Automatic control of docking and undocking of automated guided vehicle with forklift functionality

Abstract

The invention relates to a method for controlling an automated guided vehicle on a path between a starting position and an object handling position. The control includes navigating the automated guided vehicle in a predefined lane along a first portion of the path toward an aligned position. The control further includes navigating the automated guided vehicle along a second portion of the path towards the docking start location based on at least one predetermined navigation step. The control includes navigating the automated guided vehicle along a third portion of the path based at least in part on input from the primary sensor and stopping movement of the automated guided vehicle when the automated guided vehicle is aligned with the object handling location.

Description

Technical Field

[0001] The invention relates to a method of controlling an automated guided vehicle comprising a fork and an automated guided vehicle comprising at least one tine. Background Art

[0002] Controlling the engagement and disengagement of a vehicle with a pallet is known in the art from EP3729345. The vehicle is positioned based on a baseline geometric representation of the face of the pallet before penetrating the pallet with the tines. This baseline geometric representation is established using sensor data and automated guided vehicle positioning data. It may be determined that the updated geometric representation deviates from the baseline geometric representation by more than a threshold value, and in response, the movement of the vehicle may be adjusted.

[0003] The control method disclosed in EP3729345 is sensitive to errors in the positioning data and the proposed control method is unable to recognize situations where a pallet is pushed against a wall or other infrastructure / objects and / or where the drive wheels of the automated guided vehicle slip. Summary of the invention

[0004] The present inventors have identified the above-identified problems and challenges associated with the control of automated guided vehicles and have addressed these problems through the present invention as described below.

[0005] In one aspect, the invention relates to a method of controlling an automated guided vehicle on a path between a start position and an object handling position via an alignment position and a docking start position, the automated guided vehicle comprising at least one tine, at least one drive wheel, at least two casters, a controller, a primary sensor, and an auxiliary sensor,

[0006] wherein the controlling comprises navigating the automated guided vehicle in a predefined passage along a first portion of the path towards the alignment position,

[0007] wherein the controlling further comprises navigating the automated guided vehicle along a second portion of the path towards the docking start position based on at least one predetermined navigation step,

[0008] wherein the second portion of the path begins at the alignment position, and wherein the alignment position is automatically determined by the automated guided vehicle,

[0009] wherein the controlling includes navigating the automated guided vehicle along a third portion of the path based at least in part on input from the primary sensor, wherein the third portion of the path begins at the docking start location, and

[0010] Wherein the controlling includes stopping movement of the automated guided vehicle when the automated guided vehicle is aligned with the object handling location.

[0011] This method of controlling an automated guided vehicle is advantageous because the curve that the automated guided vehicle has to follow (the second part of the path) can be formed accurately on the first try, i.e. the automated guided vehicle does not need to move away from and towards the target position (to pick up or put down a load) several times. This makes picking up and putting down faster than with known methods and, more importantly, reduces the occupied area on the warehouse floor because the width of the aisle between two rows of loads can be reduced.

[0012] This approach is particularly advantageous on fully automated sections of the warehouse floor where manual handling of loads is not required. This is because the width of the aisle between two adjacent rows of loads can then be reduced in an ideal scenario to the length of the diagonal of the AGC (from one corner of the robot to the end of the opposite tine (if the AGC includes two tines)). This can help increase the number of rows on the warehouse floor.

[0013] The automated guided vehicle (sometimes referred to simply as an automatic mobile robot, mobile robot or simply robot) controlled according to the present invention can be of different types. It can be equipped with one or more drive wheels, which can be located at or near the center of the robot or can be located towards the side of the housing of the robot. Further, if, for example, a hollow tube is the object to be transported, the robot can include one tine, but more typically, the robot includes two tine, so that the robot can transport pallets.

[0014] Moving a load from one place to another should be understood as the main task of an automated guided vehicle. The forks of the automated guided vehicle are adapted to lift a pallet that directly carries an object or load. The pallet and the object together may be referred to as a load, but in this document references to a load or an object are the same unless otherwise mentioned. The load has a target location at which the automated guided vehicle picks up or puts down the pallet / object / load.

[0015] A channel should be understood as a passage between two adjacent rows of racks, loads, markers, etc. Alternatively, one or both sides of the channel may be virtually defined, such as by coordinates on a digital map. Thus, a channel is usually predetermined, but can also be defined as a row of loads handled by an automated guided vehicle. In the latter case, the channel can also be referred to as predetermined, because even when no load is positioned, the target position of the load is defined and therefore the position of the channel is defined.

[0016] A predetermined navigation step should be understood as a step of controlling the movement of the automated guided vehicle independently of the surroundings of the automated guided vehicle. Thus, input from, for example, a scanning LiDAR sensor may not be required to control the movement during such a step. Thus, one such step may be to continue driving 0.5 m in the current direction. Another such step may be to perform a 90 degree left turn, etc. However, it should be noted that a scanning LiDAR sensor may be used during a predetermined navigation step for safety reasons (i.e. to avoid collisions).

[0017] The starting position can be any position of the automated guided vehicle before reaching the alignment position in theory. In view of this, in this document, the starting position will usually be the position in the channel where the automated guided vehicle begins to move parallel to the channel side.

[0018] The alignment position is to be understood as a position on the path between the start position and the object handling position. Before the alignment position, i.e. along the first part of the path, the automated guided vehicle can be controlled according to the so-called normal operating mode. When the alignment position is reached, the control of the movement of the automated guided vehicle according to the normal operating mode is stopped. Therefore, the second part of the path can be said to start at the alignment position, and therefore from the alignment position, the movement of the automated guided vehicle can be controlled according to the predetermined navigation steps.

[0019] The docking start position is to be understood as the position in which the tines of the automated guided vehicle face the object (to be picked up) or the target position of the object (where the object is to be placed down).

[0020] An object handling position should be understood as a position or coordinate at which the automated guided vehicle should be positioned to pick up or put down an object.

[0021] In an exemplary embodiment of the present invention, the controller determines the alignment position based on predetermined information of a turning center of the automated guided vehicle, predetermined object size information, and predetermined object target position information.

[0022] A turning center is to be understood as a fixed position about which the automated guided vehicle can turn, for example, 90 degrees.

[0023] The object size information should be understood as, for example, the width, length, height, weight of the object, the position of the pallet pocket on the object, the size of the pallet pocket, etc.

[0024] Object target location information should be understood as information defining the location of the target location of the load / object (i.e. the location on the warehouse floor where the load / object is to be picked up or positioned). The object target location information may specify the row in which the object is to be picked up or dropped off and the location of the object in the row (its target location). Thus, the information may include the warehouse number, the row number, the number in the row (i.e. the first, second, third, etc. target location in the row), the first available target location, etc.

[0025] Before the automated guided vehicle reaches the alignment position, the controller preferably knows the turn center, information about the object size, and target position information. The controller can access this information, for example, from a data store of the automated guided vehicle. Thus, based on this predetermined information, the controller can navigate the automated guided vehicle to the correct row, and if the object target position is the first in the row, determine the alignment position based on this information.

[0026] In an exemplary embodiment of the present invention, the controller further determines the alignment position based on a comparison of predetermined neighboring object size information of an object located at a neighboring object target position with input from the primary sensor scanning the surroundings of the automated guided vehicle.

[0027] Neighboring object size information should be understood as information about the size of objects located near the predetermined target location of the object to be picked up or put down (i.e., at the neighboring object target location). This information may include the height, length, width (if the object stands on a pallet), etc. of the object.

[0028] Scanning the surroundings is to be understood in particular as scanning objects already present in the row in order to be able to match the scan data with predetermined adjacent object size information.

[0029] Therefore, comparing the scan information with the predetermined adjacent object size information is advantageous because the automated guided vehicle can automatically locate the target location of the object to be picked up or dropped off. With knowledge of the turning center and the object size information, the automated guided vehicle can also determine an alignment position from which predetermined navigation steps are used to ensure correct docking of the automated guided vehicle.

[0030] Docking of an automated guided vehicle should be understood as navigating the automated guided vehicle from a docking start position to an object handling position by one or more predetermined navigation steps. It should be mentioned that docking may include navigating the AGV according to one or more predetermined navigation steps starting from an alignment position.

[0031] The invention thus has the effect that the alignment position (from which the predetermined navigation step is to be performed) can be determined automatically and thus in real time (i.e. as the automated guided vehicle moves towards the object target position). The invention thus solves the problem of mismatch between the digital representation / digital coordinates of the position of the object or the position of the object to be positioned and the actual (real world) coordinates of the position of the object or the object to be positioned. Thus, even small displacements of one or more objects in the row can be taken into account.

[0032] This small displacement may also be referred to as object drift, i.e. when an object is not positioned exactly at the expected coordinates, subsequent objects may then also not be positioned correctly, and thus, if, for example, five objects are not positioned accurately (e.g. according to a digital map of the warehouse floor), the fifth object may overlap the target position of the sixth object. In this case, there is a risk that the automated guided vehicle or the sixth object will collide with the fifth object, or that the automated guided vehicle will cease to operate because the object is already at least partially positioned at the coordinates it has in order to position the sixth object.

[0033] In an exemplary embodiment of the invention, the alignment position is determined as a position on said first portion of said path, preferably in front of said turn center of said automated guided vehicle.

[0034] In front of the automated guided vehicle should be understood with reference to the direction of movement along the first portion of the path.

[0035] Based on the inputs described above, the controller may determine the alignment position before the automated guided vehicle reaches this position, so the automated guided vehicle may continue to move along the first portion of the path even when the alignment position has not yet been determined.

[0036] It should be mentioned that the automated guided vehicle may have passed the alignment position when the controller has established the alignment position. Therefore, in this case, the movement of the automated guided vehicle is stopped and then still continues to return in the direction it came along the first part of the path.

[0037] In an exemplary embodiment of the present invention, when the turn center is aligned with the object target position centerline, the automated guided vehicle is positioned at the aligned position.

[0038] The object target position centerline should be understood as a virtual line perpendicular to the direction of the first path from the center of the object target position. When the turning center of the automated guided vehicle is aligned with this line, the automated guided vehicle can perform a 90 degree turn, thereby facing the object target position with its tines.

[0039] The turning center of one type of automated guided vehicle may be significantly different from another type of automated guided vehicle because the wheel configuration of the automated guided vehicle may be different. Thus, in one automated guided vehicle, the turning center may be between the tines, while in another automated guided vehicle, the turning center may be below the body of the automated guided vehicle.

[0040] In an exemplary embodiment of the present invention, a controller determines a perpendicular intersection of the centerline of the object target location and the first portion of the path based on information received from the primary sensor.

[0041] This is advantageous because in this way the controller is able to determine whether the front face of the object to be picked up is parallel to the direction of movement of the first portion of the path.

[0042] In an exemplary embodiment of the invention, a controller determines the angular adjustment required to make said first portion of said path parallel to the front face of said object.

[0043] In an exemplary embodiment of the invention, the predetermined navigation step comprises performing a 90 degree turn of said automated guided vehicle such that said at least one tine is directed towards said object target location.

[0044] Thus, the second portion of the path may be a 90 degree turn of the automated guided vehicle. If the front of the object or the engaging position of the object, such as a pallet pocket, is not perpendicular to the first direction of the path, the predetermined navigation step may not necessarily be 90 degrees. The predetermined navigation step determines the parallel alignment of the tines with, for example, a pallet pocket.

[0045] In an exemplary embodiment of the present invention, navigation of the automated guided vehicle comprises another predetermined navigation step, wherein the another predetermined navigation step comprises moving the automated guided vehicle along the third portion of the path toward the object target position a distance determined by the object size information.

[0046] Now that the automated guided vehicle is aligned with the object target location, the automated guided vehicle can be navigated a certain distance towards the object target location based solely on the object's information. When the exact position of the automated guided vehicle relative to the object target location is known, the controller can determine the distance to move the automated guided vehicle in order to reach the object handling location, where the location of the object is aligned with the object target location / the location where the object is picked up.

[0047] The distance of the third portion of the path when docked may be determined by the size of the object to be positioned or the position on the tine where the object is positioned (the latter of which may be determined by the primary sensor).

[0048] In an exemplary embodiment of the invention, during said predetermined navigation step and / or said further predetermined navigation step, a controller controls the movement of said automated guided vehicle independently of inputs from said primary sensor and from said auxiliary sensor.

[0049] This is advantageous as it has the effect of providing a guaranteed movement of the automated guided vehicle, i.e. without interruptions due to obstacles which the sensors may or may not be able to register correctly. Thus, a faster and more accurate navigation of the automated guided vehicle is obtained.

[0050] Input from sensors may be avoided due to the assumption that there may be no obstacles within the turning radius of the automated guided vehicle (ie, the area covered by the automated guided vehicle when turning 90 degrees (ie, being navigated along the second portion of the path)).

[0051] In an exemplary embodiment of the present invention, the minimum width of the predefined channel is determined by the diagonal dimension of the automated guided vehicle.

[0052] An aisle width that is equal to the diagonal dimension of the automated guided vehicle is advantageous because the less space required for the aisle, the higher the number of rows of objects that can be provided on the warehouse floor.

[0053] In an exemplary embodiment of the invention, the channel includes at least one side defined by a row of object target locations.

[0054] In an exemplary embodiment of the invention, at least a portion of said first portion of said path is located on the half of said channel closest to a target location of an object to be picked up or dropped off by said automated guided vehicle.

[0055] In one aspect, the present invention relates to an automated guided vehicle comprising two tines each comprising a claw wheel, a drive wheel positioned below the automated guided vehicle between two casters positioned at opposite sides of the automated guided vehicle, a main sensor, a front sensor, and a controller,

[0056] wherein the controller is associated with a data store comprising control software, the controller being configured to control the automated guided vehicle based on the control software,

[0057] wherein the data storage further includes position information of a turning center located between the claw wheels and length information of a diagonal dimension of the automated guided vehicle,

[0058] wherein the controller is configured to automatically determine the alignment position as an intersection of a centerline of the object target position and a first portion of a path along which the automated guided vehicle is configured to move, and

[0059] Wherein the controller is configured to control the automated guided vehicle so as to stop the movement of the automated guided vehicle when the turning center coincides with the alignment position.

[0060] In an exemplary embodiment of the present invention, the controller is further configured to automatically determine the object target position centerline based on information stored in the data store and information provided by the front sensor and / or the main sensor. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] For a more complete understanding of the present disclosure, reference is now made to the following brief description in conjunction with the accompanying drawings and detailed description, wherein like reference numerals represent like parts. The accompanying drawings illustrate embodiments of the present invention, and elements of different drawings may be combined within the scope of the present invention:

[0062] Figure 1 An AGV according to an embodiment of the present invention is illustrated,

[0063] Figure 2 The different positions of the AGV from the starting position to the object handling position of the embodiment in which the AGV is to pick up an object are illustrated,

[0064] Figures 2 to 7 illustrates different positions of the AGV in an implementation scheme where the AGV is to drop an object,

[0065] Figure 8 illustrates an embodiment where an AGV is to pick up an object that is not positioned correctly, and

[0066] Fig. 9 An implementation is illustrated in which the AGC is to place a first object in a row of object target locations. DETAILED DESCRIPTION

[0067] The present invention has been described in view of exemplary embodiments intended only to illustrate the principles and specific implementations of the invention. A skilled person will be able to provide several embodiments within the scope of the claims.

[0068] Figure 1 An example of an automated guided vehicle AGV according to the present invention is schematically illustrated. The automated guided vehicle AGV is also referred to as AGV in this document. Figure 1 The main parts of the AGV are illustrated in a top view.

[0069] Figure 1The AGV illustrated in comprises a fork with two tines TI, such that the AGV is suitable for lifting a pallet carrying a load LO. The load may comprise an object OB positioned on the pallet and may also be referred to as an object. The actual position of the object OB may not be known in advance by the controller and is only determined by the controller CO with reference to the position of the AGV and, for example, also based on sensor input from the surroundings of the AGV. Thus, with the known position of the AGV and, for example, knowledge of the dimensions of the objects (load and pallet), the controller CO is able to determine and record the actual position of the object OB at least when establishing engagement between the tines TI and the pallet.

[0070] It should be mentioned that when referring to an AGV with a fork, that is to an AGV comprising one or more tines / prones, the AGV is adapted to lift objects such as for example a pallet carrying a load or a pipe around which a load can be wrapped.

[0071] The illustrated AGV includes a controller CO that controls the movement of the AGV. The movement includes driving the AGV forward and backward relative to the tine IT, turning the AGV clockwise and counterclockwise, lifting the tine TI, stopping the AGV, adjusting the movement speed, etc. The control is based on control software stored in a data storage associated with the controller (i.e., part of the controller CO or a part that communicates with the controller CO) and sensor inputs. The controller CO further controls the movement according to different operating modes. The controller can be a standard programmable logic controller or a dedicated AGV controller specially designed for this purpose. As known to those skilled in the art, in addition to data storage, the controller also communicates with / includes the following: a communication interface, a position determination device such as a global positioning system, and another element required to safely navigate the AGV.

[0072] The controller may allow the AGV to travel at a speed of, for example, 2 m / s during normal operation, and at one or more reduced travel speeds in docking / undocking mode. An example of this reduced travel speed may be 0.3 m / s. The speed and when which speed is allowed may be determined, for example, by the distance to the load, such that the slowest speed is when the forks are closest to the load. In this way, speed control may be provided based on the distance to the load.

[0073] The control software may include a map of the area in which the AGV is to move. The map may include obstacle locations, allowed / not allowed areas, target locations, starting locations, channels, etc. It should be mentioned that the map may be a virtual map that only includes the coordinates of map elements such as the above-mentioned obstacles, locations, and obstacles.

[0074] The sensor input may be provided by a series of sensors. Of primary importance for the present invention are the so-called primary sensors PS and the secondary sensors SS. Furthermore, tertiary sensors and additional sensors may provide relevant inputs.

[0075] The main sensor PS is also called a rear sensor because it is positioned opposite to the direction of movement of the AGV when controlled according to the normal operating mode. The main sensor PS is typically a distance sensor type and can be implemented as a scanning LiDAR sensor, which is an example of a common sensor type for AGVs. Alternatively, the main sensor PS can be implemented as a laser distance sensor.

[0076] Thus, the variation of the distance between the AGV and the object OB over a period of time can be established from the main sensor PS. The characteristic of the main sensor PS is that it relies on reflecting the object to measure the distance. The AGV may also include a front sensor, which is also implemented as a scanning lidar sensor, for example. Thus, the controller can establish a digital representation of the surroundings of the AGV from the rear sensor and the front sensor and thereby be able to locate itself and the object OB in the digital representation.

[0077] The second sensor SS is typically a motion sensor type from which the position over time can be measured or derived. The auxiliary sensor SS can be implemented as a distance measuring sensor that measures the distance traveled or a representation of the distance traveled during a period of time. The characteristic of the auxiliary sensor SS is that it is independent of the object so as to be able to perform measurements of the distance traveled by the AGV.

[0078] The tertiary sensor can be implemented as an inertial measurement unit that uses, for example, accelerometers and gyroscopes to determine the movement characteristics of the AGV and the distance traveled by the AGV. Additional sensors can also be used. Such additional sensors can be of the types mentioned above or other types. An example of an additional sensor for an AGV is called a front sensor, which is located opposite the tines and is used when the AGV moves in normal operating mode. Angular sensors also belong to the category of additional sensors, and such angular sensors can be of the scanning lidar type like the main sensor. Further, angular odometers also belong to the category of additional sensors, via which the angle at which the AGV turns can be determined.

[0079] The illustrated AGV includes different types of wheels, including a drive wheel DW, two casters CW, and two claw wheels TW. The drive wheel DW is usually positioned below the center of the AGV, while the casters CW can be positioned at the center of each side of the AGV. The claw wheels TW are positioned as supports for the tines TI when lifting the object OB / load.

[0080] In addition to the controller CO, the AGV control system may also include a safety controller (not illustrated). The robot control system may control the establishment of a planned robot route, the establishment of subroutes, the manipulation of the AGV (via wheels), and the execution of a Monte Carlo localization algorithm for determining the position of the AGV in a physical environment including various obstacles. The safety controller provides safety functions such as emergency braking for avoiding collisions with people and obstacles.

[0081] The robot control system and the sensing system including the above-mentioned sensors are communicatively connected so that the measurements performed by the scanning lidar sensor can be provided to the control system so that these measurements can be processed. The input from the sensors of the sensing system to the control system can therefore be used as a basis for maneuvering and positioning the AGV.

[0082] Optionally, the control system further provides an AGV state indicating the current operating state of the AGV. For example, a first possible state indicates normal operation of the AGV, and a second state indicates limited mobility of the AGV.

[0083] It should be noted that other types of AGVs can be controlled according to the present invention. One type includes a sensing system in the form of two scanning LiDAR sensors placed in opposite corners of the AGV, and are therefore referred to as corner sensors. Given the positioning of the corner sensors, scanning signals are permitted to travel from the corner scanners into the robot's surroundings, and each of the two corner scanners may be able to scan an angle of approximately 270 degrees. The corner scanners are therefore able to effectively cover a full 360 degrees horizontally around the AGV.

[0084] In this embodiment, the AGV includes casters in each corner and drive wheels in the center of each side of the AGV. Some of these wheels are responsible for supporting the AGV and some are responsible for moving and steering the AGV. Some wheels are associated with one or more odometers that provide an indication of the distance traveled by the AGV. This is useful for supporting navigation, especially when there are no obstacles in the surrounding environment of the AGV that can be used to determine its position, for example, relative to a map of the area. In other embodiments, the same wheels are used for guidance and distance measurement.

[0085] In addition, an AGV according to this embodiment may include a support surface for receiving an item to be transported by the AGV. The support interface may be implemented as a fork for lifting, for example, a pallet at one end of the AGV. Alternatively, the AGV may be designed to travel under an object / load for lifting and transporting a load on a bracket that may be positioned on the body of the AGV.

[0086] The diagonal dimension of the AGV, denoted ADS, is illustrated as the angle from one corner of the AGV to (in this embodiment) the opposite tine. The diagonal dimension specifies the space required to make the AGV turn, such as the minimum width of a passageway in which the AGV is to perform a 90 degree turn. It should be noted that if the AGV is carrying an object, the diagonal dimension ADS of the AGV may increase. This is the case if the length from one corner of the AGV to the opposite corner of the object is longer than the length of the illustrated diagonal dimension ADS.

[0087] The turning center TC of the AGV, denoted as TC, is illustrated as being between the two claw wheels. For this particular AGV, the turning center TC is between the claw wheels TW, and for other types, the turning center TC may be different relative to the AGV. When the AGV turns by turning the drive wheels 90 degrees, the turning center TC is defined as the point or center of a circle (whose circumference is defined by, for example, the drive wheels DW).

[0088] Figure 2 The different positions of the AGV are illustrated as the AGV is navigated along a path in channel COR to an object target position OTP where the AGV is to pick up an object OB. The path may be Figure 2 , and ends at the object handling position OHP. Since the AGV is illustrated as being at different positions on the path, it is illustrated in dotted lines. Further, given that the main sensor is part of the AGV in all positions, the scanning area SA of the main sensor PS is only included on the AGV positioned in the starting position SP. It should also be noted that the illustrated scanning area SA can be wider than illustrated, so the illustrated scanning area SA is only an example to support the understanding of the principles of the present invention. It should also be noted that some scanning areas SA are hidden by objects and are therefore illustrated in dotted lines.

[0089] An important function of the main sensor PS is to scan the area on the side of the AGV at least while approaching the target position TP / object handling position OHP of the object to be transported, so as to identify and determine the exact position of the object OB, pallet or other elements near the target position TP, partially on the target position TP and / or on the target position TP. As a result, the AGV controller can calculate the optimal alignment position AP and the time when the AGV reaches this position. Further, the controller can calculate the optimal docking start position DSP and ensure smooth driving and docking operations from the alignment position AP to the object handling position OHP and thereby ensure that the objects, pallets and / or other elements are placed at the desired distance relative to each other. The desired distance should be understood as the optimal distance from the object to be placed to the adjacent object that has been positioned. Further, it should be noted that, similar to placing an object, when picking up an object, the main sensor PS also advantageously scans the side of the AGV so that its controller ensures smooth driving, docking and undocking between the alignment position and the object handling position.

[0090] The width of the channel COR is at least the same as the length of the diagonal dimension ADS of the AGV, so that the AGV can perform a 90 degree turn within the defined channel COR. It should be mentioned that, preferably, the width of the channel is equal to the diagonal dimension of the AGV plus an additional distance ( Figure 2 This is to compensate for misalignment of the AGV in the channel, inaccurate positioning of the AGV, objects of different sizes to be carried by the AGV, etc.

[0091] The turning center TC of the illustrated AGV is located between the claw wheels TW towards the ends of the tines TI. Therefore, when turning around the turning center, the AGV requires more space towards the drive wheel end than towards the claw wheel end. Therefore, when navigating along the first portion FPA of the path, the main part of the AGV moves between the middle of the channel COR and the sides of the channel defined by the objects / target locations (rows) to be transported by the AGV. As mentioned, other types of AGVs may have other turning centers TC and therefore have another position when navigating along the first portion FPA of the path in the channel COR.

[0092] The AGV is illustrated as being at a starting position SP at a first portion of the path FPA. Along the first portion of the path FPA, the AGV is controlled according to a normal operating mode based on inputs (acceleration, velocity, distance traveled, etc.) from the above-mentioned sensors that scan the area around the AGV and monitor the AGV. The first portion of the path FPA can be any distance traveled by the AGV before the AGV reaches the alignment position AP.

[0093] The exact position of the AGV following the first portion of the path FPA in the aisle COR may be determined by the drive wheels DW or by the position of the turning center TC in the aisle (ie, the floor below the turning center TC).

[0094] As mentioned, the scanning area SA of the main sensor PS of the AGV is illustrated as being at the starting position SP. Based on the input from this main sensor PS, the controller CO is able to determine when an object OB located at an object target position OTP in the object row is passed. This information can be used by the controller CO to track where it is relative to the object target positions OTP that define the row of objects OB. In other words, the controller CO is therefore able to find the desired object target position OTP where it is instructed to pick up or drop off an object by tracking how many objects OB or empty object target positions OBT it has passed.

[0095] According to the present invention, the controller CO uses the input from the primary sensor to determine the object target position OTP at which the desired object DOB is to be picked up or dropped off. This is the case at least when the AGV reaches a target position that is an adjacent object target position AOTP of the desired object DOB. In this way, the adjacent object target position AOTP is used to find the desired object DOB.

[0096] exist Figure 2 The controller CO of the AGV in the starting position SP knows the number of the object target position OTP in the row where the AGV is to pick up the desired object DOB (in this particular embodiment). Further, the controller CO may know the geometry of the objects OB positioned at the individual object target positions OTP. Thus, the controller CO is able to identify part of the information received from the main sensor PS as, for example, the corners of the object and thereby how many objects it has passed. Alternatively or additionally, if the object target position OTP is empty, the controller CO may know this and in the same way know how many objects OB / object target positions OTP it has passed.

[0097] Therefore, in Figure 2 The controller CO of the AGV in the starting position SP knows that the currently scanned object OB is the second one before the desired object target position OTP. Therefore, the controller CO knows that the next object to be scanned is located at the object target position adjacent to the object target position OTP where the AGV is to pick up the desired object DOB.

[0098] As the AGV continues along the first portion FPA of the path, the main scanner continues to scan the area passed by the AGV, and when the scanned area SA reflects an object adjacent to the desired object, the controller CO preferably calculates the position of the alignment position AP as the AGV continues to move along the first portion FPA of the path.

[0099] The position of the first portion of the path FPA in the aisle is automatically determined by the controller based on information about the turn center TC and the diagonal dimension ADS. This position (i.e., the distance from the side of the aisle) is determined so that the intersection of the object target position centerline OTPCL and the first portion of the path becomes the alignment position AP. More specifically, making the alignment position AP allows the AGV to turn in the aisle without colliding with objects or obstacles that define the sides of the aisle.

[0100] In view of this, the AGV may move through the middle of the aisle COR and move toward one side of the aisle COR when approaching the object target position OTP to allow alignment of the turn center TC with the alignment position AP.

[0101] The object position centerline OTPCL can be determined from the input from the scanner and the information of the object geometry. The input from the scanner can make it possible to locate the adjacent object AOB (if any), which can then be used by the controller to determine the actual position of the object target position OTP / adjust the predetermined digital position (coordinates) of the object target position OTP. Therefore, when determining the actual position of the object target position, the controller easily finds the middle of the object target position and establishes a centerline parallel to the sides of the established object target position.

[0102] This manner of determining the centerline of the object target location can be used independently of the orientation of adjacent objects in a drop situation and independently of the desired object to be picked up in a pick situation.

[0103] It should be noted that even though the invention in this document describes determining the alignment position AP from the object target position centerline OTPCL, the alignment position AP can be found without information of the object target position centerline OTPCL. Instead, for example, the side of the desired object DOB or other scannable object or the measured distance to a known position can be used to determine the position of the alignment position AP.

[0104] When determined, the controller can draw / establish the coordinates of a virtual line and therefore be able to determine when the tines of the AGV are parallel to that line and therefore ready to move towards the object target location.

[0105] The alignment position AP is determined based on the information of the turning center TC of the AGV, the adjacent objects AOB (if any) and the desired object DOB. This information is received from sensors of the preferably AGV or from information stored in a data storage of the preferably AGV.

[0106] Relevant information related to the adjacent object AOB may include information related to the position of the adjacent object's corner facing the desired object DOB, the distance from the adjacent object's corner facing the desired object DOB to the desired object's corner facing the adjacent object AOB, the angle between the adjacent object AOB's front and the direction of the first part of the path FPA (used to determine how far is from parallel to this front to non-parallel to this direction), etc.

[0107] Relevant information related to the desired object DOB may include information related to the position of the corner of the desired object facing the adjacent object AOB, the geometry including the distance from the corner of the desired object facing the adjacent AOB to the center line OTPCL of the object target position, the angle between the front of the desired object DOB and the direction of the first part of the path FPA (used to determine how far is from parallel to this front to non-parallel to this direction), etc.

[0108] It should be noted that the object target position center line OTPCL may refer to the center line of the object and the object target position without the object.

[0109] Thus, the controller CO is able to calculate or determine how far to continue moving along the first portion of the path FPA before reaching the alignment position AP. The alignment position AP is reached when a 90 degree turn of the AGV positions the AGV in the docking start position. Thus, for example, when the turn center TC coincides with the object target position centerline OTPCL.

[0110] In this way, at least the X coefficient / value of the alignment position AP can be found. The X value can be defined as the position of the first part FPA along the path. Therefore, the alignment position AP can also include a Y coefficient / value defining the distance from the object, such as from the front of the object target position to the alignment position AP. Therefore, the automatic establishment of the alignment position AP can also include defining the Y value. In an ideal scenario where the fronts of all objects are aligned to form one side of the channel COR, the Y value can be predetermined. However, if the front of one or more objects deviates from this channel side, the Y value needs to be adapted accordingly to ensure that there is sufficient space for the AGV to turn. Therefore, based on information received, for example, from a scanner, the controller is able to determine the Y value for a specific object to be picked up (or put down).

[0111] For example, if one or two adjacent objects AOB are available, their corner / front positions are identified from the scan information and a "front line" / channel side can be established from this information. With reference to this front line / channel side, the controller can define the minimum distance along the object target position centerline OTPCL from the front line / channel side for the position of the alignment position AP.

[0112] Further, it should be mentioned that the alignment position can be determined by the controller CO based on the geometry information of the desired object DOB. Therefore, if the size of the desired object is large, thereby increasing the total diagonal size ADS of the AGV, the alignment position can be moved accordingly to ensure a safe turn of the AGV. In order to allow this, the channel width should be larger than the diagonal size of the individual AGV, i.e., the width should be the diagonal size ADS plus an additional distance, which can be determined by the diagonal size ADS of the AGV carrying this large object.

[0113] In an embodiment, before controlling the AGV along the second portion SPA of the path based on a predetermined navigation step, the angle at which the front of the desired object DOB is parallel to the direction of the first portion FPA of the path can be determined by scanning the front of the desired object DOB. Such scanning may require the AGV to pass the desired object, i.e., pass the alignment position AP or use information from a front scanning lidar sensor and a rear scanning lidar sensor, etc.

[0114] Moving the AGV along the second portion SPA of the path according to the predetermined navigation steps is preferably performed without input from any scanning sensor. Turning can be achieved by controlling the rotation of the drive wheel DW (preferably turning the drive wheel 90 degrees) and measuring the time to move along the second portion of the path, measuring the degree of rotation of the angular odometer, using a positioning system (global or local), etc.

[0115] Typically, in order to achieve a desired turn, ie, the degree required to turn the AGV to ensure that the object target position centerline OTPCL is parallel to the tine TI, the controller may receive input from one or more encoders.

[0116] Before initiating movement along the second portion SPA of the path, the angle encoder can provide information on the angular position of the drive wheel DW, such as the drive wheel angle with reference to the direction of movement of the first portion FPA of the path. To ensure that the AGV turns around the turning center TC, the angle of the drive wheel DW should be turned 90 degrees.

[0117] The wheel encoder can provide information on the distance traveled by the drive wheel DW. In an ideal scenario where the object OB is parallel to the direction of the first portion of the path FPA, the distance traveled is one quarter of the circumference of the circle followed by the drive wheel DW (whose center is at the turn center TC). But depending on the angle between the front of the desired object DOB and the direction of movement of the AGV when the alignment position AP is reached, more or less than one quarter of the circumference may be traveled.

[0118] When the alignment position AP is reached and the front of the desired object DOB is in the direction of the first portion FPA and therefore parallel to the AGV, the controller CO controls the AGV along the second portion SPA of the path according to a predefined navigation step. In the case where the front of the desired object DOB is parallel to the direction of the first portion FPA of the path, the predefined navigation step is a 90 degree turn of the drive wheel DW and a subsequent movement of the AGV along one quarter of the circumference of a circle (whose center is at the turn center TC) so that the tine TI points towards the desired object DOB and parallel to the object target position centerline OTPCL.

[0119] When the alignment position AP is reached and the front face of the object DOB is not parallel to the direction of the first portion FPA and therefore to the AGV, the controller CO first calculates the misalignment angle relative to parallelism and then compensates the predetermined navigation step with this alignment angle. Thus, in this case, the predetermined navigation step still includes turning the drive wheel DW by 90 degrees, but traveling along a portion of the circumference of a circle (whose center is at the turning center TC) with the misalignment angle compensation. In this way, the traveled portion of the circumference can be a little more or a little less than a quarter of the circumference of the circle (depending on how far parallel to the direction of the front face of the object and the first portion is).

[0120] As described, the controller CO determines the alignment position automatically, i.e. without prior knowledge of its position and without the assistance of external input from the AGV. Further, the controller CO performs predetermined navigation steps to guide the AGV along the second portion SPA of the path without using a scanning LiDAR sensor (it should be noted that a scanning LiDAR may be necessary for safety reasons and not for control purposes). Thus, the AGV can be precisely controlled so that the width of the aisle COR can ultimately be reduced to the length of the AGV diagonal dimension ADS.

[0121] The resulting position of the AGV after the predetermined navigation step is in the docking start position DSP.When the AGV is positioned in the docking start position DSP, the tines TI should be parallel or at least substantially parallel to the object target position centerline OTPCL.

[0122] like Figure 2 As illustrated in FIG, during the second portion of the path SPA, the AGV turns around the turning center TC, which is key to reducing the width of the channel COR and reducing the time of the docking procedure. This is because the AGV does not need space and time to move back and forth to align the tine TI with the object target location centerline OTPCL, for example based on input from a scanner.

[0123] From the docking start position DSP, the controller CO then controls the AGV along the third portion TPA of the path towards the desired object DOB. This portion of the control of the AGV's navigation can be based on the above-mentioned information available of the desired object DOB. Therefore, since the AGV's tines TI are parallel to the object target position centerline OTPCL and the turn center TC is aligned with the alignment position AP, the controller CO knows the distance to move the AGV along the third portion TPA of the AGV's path to reach the object handling position OHP.

[0124] Thus, the controller CO may only require input from an odometer, accelerometer, etc. to determine the distance traveled by the AGV to reach the object transfer location OPH, and thus terminate the AGV's travel along the third portion TPA of the path.

[0125] In this regard, a scan or distance measurement may be performed before the controller CO initiates movement along the third portion TPA of the path. This scan or measurement may be used to identify or verify the engagement location of the object such as the location of a pallet pocket, the distance from the AGV to the object OB, etc.

[0126] It should be noted that input from sensors other than those mentioned above can be used to improve control of the AGV, improve safety, ensure redundancy, confirm that control of the AGV is performed as expected, etc. Such sensors may include main sensors (rear sensors), front sensors (such as scanning LiDAR sensors as rear sensors), laser distance sensors, stop sensors, emergency sensors, etc.

[0127] The information relating to, for example, the object known / accessible to the controller CO may be information stored digitally in a memory with which the controller CO is able to communicate.

[0128] It should be noted that the AGV can also be navigated based on principles other than those described above. Further, it can also use the front sensor (scanning the area the AGV will pass through) instead of the main sensor (scanning the area the AGV has just passed through) to determine the desired object target position OTP.

[0129] Figures 3 to 6 An embodiment of the present invention is illustrated wherein an AGV is approaching an object target location OTP having a desired object DOB to be dropped off at the object target location OTP.

[0130] In addition to the rear sensor (main sensor PS), the AGV of this embodiment also includes a front sensor FS. The scanning area SA of these sensors is illustrated as a dashed semicircle, which scans adjacent objects AOB on both sides of the object target position OTP as the AGV moves along the first portion of the path FPA. Based on the information received from the scanner and the information included about the geometry of the object, the controller is able to identify the object target position OTP and its object target position centerline OTPCL.

[0131] Figure 3 The position of the AGV illustrated in FIG is on / along the first portion of the path FPA. The position of the AGV is such that the turning center TC is no longer aligned with the alignment position AP. Therefore, the AGV is still moving in the direction of the arrow, which illustrates the direction of movement along the first portion of the path FPA.

[0132] Controlling the movement of the AGV along the first portion FPA of the path may be as described above with respect to picking up a desired object.

[0133] Figure 4 The AGV is illustrated in a position where its turn center TC coincides with the alignment position AP and thus when movement along the first portion FPA of the path is terminated.

[0134] Figure 4 The arrows on illustrate sensor readings of the adjacent object AOB, which are used by the controller CO to determine the position of the AGV relative to the object target position centerline OTPCL.

[0135] The reason for indicating scanning information from both the front sensor FS and the main (rear) sensor PS is that objects carried by the AGV may block or reduce the scanning area of ​​the main sensor PS. Therefore, in this situation, the controller can benefit from the input from the front sensor FS, which can be used in the same way as the input received from the main sensor PS as described above.

[0136] The automatic determination of the alignment position AP and the alignment of the turn center TC with the alignment position AP may be as described above with respect to picking up the desired object.

[0137] Figure 5 The AGV is illustrated on / along the second portion SPA of the path. As illustrated, the wheels are turned 90 degrees. More specifically, the drive wheels positioned between the front sensor and the main sensor are mechanically rotated under the control of the controller. The casters on the sides of the AGV are rotated only when the rotating drive wheels initiate movement of the AGV.

[0138] Control of the AGV on / along the second portion SPA of the path may be as described above with respect to picking up the desired object.

[0139] Figure 6 The AGV is shown in the docking start position DSP, that is, the AGV's tines are parallel to the object's target position centerline OTPCL. Figure 2 As described in the above illustrated picking embodiment, in an ideal scenario, the AGV has moved one quarter of the circumference of a circle (with its center at the turn center TC / alignment position AP) to reach the docking start position DSP.

[0140] Figure 7 The AGV is illustrated in an object handling position OHP, i.e., after the AGV has moved along the third portion TPA of the path. The exact location of the object handling position can be determined as described above based on scans or distance measurements when the AGV is in the docking start position DSP or when the AGV moves along the third portion TPA of the path.

[0141] As also mentioned, control of the AGV by the TPA along the third portion of the path may be based solely on the odometer providing information to the controller of the distance traveled.

[0142] Once positioned in the object handling position OHP, in this embodiment, the AGV lowers the tines and thereby positions the desired object DOB at an object target position OTP, such as on a warehouse floor.

[0143] The AGV undocks from this position, i.e. leaves the object handling position (in this embodiment without an object). The AGV's undocking can be performed in the reverse order of the docking procedure, i.e. moving towards the docking start position DSP until the turning center TC is aligned with the alignment position AP, so that the AGV turns (clockwise or counterclockwise, after which the AGV will move when leaving) and leaves, for example, parallel to the object / object target position row.

[0144] Figure 8 An AGV is illustrated to pick up a desired object DOB whose front face is not aligned with an adjacent object AOB. This is another example of the advantage of the present invention, that the AGV can automatically determine the alignment position AP when the objects are not aligned in a row as expected.

[0145] As illustrated by the prescribed AGV following the first portion of the path FPA, the AGV is approaching the object target position OTP parallel to the row defined by the objects OB. As described above, the sensors (front and rear) are scanning the adjacent objects AOB, and the controller thereby knows that the desired object DOB is positioned at the next object target position OTP.

[0146] Further, the controller receives scan information of the desired object DOB, and the controller can determine from this information the orientation of the desired object. More specifically, the angle AN between the front face of the desired object and the direction of the first portion FPA of the path.

[0147] Based on this angle AN, the controller is able to change the direction of movement of the AGV along the first portion of the path FPA. Figure 8 In , the new direction is denoted as FPA' and is parallel to the front of the desired object DOB.

[0148] It should be noted that the new direction FPA' is not necessarily parallel to the front of the desired object DOB. It is important that the alignment position AP is still automatically established by the controller of the AGV and that the turning center TC of the AGV can be positioned at the alignment position AP. As in the embodiments described above, if the direction of movement is not parallel to the front of the desired object DOB, the length of the second portion SPA of the path is adjusted accordingly. Therefore, this distance is a little more or a little less than a quarter of the circle (whose center is at the turning center TC), depending on the angle AN.

[0149] Obviously, if the front of the desired object DOB is not aligned / parallel to the aisle side / object target location, the front sensor FS can be used to ensure that the AGV does not collide with obstacles / objects once it has exited the predetermined aisle COR. Crossing the aisle side is usually not allowed and would normally require human interaction, but with the present invention, the operation of the AGV can continue. Not only continue, but continue as desired, i.e. by picking up the desired object DOB.

[0150] It should be noted that the controller may deviate from the normal control routine. In fact, the controller may need to turn the AGV 180 degrees before reaching the alignment position AP, so that the AGV moves with its tines in front. The second portion of the path SPA is then subsequently controlled to turn clockwise, which is opposite to the counterclockwise turn described above and the second portion of the path SPA.

[0151] Fig. 9 The AGV is illustrated during positioning of the first object in a row of object target locations OTP1 to OPTn. The exact object target location where the object OB is to be placed is not important, as the principle is the same. The main principle of positioning the desired object target location OTP is to measure / determine the start of the row, and then determine when the correct object target location is reached based on knowledge of the dimensions of each of the object target locations. The positioning of the object is as described above.

[0152] Typically, markers or structural obstacles can be used to assist the AGV in determining where the first object target location is located. Alternatively or additionally, rows, aisles, and / or target locations can be defined by one or more sets of X-coordinates and Y-coordinates. As described, the Y-coefficient can be used to assist the AGV in aligning the objects in a row so that the front of the objects is in a straight line, preferably defining one side of the aisle or at least parallel to one side of the aisle.

[0153] In an exemplary embodiment of the invention, the controller uses guidance markers when determining said alignment position AP.

[0154] The use of guide markers is advantageous when a new row is to be started and therefore there are no existing / adjacent objects for determining the alignment position. The guide markers may be used in conjunction with a digital map of the area of ​​the row to navigate to the first part of the path or to find / determine the alignment position. The guide markers may be structural markers such as building components or markers with a geometric shape that is easily recognizable by an automated guided vehicle.

[0155] It should be mentioned that when determining / updating the position of the automated guided vehicle, the use of guidance markers may be weighted higher than neighboring objects.

[0156] In an exemplary embodiment of the present invention, the object target position information includes at least one predetermined X or Y value specifying a set of (X, Y) coordinates of said object target position OTP.

[0157] The object target position can be specified by one or more (X,Y) coordinates. The center point of the object can be located at a specific center point of the object target position defined by a set of (X,Y) coordinates. Alternatively, the corners of the object are aligned with four sets of (X,Y) coordinates that define the corners of the object target position.

[0158] The X value of the set of (X, Y) coordinates of the target position of the object may be a position along the first portion of the path, i.e., a position parallel to the side of the passage. The Y value of the set of (X, Y) coordinates of the target position of the object may be a position at a distance from the X value, i.e., along a direction perpendicular to, for example, the side of the passage and the direction of the first path.

[0159] By specifying an X value for one or more object target positions, the distance between two objects can be predetermined. By specifying a Y value for one or more object target positions, the front faces of the objects can be aligned.

[0160] Determining the object target position from a set of (X, Y) coordinates is particularly advantageous if the object target position is the first object target position in a row of object target positions.

[0161] In an exemplary embodiment of the present invention, the alignment position AP is defined by a Y value specifying a distance from the front face of the object target position OTP to said alignment position AP.

[0162] In an exemplary embodiment of the invention, the position of the front of the desired object DOB is automatically determined by said controller CO based on information received from sensors.

[0163] This is advantageous because it is then ensured that the aiming position is sufficiently far from the desired object so that a safe turn of the AGV can be made.The sensors should include at least one of a main sensor and a front sensor that scans the surroundings of the AGV, but may also include a distance sensor.

[0164] In an exemplary embodiment of the invention, the alignment position AP is defined by a Y value calculated based on the geometry of the desired object DOB.

[0165] This is advantageous because the alignment position can be adapted to the object to be picked up or put down. This calculation of the Y value can be done in real time while the AGV is moving or when the object geometry is available. If the Y value is predetermined, the calculation can be done outside the AGV.

[0166] In an exemplary embodiment of the invention, the controller CO utilizes said angle adjustment AN to adjust the turning angle of said second portion SPA of said path.

[0167] This is advantageous because the automated guided vehicle is then controlled to approach the object perpendicularly to the front of the object, regardless of the orientation of the object.

[0168] This is advantageous as it has the effect that the automated guided vehicle can undocking and pick up an object that is not aligned with the direction of movement along the first portion of the path (ie parallel to the first portion of the path).

[0169] The required adjustment may cause the second portion of the path to turn more or less than 90 degrees, depending on the misalignment of the object.

[0170] In an exemplary embodiment of the invention, the automated guided vehicle AGV is controlled to move in three different directions, depending on which of the first portion FPA, the second portion SPA and the third portion TPA of the path the automated guided vehicle is in.

[0171] In the first part, the tines point away from the direction of movement, in the second part, the automated guided vehicle moves sideways / turns, and in the third part, the tines point towards the direction of movement. This is advantageous because it has the effect that the width of the passage can be reduced.

[0172] In an exemplary embodiment of the invention, a controller receives input from the primary sensor and the auxiliary sensor while navigating the automated guided vehicle along the first portion FPA and the third portion TPA of the path.

[0173] The primary sensor may be a scanning lidar sensor, while the secondary sensor may be an odometer type sensor.

[0174] In an exemplary embodiment of the present invention, the minimum width of the predefined corridor COR is determined by the automated guided vehicle diagonal dimension ADS plus an additional distance.

[0175] Including the additional distance to the width of the passage is advantageous because it ensures that sufficient space is provided in the passage for an automated guided vehicle carrying an object that increases the diagonal size of the automated guided vehicle. Furthermore, if the turning or positioning of the automated guided vehicle is not completely precise, space is provided for turning in this way.

[0176] In an exemplary embodiment of the invention, the channel COR is defined on both sides by a number of rows of object target positions OTP.

[0177] Having an aisle defined by two rows of object target locations is advantageous because the occupied area on the warehouse floor is reduced due to the possibility of engaging two rows of objects from the same aisle.

[0178] In an exemplary embodiment of the invention, the position of said first portion FPA of said path in said channel COR is determined by said turning center TC.

[0179] This is advantageous because it allows for a lane width that does not exceed the diagonal dimension of the AGM (possibly with an additional turning safety margin). The turning center between the claw wheels may indicate that the first portion of the path is closer to the side of the object to be picked up or dropped off, while the turning center in the middle of the AGM (without tines) may indicate that the first portion of the path is along the middle of the lane.

[0180] In an exemplary embodiment of the invention, when the neighboring object is verified by the controller, the position of the automated guided vehicle AGV in the digital representation of the operating area of ​​the automated guided vehicle AGV is updated.

[0181] The digital representation of the operating area should be understood as a map comprised by or accessible to the controller.Updating the position of the automated guided vehicles is advantageous because more accurate control and control planning of a fleet of automated guided vehicles may be provided.

[0182] In an exemplary embodiment of the present invention, the controller associates the content information of the object with the object target position OTP.

[0183] This is advantageous because the user can then specify the content and the automated guided vehicle knows where this content is.

[0184] In an exemplary embodiment of the invention, the controller CO is further configured to control the movement of said automated guided vehicle AGV from said first portion of said path FPA to an object handling position according to at least one predetermined navigation step.

[0185] In an exemplary embodiment of the invention, the invention relates to a device 10 according to claim 14 implementing the method specified in claims 1 to 13 and as claimed in claims 1 to 13 .

[0186] It is now clear from the above that the present invention relates to an automated / automatic guided vehicle, referred to throughout this document as an AGV. The AGV is capable of automatically (i.e. without human interaction) and in real time (i.e. as the AGV moves towards a target position) determining an alignment position AP. The position of the alignment position AP is in one direction (X) determined by the object target position centerline OTPCL (or the side of the object, an adjacent object, a marker, etc.), which is also automatically determined, verified or adjusted. In a second direction (Y), the alignment position is determined by the diagonal dimension ADS of the AGV, so that when the turning center TC of the AGV coincides with the alignment position, the AGV is able to perform a predetermined navigation step (such as a 90 degree counterclockwise turn) without crossing one of the two sides of the predefined channel COR.

[0187] The present invention has been illustrated above with reference to specific examples of methods and robotic systems for purposes of illustration and not limitation. Details such as specific methods and system structures have been provided for the purpose of understanding the embodiments of the present invention. It should be noted that detailed descriptions of well-known systems, devices, circuits, and methods have been omitted so as not to obscure the description of the present invention with unnecessary detail.

[0188] List

[0189] AGV.Automated Guided Vehicle

[0190] DW.Drive wheel

[0191] CW.Casters

[0192] TW.Claw wheel

[0193] TI. (fork) tines

[0194] PS. Main sensor (rear sensor)

[0195] SS. Auxiliary sensor

[0196] FS.Front sensor

[0197] CO. Controller

[0198] ADS.AGV diagonal dimensions

[0199] TC. Turning Center

[0200] SA. Scanning area

[0201] AN.Angle

[0202] SP.Starting position

[0203] TP. Target location

[0204] AP. Alignment

[0205] DSP. Docking start position / undocking stop position

[0206] OHP. Object Handling Position

[0207] OTP. Object Target Position

[0208] AOTP. Adjacent object target position

[0209] COR.Channel

[0210] OB. Object

[0211] DOB. Expected object

[0212] AOB.Adjacent Objects

[0213] OTPCL. Object target position center line

[0214] The first part of the FPA path

[0215] SPA. The second part of the path

[0216] The third part of the TPA path

Claims

1. A method of controlling an automated guided vehicle (AGV) on a path between a start position (SP) and an object handling position (OHP) via an alignment position (AP) and a docking start position (DSP), the automated guided vehicle (AGV) comprising at least one tine (TI), at least one drive wheel (DW), at least two casters (CW), a controller (CO), a primary sensor (PS) and a secondary sensor (SS), wherein the controlling comprises navigating the automated guided vehicle (AGV) along a first portion of the path (FPA) towards the alignment position (AP) in a predefined corridor (COR), wherein the controlling further comprises navigating the automated guided vehicle (AGV) along a second portion (SPA) of the path towards the docking start position (DSP) based on at least one predetermined navigation step, wherein said second portion (SPA) of said path starts at said alignment position (AP), and wherein said alignment position (AP) is automatically determined by said automated guided vehicle (AGV), wherein the controlling comprises navigating the automated guided vehicle (AGV) along a third portion (TPA) of the path based at least in part on input from the primary sensor (PS), wherein the third portion (TPA) of the path begins at the docking start position (DSP), and Wherein the controlling includes stopping the movement of the automated guided vehicle (AGV) when the automated guided vehicle (AGV) is aligned with the object handling position (OHP).

2. The method for controlling an automated guided vehicle (AGV) according to claim 1, wherein the controller (CO) determines the alignment position (AP) based on predetermined information of a turning center (TC) of the automated guided vehicle (AGV), predetermined object size information, and predetermined object target position information.

3. A method for controlling an automated guided vehicle (AGV) according to claim 1 or 2, wherein the controller (CO) further determines the alignment position (AP) based on a comparison of predetermined adjacent object size information of an object located at an adjacent object target position (AOTP) with input from the primary sensor (PS) scanning the surrounding environment of the automated guided vehicle (AGV).

4. Method for controlling an automated guided vehicle (AGV) according to any of the preceding claims, wherein the alignment position (AP) is determined as a position on the first portion (FPA) of the path, preferably in front of the turning center (TC) of the automated guided vehicle (AGV).

5. A method of controlling an automated guided vehicle (AGV) according to any one of the preceding claims, wherein the automated guided vehicle (AGV) is positioned at the alignment position (AP) when the turn center (TC) is aligned with an object target position centerline (OTPCL).

6. Method for controlling an automated guided vehicle (AGV) according to any of the preceding claims, wherein the controller (CO) determines a perpendicular intersection of the object target position centerline (OTPCL) with the first portion (FPA) of the path based on information received from the primary sensor (PS).

7. Method of controlling an automated guided vehicle (AGV) according to any of the preceding claims, wherein the controller (CO) determines the angular adjustment (AN) required to make the first portion of the path (FPA) parallel to the front face of the object.

8. A method of controlling an automated guided vehicle (AGV) according to any one of the preceding claims, wherein the predetermined navigation step comprises performing a 90 degree turn of the automated guided vehicle (AGV) such that the at least one tine (TI) points towards the object target position (OTP).

9. A method for controlling an automated guided vehicle (AGV) according to any one of the preceding claims, wherein the navigation of the automated guided vehicle (AGV) comprises a further predetermined navigation step, wherein the further predetermined navigation step comprises moving the automated guided vehicle along the third portion (TPA) of the path towards the object target position (OTP) a distance determined by the object size information.

10. Method for controlling an automated guided vehicle (AGV) according to any of the preceding claims, wherein the controller (CO) controls the movement of the automated guided vehicle during the predetermined navigation step and / or the further predetermined navigation step independently of the input from the primary sensor (PS) and from the auxiliary sensor (SS).

11. Method of controlling an automated guided vehicle (AGV) according to any of the preceding claims, wherein the minimum width of the predefined channel (COR) is determined by the automated guided vehicle diagonal dimension (ADS).

12. Method of controlling an automated guided vehicle (AGV) according to any of the preceding claims, wherein the corridor (COR) comprises at least one side defined by a row of object target positions (OTPs).

13. A method for controlling an automated guided vehicle (AGV) according to any of the preceding claims, wherein at least a portion of the first portion (FPA) of the path is located on the half of the channel (COR) closest to the object target position (OTP) where the automated guided vehicle (AGV) is to pick up or put down an object (OB).

14. An automated guided vehicle (AGV), comprising two tines (TI) each comprising a claw wheel (TW), a drive wheel (DW) positioned below the automated guided vehicle (AGV) between two casters (CW) positioned at opposite sides of the automated guided vehicle (AGV), a main sensor (PS), a front sensor (FS) and a controller (CO), wherein the controller (CO) is associated with a data store comprising control software, the controller (CO) being configured to control the automated guided vehicle (AGV) based on the control software, wherein the data storage further includes position information of a turning center (TC) positioned between the claw wheels (TW) and length information of a diagonal dimension (ADS) of the automated guided vehicle (AGV), wherein the controller (CO) is configured to automatically determine an alignment position (AP) as an intersection of an object target position centerline (OTPCL) and a first portion (FPA) of a path along which the automated guided vehicle is configured to move, and The controller (CO) is configured to control the automated guided vehicle (AGV) so as to stop the movement of the automated guided vehicle (AGV) when the turning center (TC) coincides with the alignment position (AP).

15. The automated guided vehicle according to claim 14, wherein the controller (CO) is further configured to automatically determine the object target position centerline (OTPCL) based on information stored in the data store and information provided by the front sensor (FS) and / or the main sensor (PS).

Images