Sensors 43 and 44 are responsible for checking the entire overhang of the load. To verify the general presence of a load, the fourth auxiliary sensor 45 is installed at the bottom of the loading area 3. Figure 3 is a schematic and perspective view of the loading area 3 with a load 2 placed on it. The remaining components correspond to those shown in Figure 2.

[0043] Figure 4A is a schematic representation of an area or section monitored by the sensor device 4 on the charging area 3. Figure 4A shows a view of the charging area 3 from the left side. The main sensor 41 is visible in the upper right area. The first secondary sensor 44 is located in the upper left area. These two sensors can detect all areas 61, 62, 63, 64, and 65.

[0044] Figure 4B is a schematic view from the right side showing sensors 42 and 43. These are designed to monitor an area into which an overhang of the load may protrude.

[0045] Figure 4C is ultimately a top view of the charging area with the fourth secondary sensor 45, which can detect a general presence of a charge 2 on the charging area.

[0046] According to one embodiment of the present invention, the method is based on a combination of the various sensors of the sensor device 4. Each sensor outputs the occupancy of the defined fields, thus enabling a digital assignment of the load (i.e., the determination of the current state). A classification of the load type is based on the main sensor 41, which is located at the rear of the loading area 3. The main sensor 41 can detect all types of loads 2 except for the rearmost load (area 63). A first secondary sensor 44 is installed at the front of the loading area 3, which, in combination with the main sensor 41, can fully detect the rearmost load. During operation, a warehouse is supplied with loads from production. When a load arrives at the input point, the storage and retrieval machine 1 receives an order from a control unit to pick up the load.The storage and retrieval machine 1 positions itself according to the entry point, extends forks to grasp the load, and retracts the forks so that the load 2 is placed onto the loading area 3. During the insertion process, the control unit monitors the feedback from the sensor device 4. When the forks reach the center (end position) of the loading area 3, a classification is performed. Using the reliable and accurate actual state of the load 2 obtained from the sensor device 4, the control unit compares the actual state with a target state. The height of the load is also detected. If a height exceeding a permissible maximum is detected, an error is triggered. If an error occurs, the operator can check the cause and decide whether to reclassify the load or return it to the delivery station.The load control method is based on a combination of different sensors. The first auxiliary sensor 44 and the second auxiliary sensor 43 on the left side of the loading area monitor the overhang on the left side of the loading area. Additionally, the first auxiliary sensor 44 ensures that the load overhang at the front of the loading area 3 is not exceeded. The same method is applied on the opposite side of the loading area 3, with two sensors 42 and 43 (second auxiliary sensor and third auxiliary sensor) that scan the load to be monitored, including the total overhang. During operation of the storage and retrieval machine 1, the load 2 is monitored by the sensor device 4. During monitoring, the front and rear profiles, as well as the height profile, are also checked by the sensor device 4.This prevents the storage and retrieval machine and / or any load it is transporting from colliding with a shelf. The LiDAR sensors monitor the safety zones, and the control unit triggers an error and stops the machine if anything is detected in a restricted area.

[0047] Figure 5 is a schematic flowchart of a method according to an embodiment of the present invention. In step S1, a load 2 is provided. In step S2, the load is transferred to the loading area 3 of the storage and retrieval machine. In step S3, the actual state of the load is determined. In step S4, the actual state is compared with a target state. In step S5, an operating state of the storage and retrieval machine is determined based on the comparison of the actual state with the target state.

[0048] Reference symbol list:

[0049] 1 storage and retrieval machine

[0050] 2. Load, cargo

[0051] 3 Charging area

[0052] 4 Sensor device

[0053] 5 Control unit

[0054] 7 frames

[0055] 11 masts

[0056] 12 Drive element

[0057] 13 rail

[0058] 41 Main sensor

[0059] 42 third secondary sensor

[0060] 43 second secondary sensor

[0061] 44 first secondary sensor

[0062] 45 fourth secondary sensor

[0063] HR Horizontal Direction

[0064] VR Vertical Direction

Claims

Claims 1. Storage and retrieval machine (1) for transporting a load (2), comprising: a loading area (3) configured to receive a load (2), a sensor device (4) configured to determine an actual state of the load (2), a control unit (5) configured to compare the actual state of the load (2) with a target state in order to determine deviation information, wherein the control unit (5) is further configured to determine an operating state of the storage and retrieval machine (1) based on the deviation information.

2. Storage and retrieval machine (1) according to claim 1, wherein the target state is a target state of the load (2).

3. Storage and retrieval machine (1) according to claim 1 or 2, wherein the sensor device (4) comprises at least two sensors.

4. Storage and retrieval machine (1) according to one of the preceding claims, wherein the sensor device (4) comprises at least one LiDAR sensor.

5. Storage and retrieval machine (1) according to one of the preceding claims, wherein the sensor device (4) comprises at least one image sensor.

6. Storage and retrieval machine (1) according to one of the preceding claims, wherein the sensor device (4) is configured to determine the actual state of the load (2) while the load (2) is being moved into the loading area (3).

7. Storage and retrieval machine (1 ) according to one of the preceding claims, wherein the operating state is indicative of an operational readiness or a fault of the storage and retrieval machine (1 ).

8. Storage and retrieval machine (1 ) according to one of the preceding claims, wherein the sensor device (4) is configured to monitor the state of the load (2) during operation of the storage and retrieval machine (1 ).

9. Storage and retrieval machine (1) according to one of the preceding claims, wherein the control unit (5) is configured to determine a type of load (2) based on the current state.

10. Storage and retrieval machine (1 ) according to one of the preceding claims, wherein the control unit (5) is configured to perform a classification based on the deviation information in order to determine the operating state of the storage and retrieval machine (1 ).

11. Storage and retrieval machine (1) according to one of the preceding claims, wherein the actual state comprises a representation of the load (2).

12. Storage and retrieval machine (1 ) according to one of the preceding claims, wherein the sensor device (4) comprises a plurality of sensors and wherein the control unit (5) is configured to determine the actual state of the load (2) based on the sensor output and the position of the respective sensor.

13. Storage and retrieval machine (1 ) according to one of the preceding claims, wherein the actual state comprises a point cloud that characterizes the load (2).

14. Storage and retrieval machine (1) according to one of the preceding claims, wherein the sensor device (4) comprises a plurality of sensors, and wherein the The control unit (5) is designed to determine the current state of the charge (2) based on the sensor output and the position of the respective sensor.

15. Storage and retrieval machine (1 ) according to one of the preceding claims, wherein the sensor device (4) comprises a main sensor (41 ) and at least one secondary sensor (42, 43, 44, 45).

16. Storage and retrieval machine (1 ) according to one of the preceding claims, wherein the control unit (5) is configured to perform a classification based on the sensor output of the main sensor (41 ) in order to determine a type of load (2).

17. Storage and retrieval machine (1 ) according to one of the preceding claims, wherein the sensor device (4) comprises a plurality of sensors, wherein at least one sensor is configured to detect whether a load (2) is arranged in the loading area (3).

18. Storage and retrieval machine (1) wherein the loading area (3) is divided into a plurality of sections, and wherein each sensor of the plurality of sensors is assigned to a section.

19. Storage and retrieval machine (1 ) according to one of the preceding claims, wherein the control unit (5) comprises a learning algorithm configured to determine an operating state of the storage and retrieval machine (1 ) based on the deviation information.

20. Storage and retrieval machine (1 ) according to one of the preceding claims, wherein the storage and retrieval machine (1 ) has an interface designed to receive user input that is indicative of which operating state is to be assigned to a given deviation information.

21. Method for loading a storage and retrieval machine (1) with a load (2), comprising: Providing a load (2), Transfer of the load (2) to a loading area (3) of the storage and retrieval machine (1 ), Determining the current state of the charge (2), Comparing the current state with a target state, Determining an operating state of the storage and retrieval machine (1 ) based on comparing the actual state with the target state.

22. Method according to claim 21, wherein the current state is determined during the transfer.

Images