Routing of container handling vehicles operating in an automated storage system

Abstract

A method, control system, and computer program for routing and re-routing container handling vehicles that handle storage containers in an automated storage and retrieval system comprising a skeletal structure that forms a three-dimensional storage grid structure for storing storage containers within storage columns, the skeletal structure including a grid-based rail system arranged above the storage columns, the rail system providing available routes for the container handling vehicles that handle and transport the storage containers to and from the storage columns.

Description

Technical Field

[0001] (Field of the Invention) The present invention relates to an automated storage and retrieval system for storage containers handled by a container handling vehicle that travels on a grid-based rail system of a storage and retrieval system, and more particularly to a method, system, and computer program for routing and rerouting a container handling vehicle according to available routes on the rail system.

Background Art

[0002] (Background) FIG. 1 discloses a typical prior art automated storage and retrieval system 1 with a skeletal structure 100, and a container handling vehicle 201, also known as a robot, operates on the system 1.

[0003] The skeletal structure 100 includes upright members 102, horizontal members 103, and a storage volume including storage columns 105 arranged side by side between the upright members 102 and the horizontal members 103. In these storage columns 105, storage containers 106, also known as containers, are stacked on top of each other to form stacks 107. The members 102, 103 may typically be made from metal, for example, extruded aluminum profiles.

[0004] The skeletal structure 100 of the automated storage and retrieval system 1 includes a rail system 108 arranged across the upper part of the skeletal structure 100, and a plurality of container handling vehicles 201 are operated to lift storage containers 106 from the storage columns 105 on the rail system 108, lower the storage containers 106 into it, and also transport the storage containers 106 above the storage columns 105.

[0005] The rail system 108 comprises a first set 110 of parallel rails arranged to guide the movement of a container handling vehicle 201 in a first direction X that crosses the top of the frame structure 100, and a second set 111 of parallel rails arranged perpendicular to the first set 110 for guiding the movement of a container handling vehicle 201 in a second direction Y that is perpendicular to the first direction X.

[0006] The first rail 110a in the first direction X, the second rail 110b in the first direction X, the first rail 111a in the second direction Y, and the second rail 111b in the second direction Y are also shown in Figure 1. The container handling vehicle 201 moves laterally above the storage column 105, that is, it moves in a plane parallel to the horizontal XY plane.

[0007] The type of container handling vehicle 201 used may be any publicly known in the art, and may be, for example, one of the automated container handling vehicles disclosed in WO2014 / 090684A1 (Patent Document 1) or WO2015 / 193278A1 (Patent Document 2), having an area that covers one or two storage columns 105. The rail system 108 may be arranged in a single-track and / or double-track configuration.

[0008] The storage containers 106 are stored in columns 105 that define a third direction Z perpendicular to a first direction X and a second direction Y. The storage containers 106 are accessed by container handling vehicles 201 through access openings 112 in a rail system 108, i.e., the rail system 108 is arranged on a skeletal structure 100 that defines the outer perimeter of each access opening 112 above each storage column 105. Upright members 102 of the skeletal structure 100 may be used to guide the storage containers when moving them out of the columns 105 and when lowering containers into them. The stacks 107 of containers 106 are typically freestanding.

[0009] The storage volume of the skeletal structure 100 is often referred to as a grid 104, and the possible storage locations within the storage columns 105 in this grid are referred to as storage cells. Each storage column 105 may be identified by its position in the X and Y directions, while each storage cell may be identified by the number of containers in the X, Y, and Z directions.

[0010] In the skeletal structure 100, most of the columns 105 are storage columns 105, that is, columns 105 into which storage containers 106 are stored within the stack 107. However, some columns 105 may serve other purposes.

[0011] In Figure 1, columns 119 and 120 are dedicated columns used by a container handling vehicle 201 to load and unload storage containers 106 so that they can be transported to an access station (not shown), where the storage containers 106 can be accessed from outside the frame structure 100 or transported inside and outside the frame structure 100 at the access station. The access station is typically a picking or stock station where product items are removed from or positioned in the storage containers 106.

[0012] In the art, columns for transporting storage containers into and out of a storage system are typically referred to as port columns 119, 120 or transport columns. Storage containers are typically transported into and out of port columns 119, 120 via ports 119', 120' located at openings at the ends of the port columns 119, 120; that is, storage containers enter or exit port columns 119, 120 at ports 119', 120'. The ports may be located elsewhere, such as on an intermediate or ground level of port columns 119, 120.

[0013] The transport and transfer of the storage containers 106 to the access station may be in any direction: horizontal, diagonal, and / or vertical. For example, the storage containers 106 may be placed in random or dedicated columns 105 within the skeletal structure 100, then loaded by any container handling vehicle 201, and transported to port columns 119, 120 for further transport to the access station. Note that the term “diagonal” refers to the transport of the storage containers 106 having a rough orientation of transport somewhere between horizontal and vertical.

[0014] In Figure 1, the first port column 119 may be a dedicated loading / unloading port column from which a container handling vehicle 201 can load and unload storage containers 106 to be transported to an access or transfer station, and the second port column 120 may be a dedicated loading port column from which a container handling vehicle 201 can load storage containers 106 being transported from an access or transfer station.

[0015] When the automated storage and retrieval system 1 is operating, each container handling vehicle 201 is given a task by receiving an instruction. The task could be, for example, to retrieve a specific storage container 106 from storage column 105 and deliver it at port column 119 for further transport to an access station, or to move a storage container 106 from one storage cell to another. This means that each container handling vehicle 201 is instructed to follow a set route on rail 111 from their current location to a target location.

[0016] When a particular storage container 106 stored in one of the columns 105 disclosed in Figure 1 needs to be retrieved, one of the container handling vehicles 201 is assigned a task and instructed to retrieve the storage container 106 from its location and transport it to port 119' of port column 119. This operation involves moving the container handling vehicle 201 to a location above the storage column 105 in which the storage container 106 is located, using a lifting device (not shown) of the container handling vehicle 201 to retrieve the storage container 106 from the storage column 105, and transporting the storage container 106 to port 119' of port column 119.

[0017] If the target storage container 106 is located deep within the stack 107, i.e., if one or more other storage containers 106 are located above the target storage container 106, the operation also involves temporarily moving the storage containers located above it before raising or lowering the target storage container 106 from the storage column 105. This step, sometimes referred to in the art as "excavation," may subsequently be carried out using the same container handling vehicle 201 used to transport the target storage container to the loading / unloading port column 119, or using one or more other cooperating container handling vehicles 201. Alternatively, or in addition, the automated storage and retrieval system 1 may have a container handling vehicle 201 specifically dedicated to the task of temporarily removing storage containers from the storage column 105. Once the target storage container 106 is removed from the storage column 105, the temporarily removed storage container 106 may be repositioned in the original storage column 105. However, the removed storage container 106 may be relocated to another storage column 105 as an alternative.

[0018] When a storage container 106 is to be stored in one of the columns 105, one of the container handling vehicles 201 is commanded to load the storage container 106 from port 120' of port column 120, which transports storage containers from access or transfer stations, and to transport it to a position above the storage column 105 where it is to be stored. After any storage containers located in or above the target position in the storage column stack 107 are removed, the container handling vehicle 201 positions the storage container 106 in the desired position. The removed storage container 106 may then be lowered back into the storage column 105 or repositioned in another storage column.

[0019] The location of each storage container 106 within the skeletal structure 100, as well as the location and movement of each container handling vehicle 201 operating on the storage and retrieval system, are continuously monitored and controlled by the control system 500 and by referencing the contents of each storage container 106, so that the desired storage container 106 can be delivered to the desired location at the desired time without the container handling vehicles 201 colliding with each other.

[0020] An embodiment of a typical control system 500 is shown in Figure 2. The control system 500 is adapted to transmit operation and movement commands to a vehicle controller 230 in each container handling vehicle 201 in order to control all movement and operation on the storage and retrieval system 1. The control system 500 commands each vehicle 201 to store or retrieve the storage containers 106. The current position of each vehicle 201 is communicated from the vehicle 201 to the control system 500, so that it can control the movement of all vehicles 201 on the rail system 108 in an optimal manner without involving vehicles 150 queuing or collisions.

[0021] In the embodiment shown, the control system 500 comprises a master controller 220, a database 210, a routing planner 200, and a transmitter / receiver 225 for communicating commands to each container handling vehicle 201.

[0022] The database constantly tracks not only the location of the storage containers 106, but also which storage containers 106 should be handled, i.e., which storage containers 106 should be retrieved or stored in the storage grid 104. The routing planner 200 is adapted to find the optimal route for the container handling vehicles 201.

[0023] The master controller 220 is adapted to integrate information from the database 210 and the routing planner to generate operational commands for each container handling vehicle 201 (for example, which storage containers to handle and which route on the rail system 108 to follow according to input from the routing planner 200).

[0024] The control system 500 communicates with a central computer to which commands and tasks are transmitted.

[0025] The automated storage and retrieval system 1 typically operates with multiple container handling vehicles 201 that transport storage containers from one location to another. Larger storage and retrieval systems may have hundreds of additional container handling vehicles 201, each assigned and given tasks by receiving commands transmitted from the control system 500. For example, a task might be to retrieve a particular storage container 106 from a storage column 105 where it is located, traverse a set route by driving a set distance in a set direction, and deliver the storage container 106 at a port 119', etc.

[0026] The commands are given by the control system 500, which commands each container handling vehicle 201 to perform tasks and follow a specific route from the current location to the target destination. The overall goal is to find the best routes for each container handling vehicle 201 and routes that do not collide with each other. The best route is typically the shortest route that does not collide with other routes within the same time frame.

[0027] The Multi-Agent Path Finding (MAPF) algorithm is known in the technical field of computer science and is used to solve the problem of finding paths for multiple agents from their current locations to their target locations without colliding with each other, while simultaneously optimizing a cost function such as the sum of the path lengths of all agents.

[0028] Regarding the storage and retrieval system 1, the container handling vehicle 201 represents an agent. However, most MAPF algorithms are one-shot algorithms for solving specific scenarios, and the container handling vehicle 201 follows the set route from the current position to the final position for performing the task until it finishes its planned tasks before creating a new route according to a new task. It is not obvious to adapt these multi-agent path finding algorithms to a dynamically changing environment where the task or target type can be changed before the container handling vehicle 201 finishes its first assigned task or where there are new active container handling vehicles 201 operating on the storage system.

[0029] By using the MAPF algorithm, the container handling vehicles 201 can be routed simultaneously, but since the MAPF algorithm has a very long execution time, the container handling vehicles 201 must wait while a new route is being generated.

[0030] In the 2005 research paper "Cooperative pathfinding" by David Silver (www.aaai.org), a sliding window that defines a restricted area is used for routing to search for non-collision routes within the area in a real-time environment. However, the collaborative search for possible routes is limited to a certain depth specified by the current window. Each agent searches for a partial route to its destination and then begins to follow that route. At fixed intervals, for example, when the agent is in the middle of its partial route, the window is shifted forward and a new partial route is calculated. This means that the collaborative search only considers agents within the sliding window, i.e., other agents outside the window that defines the restricted area are ignored. To search for the direction of the agent to its destination, an abstract search at full depth is performed.

[0031] The present invention presents a method for routing in which a fixed time frame is set for only the first part of each route while a complete route from the current location of a container handling vehicle to its destination is searched. This method provides a time-efficient routing in which rerouting can be performed without stopping the container handling vehicle.

Prior Art Documents

Patent Documents

[0032]

Patent Document 1

Patent Document 2

Summary of the Invention

Means for Solving the Problems

[0033] (Summary of the Invention) The object of the present invention is to provide improved methods for routing and rerouting container handling vehicles, which perform these operations while running on an automated rail system. Novel methods, systems, and computer programs provide more route alternatives for all container handling vehicles than previous routing methods have provided.

[0034] This is achieved by routing and rerouting container handling vehicles as frequently as possible while they are traveling from their current location to their target destination, and by considering all routes for all container handling vehicles from their current location to the task at the destination location.

[0035] According to this inventive method, the routes of container handling vehicles are calculated on the fly as the vehicles move toward their destinations. Container handling vehicles can be efficiently rerouted to follow new routes to adapt to changing traffic conditions, and new routes can be created for newly active container handling vehicles.

[0036] The new method requires less time to explore new routes for all container handling vehicles without stopping them all, and provides more route alternatives for all container handling vehicles. According to the method, MAPF is used to explore routes for all container handling vehicles to their final destination, but in contrast to conventional techniques where the entire route is fixed or only routes for vehicles within a restricted area defined by a sliding window are evaluated, as described in the research paper by David Silver mentioned above, only the first part of each route is fixed for a set time interval.

[0037] The new method provides more route alternatives because, as container handling vehicles move along their assigned routes, portions of each covered route are freed up for new routes and portions of assigned routes that are not yet fixed.

[0038] The method of the present invention provides a fluid, constantly changing system, and MAPF not only enables this, but also reduces journey time compared to previous planned algorithms because the path is interrupted for less time.

[0039] The present invention is defined by a method for routing and rerouting container handling vehicles that handle storage containers in an automated storage and retrieval system comprising a skeletal structure forming a three-dimensional storage grid structure for storing storage containers in storage columns, the skeletal structure comprising a grid-based rail system arranged above the storage columns, the rail system providing available routes for container handling vehicles to and from the storage columns, each container handling vehicle comprising a first set of wheels configured to move the vehicle along a first lateral direction (X) of the grid-based rail system and a second set of wheels configured to move the vehicle along a second lateral direction (Y) of the grid-based rail system, the second direction (Y) being perpendicular to the first direction (X), and the movement of the container handling vehicles being controlled by a control system that determines what tasks should be performed by a designated container handling vehicle, the location of the destination for performing the tasks, and what routes on the rail system the container handling vehicles should travel.

[0040] The method includes the following steps, which are performed by the control system. a. A step in the control system to run a multi-agent route finding algorithm, i.e., MAPF, to establish and assign a route on the rail system from the current location of container handling vehicles to their assigned task at their destination location. b. A step of determining how far a container handling vehicle can travel on the first part of an assigned route within a set time interval, wherein the first part is shorter than the assigned route to the destination. c. A step of fixing the first part of the assigned route so that container handling vehicles can travel within the set time interval. d. A step of instructing container handling vehicles to move from their current location to their destination on an assigned fixed route. e. A step that repeats steps a) to d).

[0041] According to one embodiment of the method, the upper limit of the set time interval in step b) may correspond to the execution time used by the control system to execute the MAPF algorithm and instruct the container handling vehicle to move along the established route. The distance covered on the rail system by the container handling vehicle along the assigned route within this time interval is, in fact, a fixed, complete route established by MAPF.

[0042] According to one embodiment of the method, the upper limit of the set time interval in step b) may correspond to the elapsed time relating to all container handling vehicles moving from their current position to a position on the rail system located above the nearest storage column from which they can change direction.

[0043] According to one embodiment of the method, the duration of the set time interval in step b) may be set according to the size of the automated storage and / or retrieval system and the number of container handling vehicles to be included in the MAPF algorithm.

[0044] According to one embodiment of the method, the method may include extending a fixed portion of an assigned route with respect to a container handling vehicle that has been given priority.

[0045] According to one embodiment of the method, the method may include extending a fixed portion of an assigned route with respect to a container handling vehicle that does not respond to an order.

[0046] According to one embodiment of the method, a fixed portion of the assigned route can be extended along a route established by the MAPF algorithm from the current location of the container handling vehicle to the task at the destination location.

[0047] According to one embodiment of the method, fixed routes can be excluded from the consideration of available routes while the MAPF algorithm is running.

[0048] According to one embodiment of the method, the gaps between each route established by the MAPF algorithm can be adapted to the type of container handling vehicle traveling on the assigned route.

[0049] According to one embodiment of the method, the movement of container handling vehicles can be controlled to avoid possible queuing problems at the destination location.

[0050] According to one embodiment of the method, container handling vehicles can be redirected if queuing problems are anticipated at the destination location.

[0051] According to one embodiment of the method, a container handling vehicle can be redirected to a location or route where no container handling vehicles are present or are not routed to.

[0052] According to one embodiment of the method, the movement of a container handling vehicle can be controlled according to the current load being transported and the state of the container handling vehicle.

[0053] The present invention is further defined by a control system for controlling the routing and rerouting of container handling vehicles that handle storage containers in an automated storage and retrieval system comprising a skeletal structure that forms a three-dimensional storage grid structure for storing storage containers in storage columns, the skeletal structure comprising a grid-based rail system arranged above the storage columns, the rail system providing available routes for container handling vehicles to and from the storage columns, each container handling vehicle comprising a first set of wheels configured to move the vehicle along a first lateral direction (X) of the grid-based rail system and a second set of wheels configured to move the vehicle along a second lateral direction (Y) of the grid-based rail system, the second direction (Y) being perpendicular to the first direction (X). The movement of the container handling vehicles is controlled by a control system that determines what tasks should be performed by a designated container handling vehicle, the location of the destination for performing the tasks, and what routes on the rail system the container handling vehicles should travel.

[0054] The control system comprises a routing planner for finding the optimal route for container handling vehicles, a database for tracking stored containers, and a master controller connected to a transmitter / receiver. The control system is adapted to perform the methods described above and communicate commands to each container handling vehicle.

[0055] The present invention is further defined by a computer program product, which, when executed by a processor, implements the methods described above for efficient routing and rerouting of container handling vehicles that handle storage containers in an automated storage and retrieval system, in a control system for an automated storage and retrieval system. The present invention provides, for example, the following: (Item 1) A method for routing and rerouting container handling vehicles (201) that handle storage containers (106) in an automated storage and retrieval system comprising a skeletal structure (100) forming a three-dimensional storage grid structure (104) for storing storage containers (106) in storage columns (105), wherein the skeletal structure includes a grid-based rail system (108) arranged above the storage columns (105), the rail system (108) providing available routes for the container handling vehicles (201) to handle the storage containers (106) and to and from the storage columns (105), and each container handling vehicle (20 1) comprises a first set of wheels configured to move the vehicle along a first lateral direction (X) of the grid-based rail system, and a second set of wheels configured to move the vehicle along a second lateral direction (Y) of the grid-based rail system (108), wherein the second direction (Y) is perpendicular to the first direction (X), and the movement of the container handling vehicle is controlled by a control system (500) that determines what task should be performed by a designated container handling vehicle (201), the location of the destination for performing the task, and what route on the rail system (108) the container handling vehicle (201) should travel. The method described above includes the following steps performed by the control system (500): a. Step (410) of running a multi-agent route finding algorithm, i.e., MAPF, in the control system (500) to establish and assign a route on the rail system (108) from the current location of the container handling vehicles (201) to the assigned task at the destination location, b. A step (420) of determining how far the container handling vehicle (201) can travel on the first portion of the assigned route within a set time interval, wherein the first portion is shorter than the assigned route to the destination, c. A step (430) of fixing the first portion of the assigned route that the container handling vehicle (201) can travel within the set time interval, d. A step (440) of instructing the container handling vehicles (201) to move from their current location to the end location on the assigned fixed route, e. A step that repeats steps a) to d) A method characterized by including (Item 2) The method according to item 1, wherein the upper limit of the set time interval in step b) corresponds to the execution time used by the control system (500) to execute the MAPF algorithm and instruct the container handling vehicle (201) to move according to the established route. (Item 3) The method according to item 1 or item 2, wherein the upper limit of the set time interval in step b) is the elapsed time relating to all container handling vehicles (201) moving from their current position to a position on the rail system (108) located above the nearest storage column (105) from which they can change direction. (Item 4) The method according to any of the above items, wherein the duration of the set time interval in step b) is set according to the size of the automated storage and / or retrieval system and the number of container handling vehicles (201) to be included in the MAPF algorithm. (Item 5) A method of any of the above items, comprising extending the fixed portion of the assigned route with respect to a container handling vehicle (201) that has been given priority. (Item 6) A method of any of the above items, comprising extending the fixed portion of the assigned route for a container handling vehicle (201) that does not respond to an order. (Item 7) The method according to item 5 or item 6, wherein the fixed portion of the assigned route is extended along the route established by the MAPF algorithm from the current location of the container handling vehicle (201) to the task at the destination location. (Item 8) A method according to any of the above items, wherein fixed routes are excluded from the consideration of available routes while the MAPF algorithm is running. (Item 9) The method according to any of the above items, wherein the gaps between each route established by the MAPF algorithm are adapted to the type of container handling vehicle (201) traveling along the assigned route. (Item 10) The method of any of the items wherein the movement of the container handling vehicle (201) is controlled to avoid possible queuing problems at the destination location. (Item 11) The container handling vehicle (201) is reoriented if queuing problems are anticipated at the destination location, according to the method of item 10. (Item 12) The method described in item 11, wherein a container handling vehicle (201) is redirected to a location or route to which a container handling vehicle (201) does not exist or is not routed. (Item 13) The method according to any of the items, wherein the movement of the container handling vehicle (201) is controlled according to the type of container handling vehicle (201), the current load being transported, and the state of the container handling vehicle (201). (Item 14) A control system (500) for controlling the efficient routing and rerouting of container handling vehicles (201) that handle storage containers (106) in an automated storage and retrieval system comprising a skeletal structure (100) forming a three-dimensional storage grid structure (104) for storing storage containers (106) in storage columns (105), wherein the skeletal structure (100) includes a grid-based rail system (108) arranged above the storage columns (105), the rail system (108) providing available routes for the container handling vehicles (201) to handle and transport the storage containers (106) to and from the storage columns (105), and each The container handling vehicle (201) comprises a first set of wheels configured to move the vehicle along a first lateral direction (X) of the grid-based rail system, and a second set of wheels configured to move the vehicle along a second lateral direction (Y) of the grid-based rail system (108), wherein the second direction (Y) is perpendicular to the first direction (X), and the movement of the container handling vehicle is controlled by the control system (500) which determines what tasks should be performed by a designated container handling vehicle (201), the location of the destination for performing those tasks, and which route on the rail system (108) the container handling vehicle (201) should travel. The control system (500) comprises a routing planner (200) for searching for the optimal route for the container handling vehicles (201), a database (210) for tracking the storage containers (106), and a master controller (220) connected to a transmitter / receiver (225), wherein the control system (500) is adapted to perform the methods described in items 1 to 13 and to communicate commands to each container handling vehicle (201). (Item 15) A computer program product that, when executed by a processor in the control system (500) of an automated storage and retrieval system (1), implements the methods described in items 1 to 13 for efficient routing and rerouting of container handling vehicles (201) that handle storage containers (106) in the automated storage and retrieval system. [Brief explanation of the drawing]

[0056] (Brief explanation of the drawing) The following drawings are provided to facilitate understanding of the present invention. The drawings illustrate embodiments of the present invention, which are described herein only as examples.

[0057] [Figure 1] Figure 1 is a perspective view of the skeletal structure of a conventional automated storage and retrieval system.

[0058] [Figure 2] Figure 2 shows an example of a control system for controlling a container handling vehicle operating on an automated storage and retrieval system.

[0059] [Figure 3] Figure 3 is a flowchart illustrating different steps for efficient routing and rerouting of container handling vehicles in an automated storage and retrieval system.

[0060] [Figure 4] Figure 4 illustrates examples of simple routing using the previous and new routing methods.

[0061] (reference) 1 - Automated storage and retrieval system 100 - Skeletal structure 102 - Upright members of the skeletal structure 103 - Horizontal members of the skeletal structure 104 - Storage grid structure 105 - Storage column 106 - Storage container 106' - Specific location of the storage container 107 - Stack 108 - Rail System 110 - Parallel rails in the first direction (X) 110a - First rail in the first direction (X) 110b - Second rail in the first direction (X) 111 - Parallel rails in the second direction (Y) 111a - First rail in the second direction (Y) 111b - Second rail in the second direction (Y) 112 - Access opening 119 - First port column 119' - Port 1 120 - Second port column 120' - Second Port 200 - Routing Planner 201 - Container handling vehicle 210 - Database 220 - Master Controller 225 - Transmitter / Receiver 230 - Vehicle Controller X - First direction Y - Second direction Z - Third direction 500 - Control System [Modes for carrying out the invention]

[0062] (Detailed description of the invention) The present invention will be described in more detail below with reference to the accompanying drawings. However, it should be understood that the drawings are not intended to limit the present invention to the subject matter depicted therein.

[0063] A typical conventional automated storage and retrieval system 1 with a skeletal structure 100 was described in the background section above with reference to Figure 1.

[0064] It should be understood that the skeletal structure 100 can be of any size and can be significantly wider and / or longer and / or deeper than the task disclosed in Figure 1. For example, the skeletal structure 100 may have a horizontal range of more than 700 × 700 storage columns 105 and a storage depth for storing more than 12 stacked storage containers 106, which are handled by hundreds of container handling vehicles 201 running on a rail system 108.

[0065] Furthermore, the storage grid 104 can be significantly deeper than that disclosed in Figure 1, which shows a stack of eight storage containers 106. For example, the storage grid 104 may be designed to hold twelve stacked storage containers 106.

[0066] The container handling vehicle 201 may be any type known in the art, for example, any one of the automated container handling vehicles disclosed in WO2014 / 090684 A1, NO317366, or WO2015 / 193278A1.

[0067] A rail system 108, arranged across the top of the skeletal structure 100, allows a container handling vehicle 201 to move horizontally between the storage columns 105 and the ports with which they interact (i.e., ports to which storage containers 106 are instructed to be delivered or retrieved).

[0068] The automated storage and retrieval system 1 comprises a skeletal structure 100 that forms a three-dimensional storage grid structure 104 with a grid-based rail system 108 for storing storage containers 106 in storage columns 105, the grid-based rail system 108 arranged above the storage columns 105 with the rail system 108, providing available routes for container handling vehicles 201 to handle and transport the storage containers 106 to and from the storage columns 105, each container handling vehicle 201 along the first lateral direction (X) of the grid-based rail system The container handling vehicle comprises a first set of wheels configured to move the vehicle and a second set of wheels configured to move the vehicle along a second lateral direction (Y) of a grid-based rail system 108, the second direction (Y) being perpendicular to the first direction (X), and the movement of the container handling vehicle is controlled by a control system 500, which determines what tasks should be performed by a designated container handling vehicle 201, the location of the destination for performing the tasks, and what route the container handling vehicle 201 should travel on the rail system 108.

[0069] Figure 2 shows an embodiment of a control system 500 for controlling a container handling vehicle 201 according to an embodiment of the present invention. In this embodiment, the control system 500 comprises a master controller 220, a database 210 used to track the storage containers 106, a routing planner 200 used to find the optimal route for the container handling vehicles 201, and a transmitter / receiver 225 for communicating commands to each container handling vehicle 201. The figure shows that the routing planner 200 is connected to the database 210. However, this may not be the case if the routing planner 200 can receive all the necessary information from the master controller 220.

[0070] The control system 500 typically communicates with a central computer to which commands and tasks are transmitted. The control system 500 further communicates with vehicle controllers 230 in each container handling vehicle 201 to control the traffic flow of the container handling vehicles 201 according to inputs from the routing planner 200. Communication is preferably carried out wirelessly, for example, via radio signals, optical signals.

[0071] Figure 3 illustrates different steps of Method 400 for efficient routing and rerouting of container handling vehicles 201 of the automated storage and retrieval system 1.

[0072] The first step 410 of the method is to run a multi-agent route finding algorithm, i.e., MAPF, in the control system 500 to establish and assign a route on the rail system 108 for container handling vehicles 201 from their current location to their assigned task at their destination location.

[0073] As an example, one task might be to load a designated storage container 106 and bring it to a port located in a different location. When several container handling vehicles 201 receive different tasks and destination locations, it is essential that the routes they follow on the rail system are unique for each container handling vehicle 201, thereby avoiding collisions.

[0074] As stated in the introduction, MAPF is used to find routes used to control the traffic flow of many agents simultaneously without collisions. In this case, using the automated storage and retrieval system 1, the agents are container handling vehicles 201.

[0075] The different modules of the control system 500 may be configured as a single central control unit that performs and controls all the different steps of the method described herein, or as separate units that are interconnected to make up the control system 500. If separate modules are used, as shown in the embodiment of Figure 2, a module named routing planner 200 runs a MAPF for container handling vehicles 201 to search for feasible routes from their current location to their destination and to fix the first part of the route which may be changed. The resulting routes searched for the container handling vehicles 201 are communicated via transmitter / receiver 225 to a master controller 220 that controls each container handling vehicle according to the established route.

[0076] MAPF may be used to find routes for all container handling vehicles 201 operating on a grid-based rail system 108, or for a selected group of container handling vehicles 201.

[0077] For example, with respect to a larger automated storage and retrieval system 1 operated by several hundred container handling vehicles 201, one or more container handling vehicles 201 can be specialized to perform a specific task and to operate in an assigned area along a predetermined route on a rail system 108.

[0078] Another embodiment is one in which a first group of container handling vehicles 201 is controlled according to one MAPF in one area of ​​the rail system 108, while a second or further group of container handling vehicles 201 is controlled according to another MAPF in another area of ​​the rail system 108. To speed up execution, different MAPF algorithms may be executed in parallel by different computer systems.

[0079] An example of a container handling vehicle 201 performing a specially specialized task is an excavation vehicle that, for instance, prepares a specific storage container 106 for further handling by other container handling vehicles 201 by first removing other storage containers 106 stacked on top of a particular storage container 106 stored in a storage column 105. A container handling vehicle 201 performing a specially specialized task may have a different movement pattern and speed compared to other container handling vehicles 201. Therefore, these may be controlled separately, or adjustments to take these differences into account may be made in routing decisions.

[0080] Routes for container handling vehicle 201 are established and assigned after performing the first step 410 of the method, i.e., after running the MAPF algorithm. These routes extend from the current location of container handling vehicle 201 to the assigned task at the destination location.

[0081] The next step in the method 420 is to determine how far the container handling vehicle 201 can travel on the first portion of the assigned route within a set time interval. The duration of the set time interval may be set according to the size of the automated storage and retrieval system 1 and the number of container handling vehicles 201 to be included in the MAPF algorithm.

[0082] According to one embodiment, the set time interval corresponds to the time it takes for the container handling vehicle 201 to move from its current position to a position on the rail system 108 above the nearest storage column 105. Once the container handling vehicle 201 has just passed the position above the storage column 105, it cannot change direction until it reaches the next nearest storage column. The time interval is, in fact, the time it takes to move to the nearest storage column 105, at which point the container handling vehicle 201 may change direction if necessary.

[0083] According to another embodiment, the set time interval corresponds to the execution time used by the control system 500 to run the MAPF algorithm, i.e., the execution time used to establish a route and instruct the container handling vehicle 201 to move along the established route. This provides an optimal routing method because, when the container handling vehicle 201 reaches the end of the fixed route, a new possible route may have already been established by the MAPF algorithm. Thus, the container handling vehicle 201 does not need to stop at the end of the fixed route to receive a new drive command, but can continue and follow the initially assigned route without stopping, or be routed to a new route.

[0084] The next step in the method 430 is to fix the first portion of the assigned route that the container handling vehicle 201 can travel within the set time interval. This means that this portion of the route cannot be changed, while the remainder of the assigned route established by MAPF can be changed to a new route starting from the end point of the fixed route.

[0085] The next step, 440, is to instruct the container handling vehicles 201 to move from their current location to their assigned destination on a fixed route.

[0086] Steps 410–440 are repeated for each container handling vehicle 201 until they complete their assigned tasks. Each time the method is repeated, it may be considered an iteration, and each iteration allows for the rerouting of the container handling vehicles 210 from the routes established by MAPF in the previous iteration.

[0087] According to the method, a container handling vehicle 201 can easily change the type of task or target before completing its initial assigned task, or a new active container handling vehicle 201 can be added to operate on the storage system.

[0088] According to one embodiment of the present invention, if a container handling vehicle 201 is given priority, the fixed route is extended. An embodiment of this is when a port sends a signal that it is ready to receive a storage container 106 from the container handling vehicle 201. In this case, the complete route for the container handling vehicle 201, determined by the MAPF and extending from its current location to the port, is fixed and cannot be changed until the assigned task (e.g., delivering the storage container to the port) is completed.

[0089] According to another embodiment of the present invention, the fixed route is also extended if a container handling vehicle 201 fails to respond to a command from the control system 500. This may be due to a malfunctioning container handling vehicle 201, which may be stopped at some point along its fixed route or fail to respond to a command such as a rerouting command at the end of the fixed route. In that case, it may then continue traveling at some point along its assigned route (i.e., that portion of the assigned route that is not fixed). By fixing the complete route established by the MAPF, other container handling vehicles avoid collisions with it.

[0090] Possible collisions with other container handling vehicles 201 are avoided by extending a fixed route to include the complete route established by the MAPF algorithm (i.e., from the current location of container handling vehicle 201 to the task at the destination location). This is preferable, for example, when a container handling vehicle 201 is given priority or when a port sends a signal that it is ready to interact with the assigned container handling vehicle 201.

[0091] According to one embodiment of the method, fixed routes are excluded from execution in the MAPF algorithm. With respect to container handling vehicles 201 operating normally without priority, this means that their current location is used as input to the MAPF algorithm, i.e., the current location provides one set of destination locations for a fixed route. With respect to container handling vehicles 201 with priority or faulty container handling vehicles 201, the complete established route to the assigned task at the destination location is excluded from further execution in the MAPF algorithm.

[0092] Determining which container handling vehicle should be given priority can be based on several factors. Some tasks have a greater impact on performance than others. A container handling vehicle 201 assigned to deliver storage container 106 within a port is typically given priority over a container handling vehicle 201 assigned to deliver storage container 106 to storage column 105. Typically, the optimal route for the preferred container handling vehicle 201 is created first. Different routing algorithms exist for route prioritization and determining the optimal route. These are based on cost functions that use different sets of input parameters such as time and distance to reach the destination, priority of the assigned task, and expected waiting time at the port.

[0093] As described, the MAPF algorithm is used to establish and assign routes on rail system 108 for container handling vehicles 201 from their current location to the task at the destination location. Inputs to the MAPF algorithm are the arrangement of rail system 108, the container handling vehicles 201 in operation, and the current and destination locations of each container handling vehicle 201.

[0094] According to one embodiment of the method, an additional input to the MAPF algorithm is, for example, the required gap between each route established by the MAPF algorithm to provide sufficient clearance to avoid collisions with margin. The required gap is adapted to the type of container handling vehicle 201 used for the task and the type of rail used on the route to the task at the destination location.

[0095] A double-track configuration allows two container handling vehicles to pass each other on rails above two adjacent storage columns 105. A single-track configuration does not allow a container handling vehicle 201 to pass if it is not possible to move an obstructing container handling vehicle out of the way, and may require a passing gap between two passing container handling vehicles 201 that corresponds to at least the area occupied by one storage column.

[0096] According to one embodiment of the method, the movement of the container handling vehicles 201 is further controlled to avoid potential queuing problems at the destination locations. This can be done in different ways. For example, if several container handling vehicles 201 are expected to arrive at the same port at approximately the same time, they may be ranked, the driving speed of the lowest-ranked container handling vehicle 201 may be reduced, or they may be stopped off-course at some location until the first-ranked container handling vehicle 201 has completed its task at the port. Another way to avoid anticipated queuing problems is to redirect the container handling vehicles, for example, to deliver a storage container 106 at a different port, to complete an assigned task at another destination location.

[0097] Different container handling vehicles exist, each with different specifications such as acceleration, speed, and maximum load. Additional inputs to the MAPF algorithm may provide details such as the type of container handling vehicle 201 assigned to perform the task, the current load on the container handling vehicle 201 based on the weight of the storage container 106 being transported, and the state of the container handling vehicle 201, such as low battery, worn parts, and wheel control force. Such additional inputs may affect how the movement of the container handling vehicle 201 is controlled and whether the gap between vehicles needs to be increased.

[0098] Figure 4 illustrates a simple example of how routes are established according to the old routing method and how routes are established according to the improved routing method described above.

[0099] R1 and R2 indicate the current locations of container handling vehicles 1 and 2. D1 and D2 indicate the destination locations for container handling vehicles 1 and 2.

[0100] In this embodiment, robot 1 at R1 receives a task immediately before robot 2 at R2 receives the task. According to the previous routing system, the route extending from robot 1's current location R1 to its destination D1 is created and fixed first. Then, a route for robot 2 is created, which is made so as not to conflict with robot 1's established fixed route. As can be seen from Figure 4, the route for robot 2 is not optimal in terms of distance traveled because it is necessary to avoid grid space D1 (and possibly any other fixed grid space on the assigned route from R1 to D1).

[0101] According to the improved routing method disclosed herein, only the first portion of the established route for container handling vehicles is fixed, while the last portion of the established route is a planned route that can be changed. This means that when container handling vehicles reach the end of the fixed route and the MAPF algorithm is run again, they can be rerouted to follow a new route from their current starting position at the end of the fixed route.

[0102] Therefore, assuming that robot 1 has not yet reached the endpoint of the first fixed section when the route for robot 2 is established by the MAPF algorithm, the final section of the route for robot 1 is not fixed, resulting in an optimal route for both container handling vehicles.

[0103] This new routing method is flexible and efficient, considering all routes for all container handling vehicles 201 to their assigned destinations at multiple stages while a given container handling vehicle is traveling along its assigned route.

[0104] The present invention is further defined by a control system 500 for controlling the efficient routing and rerouting of container handling vehicles 201 that handle storage containers 106 in an automated storage and retrieval system 1, as described above with reference to Figure 1.

[0105] The control system 500 is described above with reference to Figure 2. It further comprises a processor arranged to run a computer program that, when executed, carries out the method described above with reference to Figure 3, thereby enabling efficient routing and rerouting of container handling vehicles 201 that handle storage containers 106 in the automated storage and retrieval system 1. The computer program may run a processor in the routing planner 200.

Claims

1. A method for routing container handling vehicles (201) in an automated storage and retrieval system, wherein the movement of the container handling vehicles is controlled by a control system (500), The above method is performed by the control system (500) by the following steps, namely, a. Step (410) of running a multi-agent route planning (MAPF) algorithm in the control system (500) to establish and assign a route for the container handling vehicles (201) from their current location to the assigned task at the destination location, b. A step (420) of determining the distance that the container handling vehicle (201) can travel within a set time interval, wherein the first portion of the assigned route is shorter than the assigned route to the destination, c. A step (430) of fixing the first portion of the assigned route that the container handling vehicle (201) can travel within the set time interval, wherein the fixed route is excluded from the consideration of available routes while the MAPF algorithm is running. d. A step (440) of instructing the container handling vehicles (201) to move from their current location to the end location on the assigned route, e. A step that repeats steps a) to d) Methods that include...

2. The method according to claim 1, wherein the set time interval in step b) corresponds to the execution time used by the control system (500) to execute the MAPF algorithm and instruct the container handling vehicle (201) to move according to the established route.

3. The method according to claim 1 or 2, wherein the set time interval in step b) is the elapsed time relating to all container handling vehicles (201) moving from their current position to a position on the rail system (108) located above the nearest storage column (105) from which they can change direction.

4. The method according to any one of claims 1 to 3, wherein the duration of the set time interval in step b) is set according to the size of the automated storage and / or retrieval system and the number of container handling vehicles (201) to be included in the MAPF algorithm.

5. The method according to any one of claims 1 to 4, wherein the method comprises extending the fixed portion of the assigned route with respect to a container handling vehicle (201) that has been given priority.

6. The method according to any one of claims 1 to 5, wherein the method includes extending the fixed portion of the assigned route with respect to a container handling vehicle (201) that does not respond to an order.

7. The method according to claim 6, wherein the fixed portion of the assigned route is extended along the route established by the MAPF algorithm from the current location of the container handling vehicle (201) to the task at the destination location.

8. The method according to any one of claims 1 to 7, wherein the gaps between each route established by the MAPF algorithm are adapted to the type of container handling vehicle (201) traveling along the assigned route.

9. The method according to any one of claims 1 to 8, wherein the movement of the container handling vehicle (201) is controlled to avoid possible queuing problems at the destination location.

10. The method according to claim 9, wherein the container handling vehicle (201) is reoriented if a queuing problem is anticipated at the destination location.

11. The method according to claim 10, wherein the container handling vehicle (201) is redirected to a location or route to which the container handling vehicle (201) does not exist or is not routed.

12. The method according to any one of claims 1 to 11, wherein the movement of the container handling vehicle (201) is controlled according to the type of container handling vehicle (201), the current load being transported, and the state of the container handling vehicle (201).

13. The method according to any one of claims 1 to 12, wherein the control system (500) determines which tasks should be performed by a designated container handling vehicle (201), the location of the destination for performing the tasks, and which route the container handling vehicle (201) should travel on the rail system (108) of the automated storage and retrieval system.

14. The method according to any one of claims 1 to 13, wherein the automated storage and retrieval system comprises a skeletal structure (100) that forms a three-dimensional storage grid structure (104) for storing storage containers (106) in storage columns (105), the skeletal structure includes a grid-based rail system (108) arranged above the storage columns (105), the rail system (108) providing available routes for a container handling vehicle (201) that handles and transports the storage containers (106) to and from the storage columns (105).

15. The method according to claim 14, wherein each container handling vehicle (201) comprises a first set of wheels configured to move the container handling vehicle along a first lateral direction (X) of the grid-based rail system, and a second set of wheels configured to move the container handling vehicle along a second lateral direction (Y) of the grid-based rail system (108), the second lateral direction (Y) being perpendicular to the first lateral direction (X).

16. A control system (500) for controlling the efficient routing of container handling vehicles (201) in an automated storage and retrieval system, wherein the movement of the container handling vehicles is controlled by the control system (500), The control system (500) is configured to perform the method described in any one of claims 1 to 15 and to communicate control commands to each container handling vehicle (201).

17. The control system according to claim 16, wherein the control system (500) comprises a routing planner (200) for searching for the optimal route for the container handling vehicle (201), a database (210) for tracking the storage containers (106), and a master controller (220) connected to a transmitter / receiver (225).

18. A computer program product that, when executed by a processor in a control system (500) of an automated storage and retrieval system (1), implements the method of any one of claims 1 to 15 for efficient routing of container handling vehicles (201) in an automated storage and retrieval system.

Images