Distribution System

Abstract

A distribution system is described that includes a transport surface for distributing objects and a carrier for transporting the objects. A drive system moves the carrier on the transport surface. A control system of the distribution system is configured to control the carrier to move on the transport surface from a start location to a final destination location along a planned route. The control system includes a route selection system configured to calculate a planned route for at least two carriers on the transport surface by modeling the transport surface with nodes and a graph and using a knowledge-based windowed hierarchical collaborative search algorithm. The route selection system is configured to determine a reserved time window and an available time window for each node. The route selection system is configured to assign a separate reservation length to each carrier for its next movement within the available time window, and assigns infinite reservation time to nodes at logical locations.

Description

[Technical Field]

[0001] The present invention relates to systems, methods, and devices for controlling the movement of carriers for transporting sample containers, particularly tubes filled with biological fluids to be analyzed, reagents, cassettes filled with reagents, specimen slides, tissue material, waste, or consumables such as pipette tips, tube caps, or empty tubes for aliquoting in diagnostic laboratories. The systems, methods, and devices are also applicable to controlling the movement of carriers capable of transporting payloads such as articles, goods in a warehouse, products manufactured on a manufacturing floor, or other objects over a transport surface. [Background technology]

[0002] Distribution systems of this type are known in the art.

[0003] Laboratory diagnostic dispensing systems are described, for example, in EP 3 095 739 or WO 2012 / 158541. These publications describe laboratory sample dispensing systems with passive or self-propelled carriers on a transport surface.

[0004] For example, WO 2008 / 148513 describes a storage or transport system comprising one or more transport devices, in particular an electromagnetic planar motor, which moves movable pallets, cars or containers mounted on wheels or rollers.

[0005] WO 2017 / 186825 describes a robotic parcel handler coordination system comprising a robotic parcel handler that, in use, can move between a number of cells arranged in a grid.

[0006] Route selection algorithms used in route selection systems are also known in the art. "Cooperative pathfinding" (Silver, D., 2005, "Cooperative pathfinding," Young, R.M., and Laird, J.E., eds., AIIDE, pp. 117-122. AAA I Press) discloses algorithms for solving the so-called multi-agent pathfinding problem, in which agents must find routes to distinct destinations given all information about the routes of other agents. These algorithms are suitable for use in real-time strategy games and other real-time environments. The proposed algorithms assume a grid to represent the space and use the so-called Manhattan distance as the basis for calculating a cost function. They use a windowed search with a limited, fixed cooperative search depth, called Windowed Hierarchical Cooperative A* (WHCA*). The drawbacks of these algorithms are also discussed. Certain types of problems cannot be solved by decoupled greedy algorithms that precompute optimal paths; that is, these problems are considered to lead to infinite computational loops.

[0007] B. Zahy et al., "Conflict-Oriented Windowed Hierarchical Cooperative A* Algorithm," 2014 IEEE International Conference on Robotics and Automation (ICRA), IEEE, May 31, 2014, pages 3743–3748, XP03265061 8, DOl:10.1 109 / ICRA.2014.6907401, describes a multi-agent path-finding problem (MAPF) that considers a map and a group of agents with distinct source and goal locations. MAPF solvers can be divided into several classes based on their objectives. One of these classes is the online MAPF algorithm, in which path search is interleaved with the actual physical movement of the agents. A prominent algorithm in this class is the windowed hierarchical cooperative A* algorithm (WHCA*), in which paths are planned individually for each agent and cooperation is achieved using a reservation table.

[0008] definition A carrier is a support structure for supporting and transporting a payload. A carrier is in particular equipped with suitable holding means for supporting and, if necessary, fixing the object in the required manner and orientation. A carrier may be self-propelled or may be propelled by and move on a transport surface.

[0009] A transfer surface is any type of two-dimensional plane, surface, floor, bed, layer, platform, or platform on which carriers can move in at least two dimensions. For example, a transfer surface can be a slideway mounted in a diagnostic laboratory or the ground in a manufacturing facility or production hall. Transfer surfaces can be vertical or horizontal, including inclined surfaces. Curved transfer surfaces are also possible. Carriers can contact the surface, allowing for frictional drive and controlled movement, or they can move without contact, for example, by air or magnetic levitation, which creates a small gap between the carrier and the surface. For the third dimension, the surface is formed with corresponding limits on upward and downward incline, or some type of levitation mechanism, such as magnetic levitation or air cushion technology, is installed with a corresponding limit on the height that can be reached without losing control. For vertical transfer in the third dimension, an elevator / paternoster mechanism can also be installed.

[0010] A drive system moves the carrier on the transport surface. The drive system may be implemented on the carrier itself, for example, as wheels connected to an electric motor with a battery and electronics. Another possibility is a linear motor. Passive carriers are also possible. For example, a magnetic device is fixed in the carrier, and a magnetic force provided by a magnetically active drivable element, such as an electromagnetic coil, generates an electromagnetic field to move the carrier. The coil can be located below, above, to the side of, or within the transport surface. For example, the arrangement of a magnetic coil below the transport surface is described, for example, in EP 2 566 787 or WO 2013 / 098202.

[0011] Logical positions are defined on the transport surface as locations where the carrier can stop, start, and / or change direction. In systems such as those described in EP 2566787 or WO 2013 / 098202, the drive system defines these logical positions through hardware limitations. In these systems, logical positions are typically defined above electromagnetic coils. At these locations, the carrier can stop and change direction on its next move. In other transport surfaces, logical positions can be defined as desired or necessary to form a useful set of intersections, junctions, start positions, and stop positions. Logical positions can be virtual positions in a route selection algorithm and / or locations on a real transport system. Logical positions can be discrete locations where the carrier can stop. In particular, logical positions can be defined by at least one physical entity of the drive system, such as an electromagnetic coil or a possible path intersection, such as a rail.

[0012] The logical location may be any location that a passive carrier can reach, or any location where the carrier can change direction, stop, or be identified by an identification or registration system. The identification and registration system may be a camera system or optical sensors and scanners that identify size, type, or any optical signature on the carrier or object, such as a barcode. Alternatively or additionally, an RFID reader system that reads the unique RFID of the carrier or object on the carrier, or a sensor inside the transport surface, may be used to identify the logical location and locate the carrier. A further option may be high-precision GPS, especially enhanced with Wi-Fi and / or GSM signals.

[0013] For route selection systems with partially stochastic behavior, i.e., carriers do not always move with the same velocity profile (acceleration, deceleration, or constant velocity variations), planning too far into the future increases the computational effort and the risk of the route becoming prone to failure. Furthermore, new instructions to transport a carrier from its starting location to its final destination location are received while other carriers' route execution is still in progress. Therefore, to increase the efficiency of the route selection system, the number of moves related to planned ahead in time can be limited. In addition, the computation time is no longer proportional to the size of the transport system. The maximum time a router plans ahead is called the cooperative search window, which has a time length T.

[0014] For each logical location, a time window is defined in which the state of that location is either reserved or free for the possibility of carrier movement. If a time window is free, it can be used to plan carrier movement at the corresponding time. Therefore, for each logical location, a time sequence is defined consisting of individual free and reserved time windows with individual durations. These durations of the time windows are discrete and continuous in the sense that the window durations for each logical location can be planned individually and their values ​​are freely definable values ​​rather than a number of discrete time steps. Therefore, a continuous duration is defined, in contrast to the discretized time windows typically known in the art.

[0015] The route selection algorithm is an algorithm that calculates a route for each carrier on the transport plane from its starting location to an intermediate destination on the way to its final destination.

[0016] The starting position is the logical position on the transport surface where the carrier is located when the algorithm begins calculating the route.

[0017] A final destination location is a logical location on the transport surface to which a carrier should go. A final destination location is a logical location on the transport surface that has, among other things, a special function, such as a logical location where a sample tube, a portion of a sample, or a consumable is transferred between the transport surface and an analyzer or a pre-analysis or post-analysis system or a storage system. In the case of a manufacturing site, a final destination location is, among other things, a logical location corresponding to a machine station that performs some manufacturing process on a semi-finished product. The starting location of one carrier may, among other things, be the final destination location of another carrier, and even more particularly, of the same carrier.

[0018] The route selection algorithm typically calculates several linear movements for each route starting from the carrier's current position on a logical location as the starting location and leading to intermediate destination locations, which are also logical locations. Each movement can have a start and a stop at a logical location. The stop of the last movement of the route is either an intermediate destination or a final destination.

[0019] A node is defined as a logical location along with time window information (booked or free) used by the windowed hierarchical cooperative search algorithm. A graph is a structure of nodes and possible connections between nodes used by the windowed hierarchical cooperative search algorithm.

[0020] Typically, a distribution system includes two parts. A transport system that executes the movement of the carriers as well as possible according to the given route calculated by the router software, and communicates all position changes or predefined position changes of the carriers as well as potential errors in the transport system to the control system. The software system receives status updates regarding the carrier's position and errors on the logical location, and calculates a new route including the following moves based on the status updates and new requests to transport the carrier to its final destination. These moves are sent to the transport system for execution.

[0021] If a problem occurs in the distribution system, for example, the plan may not be able to be executed because a location is inaccessible, or the plan may be delayed if the movement takes longer to execute than planned. In both situations, the execution of the plan is defined as "failed."

[0022] A windowed hierarchical cooperative informed search algorithm is a search algorithm using knowledge, such as the A* or D* search algorithm. For each carrier with a final destination, the search algorithm calculates a route from the starting location to an intermediate destination on nodes within a cooperative search window of time length T toward the respective final destination location. The search is cooperative because the route can only pass through the free time window of the logical location. For logical locations with a required free time window, the required duration of the free time window changes from "free" to "reserved" for the required time slot of the respective logical location. Therefore, the free time window is divided into a reserved time window and one or two additional free time windows. Therefore, the search is cooperative with respect to the cooperative time window T by respecting the reserved time windows for other carriers of the logical location.

[0023] The route selection system is configured to calculate a planned route using a knowledge-based windowed hierarchical collaborative search algorithm within a collaborative time window T. The knowledge-based windowed hierarchical collaborative search algorithm can be designed based on a realistic time travel model to reserve time windows in continuous time. The realistic time travel model can be an approximation. The realistic time travel model can be configured to allow substantially all trips to be performed within the planned time, particularly without wasting time by reserving too much time. Information for the realistic time travel model can be obtained from measurements in a real system. The realistic time travel model can include consideration of optimistic and / or pessimistic travel times. Optimistic and pessimistic travel time models refer to models that predict earlier and later arrivals than reality, respectively. The realistic time travel model can include consideration of certain accelerations, decelerations, and plateau speeds. Model parameters related to these can be measured. For small trips with only a few positions, plateau speeds may not be reached and only accelerations and decelerations may exist. The model parameters can be selected from measurements, and therefore realistic values. The model parameters can be adjusted to be pessimistic. This can ensure that there is substantially always enough time to completely execute a move. Model parameters can be adjusted to plan more time for moves. Model parameters can be adjusted automatically. Model parameters can be automatically adjusted if too many plans fail due to insufficient time or if too much time is reserved too frequently, resulting in carriers having to wait too long between moves. For example, if a short move fails, acceleration and deceleration can be modified. For example, if a long move fails, maximum speed can be adjusted to a lower value. For example, if the number of moves that follow the plan is ≧99%. Model parameters can be automatically adjusted by evaluating single position moves and / or longer position moves.For example, even a single-position movement can be evaluated if the acceleration and deceleration are fast enough to affect the maximum speed. If such a movement is often slower than a pessimistic estimate or faster than an optimistic estimate, the router can adjust the maximum speed. Specifically, the route selection system is configured to calculate a planned route using a knowledge-based windowed hierarchical cooperative search algorithm within a cooperative time window T, taking into account continuous time durations. In contrast, a knowledge-based windowed hierarchical cooperative search algorithm can reserve time windows by using discrete times. Therefore, the present invention proposes the use of a modified version of the windowed hierarchical cooperative A* (WHCA*) algorithm, referred to herein as continuous WHCA*. Continuous durations can make the algorithm more efficient because the transportation system does not use discrete time frames for operation, allowing movements to begin at any time. Furthermore, considering continuous durations can reduce computational requirements. In state-of-the-art discretized time, the discretized time is stored in a fine-grained manner, for example, 10 ms (a 2-second journey results in 200 time segments). In memory, this can lead to storing a huge array of very small time intervals. Using continuous durations avoids coarse discretization, over-reservations, and time waste. Route selection systems are used in real transport systems that face deterministic or partially stochastic travel speeds. In the presence of stochastic effects, it may not be possible to accurately predict the time required for each journey. Route selection systems can be configured to reserve time windows based on realistic travel time models. In contrast, known WHCA*s are deterministic and cannot model probabilities.In this way, the continuous WHCA* algorithm can enable route selection for processes with partially stochastic time behavior by modeling optimistic and pessimistic arrival times for each time window in such a way that most or all of the actual trips, including stochastic variations, occur within these two estimates, thereby enabling the optimal-case start and pessimistic end of subsequent trips to be planned accordingly.

[0024] The continuous WHCA* algorithm can be designed to plan each carrier's route independently, and cooperation can be achieved using a reservation table. The route selection system is configured to assign infinite reservation times to logical positions on the transportation plane. The use of infinite time windows can ensure that the reservation table is always contention-free. Contention-free can refer to no collisions occurring because a logical position can only be assigned to at most one carrier at a time. The continuous WHCA* algorithm can be designed to allow all sub-paths to end with infinite reservations to ensure that all carriers always have contention-free paths in space-time.

[0025] A continuous WHCA* can be designed to plan carrier routes using the optimal departure time of each carrier independently. In contrast to discrete time, this can avoid computational spikes because the start of a trip does not have to occur at discrete times, and therefore multiple carriers cannot depart at the same time.

[0026] The continuous WHCA* may be suitable for providing routes for transport systems, such as sample transport systems in in-vitro diagnostic laboratories, in an efficient and dynamically responsive manner. For example, the continuous WHCA* may enable one or more of the following functions: creating and deleting carriers in the WHCA* algorithm at any time, for example, when a carrier is placed on the transport system from an analyzer, or vice versa, in a real system; changing the final destination of a carrier at any time, for example, when an analyzer is no longer available and a new analyzer is assigned; considering stochastic variations in travel time, for example, due to the influence of friction between the carrier and the transport surface; considering the possibility of a travel failure, for example, due to damage to a transport element; considering the possibility of failure to execute a travel according to a planned time, for example, due to wear or dirt on the transport element; and considering the possibility of a sudden inability to pass a position, for example, due to malfunction or intentional shutoff. The execution unit may be configured to provide sudden obstacles detected by a sensor to the route selection system, which may cancel the affected plans and re-plan them.

[0027] A distribution system can be configured to distribute a large number of carriers, such as more than several thousand carriers. Carrier density can become extremely high, to the point where many carriers are blocked by neighboring carriers and cannot move until traffic density decreases. Continuous WHCA* can be designed to plan contention-free routes that take into account carrier density.

[0028] An intermediate destination is a logical location that can be reached after a time T. An intermediate destination can also coincide with a final destination if the final destination is reachable within T.

[0029] As the carrier moves, it accelerates until it reaches a steady state velocity, after which it decelerates and stops at an intermediate or destination position. For very short moves, such as one logical position, there may only be an acceleration and a deceleration immediately following the acceleration. In such cases, the steady state velocity is never reached.

[0030] The term "infinite reservation time" may refer to an additional state of a logical location that is used by a router to indicate that the location is not available for route selection. Initially, all locations may be available. If a location is set to "reserved" for a finite duration, the location will become available for travel again once the temporary reservation ends. If a location is set to "infinite reservation time" from a specific time, the router will no longer consider the location in planning from that time on. However, until the specific time, the router will also consider the location.

[0031] An infinite reservation is an entry in the reservation table that starts and / or ends at minus infinity and plus infinity, respectively. It can be used by a router to indicate that a location is unavailable for route selection. If a location is set to "reserved" for a finite duration, the location becomes available for travel again once the temporary reservation ends. If a reservation starts and ends at infinity, the router no longer considers the location in planning. If a reservation starts at a finite time and ends at infinity, the router can only use this location during the available time window before the start of the infinite reservation. Summary of the Invention

[0032] It is an object of the present invention to provide a distribution system, a method for controlling the distribution system, and an apparatus for implementing the distribution system, which allows for improved, more efficient and reliable transport in the distribution system.

[0033] This object is achieved by a system according to claim 1.

[0034] A first aspect of the present invention relates to a distribution system including a transport surface for distributing a plurality of objects. A plurality of carriers transport the objects on the transport surface. A drive system moves the carriers on the transport surface. The system further includes a control system configured to control the drive system so that the carriers move on the transport surface from a starting position to a final destination position along a planned route. The control system includes a route selection system. The route selection system is configured to calculate routes for all carriers on the transport surface by modeling the transport surface as a graph of nodes. To do so, the route selection system is configured to determine reserved time windows and free time windows for each node. To calculate the carrier routes, the route selection system uses a knowledge-based windowed hierarchical cooperative search algorithm within a cooperative time window T. In particular, the knowledge-based cooperative search algorithm is the Dijkstra algorithm or the Bellman-Ford algorithm, or more particularly the A* algorithm. T is typically in the range of 1 to 300 seconds, and is particularly about 10 seconds. The route selection system is configured to assign individual reservation lengths, as the number of nodes for the next movement, for logical locations having an available time window for each carrier so that the carriers depart and stop individually. An infinite reservation time is assigned to one logical location on the transport surface. An infinite reservation time can be assigned to a logical location on the transport surface if no movement is planned for the carrier at that location or if the location is marked as unavailable. The control system comprises at least one execution unit configured to execute the planned route and transport the carriers from their respective start locations to their respective final locations.

[0035] Calculating the planned route can include planning the route. The control system can include a route selection system for planning the carrier's route. The planning can include determining an optimal route for the carrier across the transfer surface. The optimal route can be determined with respect to at least one optimization objective, such as one or more of time, resource consumption, cost, wear balance, and good overall transfer performance. Because the route selection is collaborative, the route selection system can ensure that all carriers are transported in a net-efficient manner, sacrificing, if necessary, the shortest time to each individual carrier's final destination. The optimal route for each carrier can be a route selected from multiple possible routes that minimize an optimization objective, such as the time required to reach a final destination. The planning can consider not only one carrier moving across the transfer surface, but also multiple additional carriers present and / or obstacles and / or the failure of at least one physical entity. Planning the route can include receiving status updates regarding the transfer surface and / or drive system. The status update information may be information about a logical location, such as the presence or absence of an obstacle, a carrier passing the logical location, a carrier starting / ending a movement, or a failure of a physical entity corresponding to the logical location. Fault information may be sent separately to the control system, which can inform the route selection system to remove a particular location for route selection or reserve a location with an infinite reservation time. The route selection system may be configured to identify a congestion or stalemate by detecting no movement for a particular time. The route selection system may be configured to calculate, and in particular compute, a route for the carrier according to the received status update information. The route selection system may be designed as a computer or computer network for performing route planning.

[0036] The route selection system may be configured to propose a route for the carrier to be executed by the execution unit. The control system may include an execution unit for executing the planned route. The route selection system and the execution unit may be located on the same computer or embodied as separate devices. The control system may include at least one communication connection for transmitting information about the planned route from the route selection system to the execution unit. For example, the communication connection may be a wired connection or a wireless connection, and may be designed, for example, as an Internet connection, a Bluetooth connection, an NFC connection, an inductive coupling, etc.

[0037] Specifically, the execution unit initiates the planned route. The execution unit may be designed as a computer or computer network specifically for executing the execution process for moving carriers. The planned route can be generated so that collisions do not occur. The distribution system can be configured to prohibit collisions by selecting a good route by the following mechanism: the execution unit sends information to the route selection system about carriers that have arrived at a specific point, such as passing a module or area, reaching the end of a movement, etc. Thus, the route selection system knows that reserved positions already passed are safe and can be used for the next carrier without risk of collision. Optionally, the execution unit can also take care to prevent carrier collisions. If the execution unit (hypothetically) receives an incorrect plan that could result in a collision, it can be configured to pause without doing anything further. In this way, the execution unit can be equipped with additional safety mechanisms.

[0038] The execution unit can attempt to execute a planned route by moving the carrier. Transport methods in which the carrier is driven by a combination of electromagnetic force and friction exhibit stochasticity in the velocity profile of the moving carrier. In general, friction forces between sliding surfaces can be a significant variable for different carriers and even over time. Furthermore, the surface of the carrier and the transport sliding surface may not be uniform across the entire surface. For example, the roughness of the transport surface may vary due to differences in wear, uneven materials, dirt or dust, or moisture. Also, some module surface plates may be more recently updated than others and exhibit different friction levels than older plates. The same is true for the carrier's contact surface with the transport surface. As a result of these variable friction forces, the time required for transport may also have an unpredictable stochastic component. If the actual friction is greater than average, the carrier may accelerate slowly and fail to reach an average steady-state velocity. The carrier may also decelerate quickly. The execution unit controlling the movement can attempt to adjust for this, but variations in acceleration and speed profiles are still observed for the movements and carriers. Thus, due to stochastic effects, exceptions, and errors, not all planned routes will be executed exactly as proposed by the route selection system.

[0039] The transport surface may include at least one sensor configured to detect status information of the transport system. For example, the status information may include one or more of the following: a position of the carrier, an unavailable or broken position or node on the transport surface, or dirt on the transport surface. For example, the sensor may be or be part of a camera system and appropriate image processing. Other position detection systems, such as Hall sensors, current sensors, conductivity sensors, capacitance sensors, inductive sensors, or optical barriers, are also possible. The sensor may be configured to provide the detected status information to an execution unit. The execution unit may be configured to generate status update information by updating the latest status information according to the received detected status information. The execution unit may be configured to provide status update information regarding the transport surface, the transport carrier, and / or the drive system to a route selection system.

[0040] The route selection system is further configured to assign infinite reservation times to logical locations on the transfer surface, which avoids infinite computation loops, a known drawback of knowledge-based search algorithms, at the expense of some coordination in route selection.

[0041] In a further embodiment of the distribution system, the route selection system comprises: The logical location is the location of a carrier that does not have a final destination, The logical location is corrupted or blocked, There is no possibility of moving the carrier of the logic position during the cooperation time window T, is configured to assign an infinite reservation time to a logical position when one or more of the following is realized:

[0042] Thus, any one, a combination of two, or all three of these situations are possible for different logical positions on the transfer surface.

[0043] The reservation time assigned to a logical location may be the duration for which a carrier plans and / or expects to travel through this logical location, or may correspond to such a time. The route selection system may reserve one or more locations in its plan for as long as the carrier needs to travel through one or more locations, particularly according to its model. For example, all locations for a trip may remain reserved as long as the complete trip is made. For example, locations may be reserved as long as the trip continues within a particular logical area. For example, locations may be reserved until the carrier passes through the location. For example, locations may be released as soon as the carrier passes each logical location. For example, locations may be released when the boundary of a block of logical locations is passed. The route selection system may be configured to assign different types of reservation times to logical locations depending on whether they are temporarily or permanently unavailable for route planning. The route selection system may be configured to assign a finite reservation time to logical locations that are expected to be temporarily occupied by a carrier. Thus, if a finite reservation time is assigned to a logical location, the logical location may be available for route planning, but its temporary unavailability must be taken into account, for example, by planning a waiting time in the route. The route selection system may be configured to assign infinite reservation time to logical locations that are predicted to be unavailable for carrier movement, for example, because the logical location is the location of a carrier that has no final destination (e.g., a parked carrier), the logical location is damaged or blocked, or there is no possibility of a carrier moving on the logical location during the cooperative time window T. The route selection system may be configured to assign infinite reservation time to logical locations that are permanently unavailable. In particular, infinite reservation time may be assigned to logical locations that are unavailable for route planning. The route selection system may be configured to assign infinite reservation time to logical locations such that the knowledge-based windowed hierarchical cooperative search algorithm can consider the logical locations to be unavailable or unavailable.Specifically, the knowledge-based windowed hierarchical collaborative search algorithm can ignore logical locations that have been assigned infinite reservation times in searching for routes. Otherwise, i.e., in all cases where a logical location has no finite or infinite reservation time assigned to it, the logical location is considered free, specifically in the sense of being available.

[0044] The state of a logical location may change during operation of the distribution system. For example, a logical location that is considered permanently unavailable may change to "vacant" after maintenance. For example, a logical location that is considered vacant may change to permanently unavailable if the physical entities of the transfer surface and / or drive system are damaged.

[0045] For example, the route selection system may be configured to assign an infinite reservation time to a logical position if the logical position is considered to be damaged or blocked. A logical position may be considered damaged if it can no longer be used, for example, due to a drive system failure and / or a transport surface failure at this particular position. A logical position may be considered blocked if the logical position is usable in principle, i.e., the physical entity of this logical position is functioning properly, but the logical position is reserved by a carrier. The route selection system may be configured to take into account failures and / or changes and / or obstructions in the drive system and / or transport surface when planning the route. For example, the presence of other carriers on the transport surface may affect the possible routes of the carrier. For example, the route selection system may be configured to take into account traffic, congestion, reserved logical positions, dirt, etc. For example, the route selection system may be configured to take into account changes in the drive system and / or transport surface, such as a failure of one of the electromagnetic coils. In particular, the route selection system may be configured to prevent a carrier from moving to a logical position that would prevent further transport of the carrier. The route selection system may be configured to take into account failures and / or changes and / or obstructions by assigning infinite reservation times to these logical positions. The route selection system may be configured to assign infinite reservation times to logical positions such that the knowledge-based windowed hierarchical collaborative search algorithm can consider the logical positions to be unavailable or unavailable. The route selection system may be configured to assign infinite reservation times to logical positions occupied by carriers for which no route selection has yet been performed. For example, if a system is restarted and therefore all carriers do not have plans and must be routed simultaneously, the route selection system may assign infinite reservations to positions on all carriers except the carrier being routed initially.For example, if a failure causes some carriers' plans to become invalid, the route selection system may have to cancel those plans and still simultaneously select routes for some carriers that do not have plans. Infinite reservations, whether for physical reasons or simply software-configured reasons, can be used in the route selection system to perform route planning in which locations with infinite reservations are ignored for route generation as long as the infinite reservations are in effect.

[0046] A further aspect of the distribution system is that each has its own reservation length and duration t i The goal is to have multiple (e.g., n) trips with a coordinated time window T begin within the coordinated time window T. The last trip of the n planned trips begins within the coordinated time window T, but does not have to end within this coordinated time window. TIFF0007744363000001.tif1029. The length and duration of the movement t i depends mainly on the traffic density. At high traffic density, the interaction between carriers (e.g., path crossing) becomes more intense, resulting in a larger number of carriers, but with a shorter time t i A short move is made. The reservation length may be the number of nodes and / or logical positions reserved for the move. The term "move" may refer to an "action" and may not include a wait time before the next move is made.

[0047] The duration of a move may depend on the reservation length. A move is defined in this disclosure as a movement of a carrier in a straight line starting from one logical location and stopping at a second, different logical location. A move may consist of a displacement of a carrier through one or more logical locations. A move from a first final destination to a second final destination may be performed in one or more moves with intermediate destinations. At each intermediate destination, there may be a longer or shorter waiting time before the next move begins. The planned route selection may include all moves or only the next few moves performed until the second final destination is reached. Moves may have different reservation lengths that do not exceed the maximum reservation length. The reservation length of each move may include the logical locations to be reserved for the move. The maximum reservation length may be defined to avoid a situation where too many logical locations are reserved at once, causing these locations to be blocked for other carriers for too long. By having a model for estimating the time required for a move for a given move length, a complete time plan for reserving logical locations can be created. This time plan may include travel times and wait times until the next trip is available, as well as infinite reservation times for locations that should not be used or that contain carriers that do not have trips planned to their next destination.

[0048] For a diagnostic laboratory transport surface having a cooperative search window of about 10 seconds, n may be 1 to 12, particularly 3 to 8, and more particularly 6.

[0049] In a further embodiment, the reservation length is specific to the trip, for example carriers in certain areas may make longer trips than other areas.

[0050] A further aspect of the distribution system is that the route selection system is configured to use the largest available reservation length, particularly less than or equal to the determined maximum reservation length, for the next trip out of up to n possible trips in the cooperation time window T. In particular, the route selection system is configured to use the maximum reservation length such that n>1. The maximum reservation length for a trip is defined as a fixed number. In other embodiments, the maximum reservation length is adjustable from time to time to allow for longer trips in low traffic density situations.

[0051] In a further embodiment, the route selection system is configured to limit the maximum reservation length for travel so as not to unnecessarily block fields for an excessively long period of time. This set maximum reservation length is needed to prevent too many logical locations from being blocked for other travel carriers.

[0052] In another embodiment, the entire route in this cooperative search window of time length T need not be executed according to the calculated route. A carrier may arrive at an intermediate destination location that is its last reserved logical location earlier, and therefore a new route can or should be created sooner than after the time window T. Typically, a new route is calculated after a defined duration, such as 20% of the cooperative search window time T, has elapsed, or immediately after the transportation system sends an update that triggers the need for calculation of a new route, such as a route or movement failure or reaching a final or intermediate destination. This allows for improved throughput due to more up-to-date information on the conditions on the transportation surface for calculating the route. Therefore, routes are more efficient because the accumulation of uncertainty is reduced.

[0053] Another aspect of another embodiment of the distribution system is that the route selection system is configured to receive status information from the transport system. The received status information is stored in a storage device and compared with the latest status information stored in the route selection system. If this comparison detects a change in status, the reservation map used by the route selection system is updated and a new, final, modified route with a new reservation length is calculated for the carrier. The status information may be, for example, whether the carrier has passed a predetermined boundary on the transport surface, when the carrier has departed, passed, or arrived at any or a predetermined logical location, or when an error is detected by the transport system.

[0054] In a further embodiment of the distribution system, errors in the distribution system include, for example, abnormal and failed logical locations, failed or stranded carriers and their associated logical locations that can no longer be moved, and blocked final destination locations because the system corresponding to the final destination location is unavailable.

[0055] In another embodiment of the distribution system, the frequency of use of the transfer surface per logical location may be the status information for the transport system.

[0056] In another embodiment, the route selection system is configured to calculate the status information itself from the plan and any detected deviations thereof produced by the route selection system, also referred to herein as the router.

[0057] A further aspect of the distribution system is that the route selection system is configured to determine subsequent routes after the current route, assuming each route is successfully executed.

[0058] This allows for faster route selection, since once a move is confirmed as successful, i.e., each carrier has reached the planned intermediate or final destination along the current route, the move for the next route can be performed directly without losing time calculating the route first, since the route is already available. This further allows for better CPU utilization of the route selection system with continuous rather than peak computation loads.

[0059] Another aspect of the distribution is that the route selection system is configured to check whether all carriers arrive within their planned time windows, and if carriers that do not arrive within their planned time windows are detected, the planned time windows of those carriers that did not arrive at the stop position or intermediate or final destination position are extended, and the affected plans are extended respectively.

[0060] In a further embodiment, the affected plans, i.e. routes, are recalculated.

[0061] In further embodiments, depending on physical effects and control mechanisms such as friction and discontinuous driving forces, acceleration, steady-state velocity, and deceleration may not have constant values ​​but may vary over time. As a result, the travel time from start to stop may vary. Acceleration, steady-state velocity, and deceleration values ​​are assumed to define a realistic time window for travel, and therefore the logical position reservation. For example, the slowest expected acceleration and deceleration, and the lowest velocity, are chosen. The time window for travel can be calculated in conjunction with the number of logical positions to travel. Finally, a slightly longer time, such as 5, 10, 20, or 30 percent, is chosen to avoid generating too many failed plans.

[0062] In another embodiment, a sensible length of the cooperative search window T depends on the level of uncertainty in the future plan and the amount of computation time required to calculate the next set of moves. When stochastic movement behavior is strong, there is less certainty that future moves can be executed as planned compared to when deterministic movement behavior is strong. Thus, the time length of the cooperative search window T is selected to be shorter when stochastic movement behavior is strong and longer when deterministic behavior dominates. Creating plans with longer cooperative search windows T, and therefore more moves, requires more computation time per route. The goal is to plan with a sufficiently large cooperative search window T such that all included moves within that time window can be executed with a high degree of certainty. This reduces the frequency of re-planning and therefore requires less computational power.

[0063] In a further embodiment, instead of using a formula to calculate the time window for reserving positions, a lookup table can be used that pre-calculates times for moving 1, 2, 3, ..., nmax positions (where nmax is the maximum reservation length). These times are again based on modeled or empirically determined speed, acceleration, and deceleration. The positions do not need to be evenly spaced. For example, the distance between two logical positions on either side of a transport tile boundary may be greater than the distance between two logical positions within the tile. As a result, moving from one tile to the next may require additional time due to crossing the boundary. The model for calculating the time can take this additional time into account.

[0064] A further aspect of the distribution system is that the route selection system is configured to allow at least one of the logical locations to be designated only as a final destination location for a carrier and to prohibit its use as an intermediate destination location for transporting the carrier to the final destination location.

[0065] This allows for a smooth transfer of objects transferred by the distribution system to a separately managed transfer area or system, such as an analyzer in a diagnostic laboratory.

[0066] Another aspect of the distribution system is that the transport surface is divided into logical sub-areas.

[0067] This allows for the organization of subareas without the need to organize the entire system. The subareas can be defined by hardware modules, for example a transport system built from individual modules. The subareas can also be defined by logical areas, for example an object transfer area, a fast transport area, a sorting area, a buffer area for providing some kind of intermediate storage of objects, an area for empty carriers, or modules powered by one power supply unit.

[0068] Different carrier behaviors can be assigned to the sub-areas, such as different running speeds, carrier acceleration or deceleration behavior, special rules for carrier movement, for example first in, first out in a queue, etc.

[0069] A further aspect of the distribution system is that the route selection system is configured to allow only a limited number of simultaneous movements of carriers within a defined sub-area of ​​the transport surface.

[0070] Furthermore, for passive carriers, it is advantageous to design the power supply for the drive system with a defined maximum threshold for power consumption by limiting the maximum power consumption possible for the logic area.

[0071] Another aspect of the distribution system is that the route selection system is configured to release each node reserved by the current movement in the subarea after the carrier passes a predetermined logical position on the transport surface. Here, releasing a node means that in the node information, the respective time window of each logical position is set from reserved to free. As a first example, all nodes are directly released for further planning after the carrier passes each logical position in its planned movement and reaches the next logical position. Another possibility is to release all nodes in a subarea after the carrier passes the boundary of the respective subarea.

[0072] These boundaries may correspond to logical entities of the transport system, such as the boundary between one module of a transport surface and the next if the transport surface is constructed from tiled transport modules. Other boundaries may be artificially defined by defining logical locations that form defined boundaries, or by detection systems that detect when a carrier has passed, entered, or left a predetermined location, such as by optical gates or camera systems, inductive, capacitive, conductive, or radio frequency sensors.

[0073] A further aspect of the distribution system is that the route selection system is configured to check the success or failure of planned trips, and the parameters of the planning model are adjusted by the same value for the entire transportation area or by different values ​​for each subarea depending on the number of failed plans in the relevant area. As described above, due to stochastic effects, exceptions, and errors, not all routes planned by the route selection system are necessarily executed as proposed by the route selection system. If the execution time for executing the planned route is within the planned time range, the planned route can be executed. The route selection system may be configured to monitor the execution time and the planned time range to check the success or failure of the planned trip. For example, the route selection system may be configured to count the number of trips that exceed the planned time range. The route selection system may be configured to determine a trip, particularly an individual trip, as “successful” if the execution time for executing the trip falls within the planned time range, at least within a tolerance. However, if the execution time for executing a trip of a planned route exceeds the planned time range, it may interfere with other trips and lead to a plan failure. For example, if moves in a particular area frequently fail to execute the plan, this may signal that the model is too optimistic for this area. For example, if it is detected that a carrier frequently fails to execute individual moves regardless of where the carrier is located on the surface, the system can identify that the carrier is no longer moving according to the model of a "healthy" carrier. The reason may be, for example, wear on the carrier's sliding surfaces or dirt on the carrier's sliding surfaces. Based on this information, the control system can remove the carrier or adjust the carrier's planning model, and ultimately register the carrier as not in good condition so that the carrier can be updated or cleaned by the next system maintenance. The route selection system may be configured to determine a move as "failed" if the execution time for executing the move exceeds the planned time range.The route selection system may be configured to determine a "failure" when the number of trips beyond the planned time range exceeds a predetermined threshold. For example, a threshold may be used for a maximum allowable number of failures per time, area, or carrier. If a failure is determined, the route selection system may be configured to adjust parameters of the planning model by the same value for the entire transportation area, or by different values ​​for each subarea depending on the number of failed plans in the associated area. For example, the route selection system may be configured to adjust a parameter defining the reservation length in the planning model.

[0074] System conditions can affect the carrier's mobility. Therefore, the parameters of the planning model can be adjusted to reflect the carrier's actual mobility within the route selection system. The planning model codifies the carrier's mobility, which means achievable or determined acceleration / deceleration and maximum speed as parameters.

[0075] In another embodiment of the distribution system, the route selection system is configured to measure deviations between modeled travel times and actual travel times and adjust parameters of the planning model respectively.

[0076] A further aspect of the distribution system is that the route selection system is configured to adjust the maximum reservation length in response to a deviation between the maximum reservation length and the planned reservation length. The route selection system may be configured to execute at least one self-learning algorithm to optimize the maximum reservation length for a successful route. Specifically, the route selection system may be configured to compare the maximum number of logical positions that can be reserved for the trip with the planned reservation length. The planned reservation length may be a reservation length that the route selection system considers optimal for a successful route in the carrier's route planning. The planned scheduled length may be an individual reservation length assigned to the trip. The maximum reservation length may be a boundary condition of the planning model that sets an upper limit on the possible reservation length. The route selection system may be configured to adjust the maximum number of logical positions that can be reserved for the trip in response to a deviation between the maximum number of positions that can be reserved for the trip and the planned reservation length.

[0077] Here, adjustment means that the maximum reservation length used for route selection planning is increased when most or 80%, 90% or 100% of the carrier's planned routes use a trip length equal to the maximum reservation length, thereby speeding up the system, or at least areas with low traffic intensity or high levels of parallel movement.

[0078] A further aspect of the distribution system is that the route selection system is configured to count the number of failed plans for each carrier and / or subarea of ​​the transportation surface, and if the number of failed plans exceeds a predetermined number or the frequency of failures exceeds a predetermined number, the carrier is marked as "maintenance required" in the control system for each carrier, or nodes in the subarea are marked as "maintenance required" or are no longer used for further route selection in the route selection system, or are minimized for use only as necessary to avoid transportation problems. The route selection system may be configured to determine the frequency of engagement in failed plans for a logical location and / or determine an action the route selection system will trigger if a threshold number is exceeded, such as one or more of excluding the logical location from route selection, notifying of the need for maintenance, etc. The distribution system, such as the control system, may include at least one user interface configured to display at least one information, e.g., that maintenance is required, if the number of failed plans or the frequency of failures exceeds a predetermined number for each carrier and / or subarea of ​​the transportation surface. A user interface may be configured to interact with a surrounding environment, for example, to exchange information unidirectionally or bidirectionally, such as exchanging one or more data or commands. For example, a user interface may be configured to share information with a user and receive information by a user. A user interface may be a feature that interacts visually with a user, such as a display, or a feature that interacts acoustically with a user. A user interface may comprise, by way of example, one or more of a graphical user interface, a data interface, such as a wireless and / or wired data interface, etc.

[0079] The marking can be done in a log file or registry file stored in a storage device, memory, or database of the route selection system. This allows carriers with problems to be checked periodically and cleared or sent to a maintenance or replacement area for carriers on the transport surface by the respective route selection / destination. Furthermore, it is possible to identify areas on the transport surface where carriers are no longer moving properly. These areas can then be avoided or used less in the route selection, or checked for maintenance or dirt, or for active transport surfaces, for problems with the drive means, and the necessary measures can be taken to solve the respective problem, for example by cleaning the transport surface or replacing the transport surface or module.

[0080] In a further embodiment of the distribution system, areas and carriers may be visually marked, for example by changing the color of LEDs built into the surface or carrier, or by showing them on the screen of a control computer or mobile client device.

[0081] Another aspect of the distribution system is that the route selection system is configured to assign usage costs to nodes according to the frequency of a carrier's passage, departure, or stop at this node, and to minimize the usage costs for each planned trip.

[0082] This results in an evenly distributed use of the transfer surface. Acceleration at the start location / node and deceleration at the end location / node place a heavy load on the transfer surface, so the number of accelerations and decelerations can be included in the cost function to equalize them across nodes. This can improve the lifespan and reliability of the system and reduce maintenance.

[0083] A further aspect of the distribution system is that the route selection system is configured to check whether carriers are blocking each other, eliminating the possibility of movement for k carriers (k>1) for an infinite time. This situation is denoted as a "stalemate."

[0084] Further aspects of the distribution system include: out seconds(t out A stalemate is also declared by the route selection system if no movement is observed for a period of time (where s is configurable, for example, between 2 and 60 seconds, or 2, 5, or 10 seconds).

[0085] Another aspect of the distribution system is that the route selection system is configured to check whether a stalemate has occurred. The route selection system may be configured, if a stalemate is detected, to ignore the metrics of the final destination for all involved carriers and to make one or more moves selected from among moves possible by all involved carriers ignoring the metrics of the final destination. In particular, the moves may be selected randomly. This can allow for space to be created for the stalemate to be resolved. In particular, all carriers on the transfer surface or a defined area of ​​the transfer surface make one or more moves selected from among moves possible by ignoring the metrics of the final destination. If a stalemate is detected, the next n moves may be selected, in particular randomly, from among moves possible by all carriers located near the origin of the stalemate, ignoring the metrics to the final destination. A stalemate may start with a few carriers and may involve other carriers.

[0086] By ignoring the metric to the final destination, the carrier can travel in any direction, including backwards or perpendicular to the direction of the final destination.

[0087] Some or all of the carriers involved in the stalemate take possible moves that are different from the move that achieves the goal of reaching the final destination in the shortest time or distance. Randomness can be achieved by using a normalized random number generator for all possible moves of each carrier. Such moves can be made in random directions and lengths. Thus, for n moves, the carriers behave like Brownian motion.

[0088] In an alternative embodiment, the direction and length of the movement can be selected, for example, with values ​​determined to effect movements that result in a pattern of clear paths or clear positions in the selected direction among the carriers involved in the stall. After these movements, the normal route selection process is again applied to the carriers, including the carrier that was involved in the stall.

[0089] In another embodiment of the route selection system, the carrier's state is either "moving through reserved logical position" or "waiting at logical position" for a planned movement. If a movement cannot be calculated for a carrier, the carrier waits until a movement becomes possible. If no possibility to move the carrier is found during the cooperative time window T, the node where the carrier must wait is assigned an infinite reservation time. During subsequent planning, a possibility to move the carrier can be found, after which the node is assigned a finite-time reservation. Furthermore, nodes with carriers without destinations, such as carriers waiting to be used, or that need to be blocked for a time longer than the cooperative time window T, also receive infinite reservation time. This method avoids the cooperative A* cycle problem, which leads to an infinite cycle. Furthermore, this method can be used to prevent nodes from being used, for example, when servicing parts of a transportation system.

[0090] A further aspect of the present invention is a method for a distribution system for moving carriers on a transport surface. The distribution system includes a transport surface, a carrier for transporting objects, and a drive system for moving the carrier on the transport surface. A control system of the distribution system controls movement of the carrier on a planned route from a start location on the transport surface to a final destination location. A route selection system of the distribution system models the transport surface with a graph of nodes. A node is a logical location on the transport surface where a carrier can start and stop, i.e., a possible start location or a final or intermediate destination location for the carrier. The route selection system determines free time windows and reservation time windows for at least some of the logical locations. The route selection system calculates routes for at least two carriers on the transport surface using a windowed hierarchical cooperative search algorithm within the cooperative time window T. The route selection system assigns individual reservation lengths to each carrier for next movements within the cooperative time window T. The route selection system assigns infinite reservation times to at least one or more logical locations.

[0091] In particular, infinity here means spanning at least the time window T.

[0092] In a further embodiment of the method, the route selection system comprises: The logical location is the location of a carrier that does not have a final destination, The logical location is corrupted or blocked, There is no possibility of moving the carrier from the logic position during the cooperation time window T, is configured to assign an infinite reservation time to a logical position when one or more of the following is realized:

[0093] Infinite here means that the logical location is reserved until the condition that led to the infinite reservation is resolved, for example, a new final destination is assigned to the carrier, the logical location is repaired or unblocked, or a new movement possibility is discovered.

[0094] The route selection system may be further configured to assign an infinite reservation time to a logical location if the logical location is a location of a carrier for which a route has not yet been planned and / or is a location of a carrier for which a route has / had to be cancelled.

[0095] A heuristic is used to ensure that the carrier is directed to its final destination, and a table of pre-calculated or measured travel times from each logical location to each other logical location, or from each logical location to the associated final destination, is stored within the route selection or control system. To pre-calculate these travel times, the carrier is assumed to be able to travel to the final destination at a predetermined speed, acceleration, and deceleration without being impeded by other carriers. The predetermined speed, acceleration, and deceleration can be based on the measured speed of the actual carrier or any speed, such as 1 m / s.

[0096] In another embodiment of the method, the distance of the logical locations to the final destination locations can be used for such a look-up table to guide the carriers towards their final destination locations.

[0097] A heuristic method ensures that the route selection system can select as the next node not only a node whose logical location has an available time window, but also a node, and therefore a logical location, that reduces the travel time or distance to the final destination.

[0098] A further aspect of the method is that the route selection system uses the most available reservation length up to a predetermined maximum reservation length for each linear movement of the carrier that has an available free time window on the route.

[0099] This allows for fast and reliable route selection for carriers in the distribution system. Thus, when carriers are cooperatively searching for free time windows of logical locations on the route to a destination, the search for individual carriers is stopped if no logical location with the required free time window is available for the next trip. Theoretically, this could be the final destination. Each carrier can have a different individual reservation length planned for the next trip.

[0100] Another aspect of the method is for the route selection system to receive transport system state information, store the state information in a memory device of the route selection system, and compare the state information with previously stored state information. When the route selection system detects a change in the transport system state through this comparison, the route selection system calculates a new route for the carrier for the next trip with a new reservation length, taking into account the last known state of the transport system. Additionally, a new route is calculated when a new destination is assigned to the carrier.

[0101] The communicated status information may be information such as that a carrier has passed some software-defined or hardware boundary on the transport surface, i.e., moved from one particular area of ​​the transport surface to another, reached its final destination, or that some area of ​​the transport surface is blocked or malfunctioning, or that some carriers are malfunctioning.

[0102] In a further aspect of the method, the route selection system determines a subsequent route after the current move, assuming each move of the carrier is successful.

[0103] This allows trips to be performed more quickly, or even continuously, immediately after a particular carrier has reached the final or intermediate destination of the last planned trip, resulting in more efficient use of computer resource capacity.

[0104] A further aspect of the method is that the route selection system checks whether all carriers have arrived within the planned time window, and if a carrier is detected that does not arrive within the planned time window, extends the next planned reservation length and recalculates the plans of other affected carriers. In particular, the route selection system may be configured to extend the next planned reservation length. The distribution system may be a self-learning system. The route selection system may be configured to automatically adjust the planned reservation length, in particular without requiring manual interaction. The route selection system may be configured to automatically optimize the route plan. The optimization may include optimizing model parameters for a successful route.

[0105] A further aspect of the method is that the route selection system is configured to allow a carrier to address at least one of the reserved nodes only as a final destination and to prohibit the carrier from using that node as an intermediate point for transporting the carrier to its final destination.

[0106] This allows reserving specific points / nodes on the transport surface that are less occupied for performing specific tasks. For example, a transfer position for handling objects to or from a carrier may be such a special node on the transport surface. Another example may be a read position where some kind of information on the transported object needs to be read, or a handling position where some kind of operation is performed on the object.

[0107] Another aspect of the method is that the route selection system only allows a limited number of carriers to move simultaneously within a defined sub-area of ​​the transport surface.

[0108] In the case of transport systems with drive systems in the transport plane, this also avoids drawing too much energy in certain areas, or allows fewer or smaller power sources to be used.

[0109] A further aspect of the method is that the route selection system releases each node reserved by the current movement on the subarea after the carrier passes the designated node on the transport surface.

[0110] These designated nodes may simply be the last nth nodes (n may be an integer between 1 and 20, or between 1 and 10, or between 1 and 5) that the carrier has passed through, or may be nodes located at the boundaries of sub-areas of the transport plane, and the nodes of the passed sub-areas are released at once.

[0111] Another aspect of the method is that the route selection system is configured to check the success / failure of planned movements, particularly for defined areas and / or carriers on the transport surface, and depending on the number of failed plans for a carrier or the number of failed plans in the relevant area, the route selection system adjusts the parameters of the planning model for the entire transport surface, or with different values ​​for each sub-area, or with individual values ​​for each carrier, or for carriers within a particular sub-area.

[0112] This allows the route selection system to react to different characteristics of a transport surface, sub-areas on the transport surface, or carriers, making route selection systems more efficient and reliable, and also allowing for faster route selection.

[0113] A further aspect of the method is that the route selection system adjusts parameters of the planning model in response to the deviation of the planned time of the trip relative to the carrier's current actual travel time.

[0114] For example, if the carrier reaches the destination within the time, the moving speed of the carrier in the planning model can be increased or the moving time can be decreased, whereas if the carrier does not reach the destination within the planned time, the maximum speed can be decreased or the moving time can be increased in the planning model.

[0115] In another embodiment of the method or distribution system, the acceleration / deceleration values ​​are changed and the speed values ​​are kept the same depending on whether only short moves fail. If only plans for longer moves fail, only the speed values ​​are adjusted.

[0116] This allows for more efficient or reliable route selection.

[0117] Another aspect of the method is for the route selection system to count the number of failed moves for each carrier and / or subarea of ​​the transport surface, and if the number of failed moves exceeds a predetermined number or the frequency of failed moves exceeds a predetermined number, mark the carrier as needing maintenance, or mark the transport surface or subarea of ​​the transport surface as needing maintenance, not to be used for future route selection, or not to be used often for future route selection.

[0118] In a further embodiment of the method, the control system designates carriers needing maintenance and routes them to specific maintenance areas on the transport surface so that maintenance can be performed, such as, for example, the carrier can be replaced with a new carrier, parts of the carrier can be replaced with new parts, energy storage devices such as batteries can be charged or replaced, or parts of the carrier can be cleaned or treated with a maintenance substance / fluid.

[0119] With regard to the transfer surface, it may be a cleaning or new surface, new actuators or new actuation electronics may be needed, or some obstruction may need to be removed, and so on.

[0120] A further aspect of route selection is that the route selection system assigns usage costs to nodes depending on the frequency of carrier passage, departure, or stop at this node, minimizing the usage costs for each planned trip.

[0121] The search algorithm uses a cost function to determine the best route found. This cost function can depend on the actual distance to the final destination or intermediate destinations, or, for example, Manhattan distance. An additional cost can be added that is optimized on this cost function to find the "best" route. Thus, the best route need not be the shortest route in the sense of geometric distance. Furthermore, other aspects can be considered in the cost function, such as lowering the cost of routes that pass through areas on a transportation surface that are less used than other areas, higher costs for routes in frequently used areas, lowering the cost of routes with longer trips, lowering the cost of routes with preferred distances to other carriers, lowering the cost of routes with fewer turns, lowering the cost of routes that pass through locations with fewer carrier starts and stops, etc.

[0122] In a further embodiment, a further cost function can be implemented as a cost of resource consumption that can be calculated as the sum of all reservation lengths and carrier waiting times of all planned movements, such a cost making it advantageous, among other things, to wait a little longer rather than using more resources to make a zigzag movement for a slightly faster arrival.

[0123] A further aspect of the method is that the route selection system uses the A* algorithm as the knowledge-based search algorithm, which allows for reliable and fast route selection for carriers.

[0124] However, other embodiments are possible that involve combinations of the features disclosed herein.

[0125] In the above, all features of the systems can be used in the methods, and all method steps can be performed by the respective systems or parts of systems employed to do so.

[0126] The terms "embodiment" and "aspect" are used synonymously.

[0127] In summary, without excluding further possible embodiments, the following embodiments can be envisaged:

[0128] Embodiment 1. 1. A distribution system comprising: a transfer surface having a plurality of logical locations for dispensing a plurality of objects; a plurality of carriers for transporting said objects; a drive system for moving the carrier between the logical positions on the transfer surface; a control system configured to control the carrier to move on the transfer surface from a start location to a final destination location along a planned route; Equipped with the control system comprises a route selection system configured to model the transportation surface as a graph of nodes and calculate planned routes for at least two carriers on the transportation surface by using a knowledge-based windowed hierarchical cooperative search algorithm over a cooperative time window T; the route selection system is configured to determine a reserved time window and an available time window for each logical location; the route selection system is configured to assign a separate reservation length to each carrier for a next movement for a logical location having an open time window; A distribution system in which an infinite number of reservation times are assigned to one logical location on a transport surface.

[0129] Embodiment 2. The route selection system comprises: The logical location is the location of a carrier that does not have a final destination, The logical location is the location of a carrier for which no plan has yet been created, The logical location is the location of the carrier whose plan was or had to be cancelled, The logical location is corrupted or blocked, There is no possibility of moving the carrier from the logic position during the cooperation time window T, 2. The distribution system of embodiment 1, configured to assign infinite reservation time to a logical location when one or more of the following are realized:

[0130] Embodiment 3. 3. A distribution system as described in embodiment 1 or 2, wherein the route selection system is configured to use the largest available reservation length for linear travel that is less than the determined maximum reservation length.

[0131] Embodiment 4. the route selection system receives status information for the transportation system; In particular, the distribution system of any one of embodiments 1 to 3 is configured to compare the status information with the latest status information stored in the route selection system, and more particularly, to calculate a new planned route with a new reservation length for the next journey if a change in status is detected by the comparison.

[0132] Embodiment 5. 5. The distribution system of any one of embodiments 1 to 4, wherein the route selection system is configured to determine subsequent planned routes after a current trip assuming each trip is successful.

[0133] Embodiment 6. A distribution system as described in any one of embodiments 1 to 5, wherein the route selection system is configured to check whether all carriers have arrived within the planned reservation length, and if a carrier is detected that has not arrived within the planned reservation length, extend the next planned reservation length and recalculate the plans of other affected carriers.

[0134] Embodiment 7. 7. The distribution system of any one of embodiments 1 to 6, wherein the route selection system is configured to allow at least one of the logical locations to be addressed by the carrier only as a final destination and to prohibit it from being used as an intermediate destination location for transporting the carrier to the final destination location.

[0135] Embodiment 8. 8. A distribution system according to any one of embodiments 1 to 7, wherein the transport surface is divided into logical sub-areas.

[0136] Embodiment 9. 9. The distribution system of embodiment 8, wherein the route selection system is configured to allow only a limited number of simultaneous movements of carriers within logical subareas of the transport surface.

[0137] Embodiment 10. A distribution system as described in embodiment 8 or 9, wherein the route selection system is configured to release each logical position reserved by the current movement after the carrier passes a predetermined logical position on the transport surface.

[0138] Embodiment 11. A distribution system as described in any one of embodiments 8 to 10, wherein the route selection system is configured to check the success / failure of the planned movements, and depending on the number of failed plans, the parameters of the planning model are adjusted by the same value for the entire transport surface, or by different values ​​for each logical subarea, or by individual parameters specifically for carriers of a particular subarea.

[0139] Embodiment 12. A distribution system described in any one of embodiments 1 to 11, wherein the route selection system is configured to adjust the maximum reservation length according to the deviation between the maximum reservation length and the planned reservation length.

[0140] Embodiment 13. the route selection system is configured to count the number of failed plans for each carrier and / or subarea of ​​the transport plane; If the number of failed plans exceeds a predetermined number or the frequency of failure exceeds a predetermined number, the carrier is marked as needing maintenance; or A distribution system according to any one of embodiments 1 to 12, wherein the transport surface or the sub-area is marked as requiring maintenance and / or is avoided or used less frequently for further route selection.

[0141] Embodiment 14. A distribution system as described in any one of embodiments 1 to 13, wherein the route selection system is configured to assign a usage cost to a logical location according to the frequency of a carrier passing, departing, or stopping at this logical location, and to minimize the usage cost for each planned trip.

[0142] Embodiment 15. A distribution system according to any one of embodiments 1 to 14, wherein the route selection system is configured to check whether a stalemate has occurred and, if a stalemate is detected, to select, in particular randomly, the next n trips for all carriers in the vicinity of the stalemate from among the possible trips while ignoring the metrics to the final destination.

[0143] Further optional features and embodiments of the present invention are disclosed in more detail in the subsequent description of preferred embodiments, preferably in conjunction with the dependent claims. Each optional feature therein may be realized independently as well as in any possible combination, as will be understood by those skilled in the art. The scope of the present invention is not limited by the preferred embodiments. The embodiments are illustrated diagrammatically in the figures, where identical reference symbols refer to identical or functionally equivalent elements. [Brief explanation of the drawings]

[0144] [Figure 1] 1 shows a schematic perspective view of a distribution system. [Figure 2A] 1 shows a time-velocity diagram of a carrier moving on a transfer surface of a distribution system. [Figure 2B] 2A shows time-velocity diagrams of three possibilities for how to parameterize the movement shown in FIG. 2A. [Figure 3A] The time-velocity diagrams of the actual movement and its parameterization are shown. [Figure 3B] 3B shows time-velocity diagrams with different parameterizations of the movement shown in FIG. 3A. [Figure 4] 1 shows a schematic diagram of the distribution of reservation lengths. [Figure 5A] The reservation table and corresponding parts of the transport surface are shown, along with the movements to be performed. [Figure 5B] The reservation table and corresponding parts of the transport surface are shown, along with the movements to be performed. [Figure 6A] Two examples of impasses on the transfer surface are shown. [Figure 6B] It shows how random movement can break stalemates. [Figure 6C] It shows how the subsequent route selection can resolve the impasse. [Figure 7] 1 illustrates a schematic of the communication connections of one embodiment of a distribution system. [Figure 8] Shows a flow diagram of how the next route is calculated and executed. [Figure 9A] 1 illustrates different transport system area layouts and methods of releasing reserved fields. [Figure 9B] 1 illustrates different transport system area layouts and methods of releasing reserved fields. [Figure 9C] 1 illustrates different transport system area layouts and methods of releasing reserved fields. [Figure 10A] 1 shows the route of a carrier in a transport system layout where all logical positions are available compared to the same layout where some logical positions are blocked. [Figure 10B] 1 shows the route of a carrier in a transport system layout where all logical positions are available compared to the same layout where some logical positions are blocked. [Figure 11] 1 illustrates one embodiment of a distribution system. [Figure 12] 1 shows a Gantt diagram. [Figure 13] This shows the stochastic effect of travel on route choice. [Figure 14] The results of stochastic effects on migration are shown. [Figure 15] The discrete-time WHCA* algorithm is shown in comparison with the continuous WHCA* algorithm in games. [Figure 16] Possible moves are shown. DETAILED DESCRIPTION OF THE INVENTION

[0145] 1 shows a schematic perspective view of a distribution system 10, particularly a transport system for a diagnostic laboratory for obtaining test results for patients, required to transport objects 16 in carriers 14 between stations 18 of the distribution system 10.

[0146] The distribution system includes a transport system 11 having a transport surface 12 and carriers 14, also designated C1, C2, and C3. Each of the carriers 14, C1 and C2, carries an object 16, while the carrier 14, C3, is empty. The carriers 14 are moved or move on their own along the transport surface 12. To move the carriers 14, the transport system 11 includes a drive system. The drive system may be implemented by an electromagnetic coil below the transport surface and a permanent magnet within the carrier 14. This allows a magnetic field generated by the coil to push and / or pull the permanent magnet within the carrier 14, and thus the carrier 14, across the transport surface 12. To determine the position of the carrier 14 on the transport surface 12, sensors such as optical, magnetic, capacitive, or inductive sensors may be embedded in the transport surface 12. Another option may be a camera system 21 with image analysis software to determine the position of the carrier 14.

[0147] Other drive systems are possible in other embodiments, such as self-propelled carriers 14 that include sensors, motors, and energy storage such as batteries, especially rechargeable batteries. These self-propelled carriers 14 can therefore also travel autonomously on the transfer surface 12. In this case, the control device can be part of the carrier or distributed on the carrier 14.

[0148] The objects 16 can be delivered to stations 18. The carrier 14 can wait for the objects 16 at the delivery location or drive away, and the objects 16 are placed on another carrier 14 as needed after processing at the respective station.

[0149] In a further embodiment, the transfer surface 12 may be used within the station 18 for directly transferring carriers within the station 18 .

[0150] For example, in a diagnostic laboratory, distribution system 10 is used to transfer tubes containing, for example, biological sample fluids and / or consumables or other species such as tissues, reagents, waste, or disposables between stations 18. Stations 18 may be modules in a diagnostic laboratory, such as, for example, portion-generation stations, centrifuges, or analytical modules performing a single analysis or entire analyzers, pre-analytical or post-analytical stations, pipetting systems, incubators, mixers, or detection units. In other embodiments, distribution system 10 may be a warehouse distribution system that distributes items as objects 16 between stations 18, such as shelves and packing stations, or may be a manufacturing site where objects 16 are raw or semi-finished goods that need to be transferred between work stations, for example, in a machine shop.

[0151] A control system 20 is part of the distribution system to control the movement of the carrier 14. The control system 20 may, for example, control the drive system, e.g., the current in the coils, and / or collect position information of the carrier 14.

[0152] As shown in Figure 7, the distribution system includes a route selection system 50, which may be part of the control system 20 or may at least be connected to the control system 20 for information exchange. The route selection system 50 includes a computing device and a memory device for calculating a planned route for at least one or more carriers 14 on the transfer surface 12 from a starting location to a final destination or an intermediate destination. To do so, the route selection system 50 models the transfer surface using a graph of nodes. The final route or movement is transmitted to the control system 20, which then transmits appropriate information to the drive systems to move the carriers 14 on the transfer surface 12 along the determined route.

[0153] The dashed lines shown in FIG. 1 represent a graph or possible routes between their intersections, which define logical positions, e.g., N1 to N7. Not all logical positions are numbered in FIG. 1. In this example, the graph forms a rectangular mesh. Other models, e.g., curvilinear routes, are also possible. For some drive systems, the logical positions are given by the technically possible start-stop positions on the transport surface 12. This is the case, for example, for drive systems with coils below the transport surface 12. On the other hand, for self-propelled carriers, purely software-defined logical positions are possible, since the carriers can start and stop at any position on the transport surface 12. At the very least, the logical positions must be far enough apart on the transport surface 12 that two carriers 14 can be positioned adjacent to each other.

[0154] The route selection system 50 determines reservation time windows and free time windows for each logical location and calculates a minimum-cost route for at least two carriers 14 on the transfer surface 12 from the starting location to the final destination location using a knowledge-based windowed hierarchical cooperative search algorithm such as WHCA*. This enables cooperative route selection within the cooperative time window by not losing direction to the final destination. Typically, several intermediate destination locations are required to reach the final destination location, which depends on the dimensions of the transfer system 12. The final destination location may be a transfer location to a station 18 or an operation location at the station 18.

[0155] For example, in a diagnostic laboratory, an object 16, such as a tube containing a biological fluid, is placed at one destination location from a pre-analytical system onto a carrier 14. A further destination location can be at a station 18, such as an analyzer, where the tube is picked up and placed into the analyzer, or a portion of the fluid can be pipetted at the destination location. These types of destination locations are also called transfer locations, because the object is transferred from the carrier 14 to the station 18 or vice versa.

[0156] For stations 18 that themselves have the same or similar transport systems, the final destination location may be a logical location on transport surface 12 to which carrier 14 can move within the station. In further embodiments, the final destination location may be internal to station 18, such as a pipetting location in a diagnostic lab or a welding location on a manufacturing site.

[0157] Figure 5 shows reservation tables 13, 15 as well as two possible linear movement routes of carrier 14 on a further transport plane 12 along further logical positions N1 to N15 spanning three separate modules 1, 2, 3. On the right hand side, logical positions N1 to N15 are shown on transport plane 12 and three movements 4, 5, 6 of carrier 14 are shown. The diagram on the left hand side of Figure 5 illustrates reservation in time for the situation when a boundary crossing release is performed to release the reserved logical positions earlier than the end of each of movements 4, 5, 6.

[0158] The carrier's first move 4 on its first route in FIG. 5A is from logical position N4 to logical position N8, using a predetermined maximum reservation length for a move consisting of four logical positions. As shown in first reservation table 13, all logical positions N4-N8 of this move are reserved for the duration of this first move 4, but when moving from N6 to N7, the boundary between module 1 and module 2 is crossed, and the reservations of logical positions N4-N6 are released before the end of first move 4 at logical position N8. Logical positions that are released early, after the carrier crosses a module boundary, for example from module 1 to module 2, are indicated by cross-hatching in first reservation table 13.

[0159] Further embodiments of methods for releasing reserved fields are described below with reference to Figures 9A, 9B and 9C.

[0160] As shown in the embodiment of Figure 5A, after reaching logical position N8, logical position N8 remains reserved for a second move 5 that ends at logical position N12, which has a maximum reservation length of four logical positions. This second move 5 does not cross a module boundary, so no logical positions are released early. As can be seen in the first reservation table 13, the reserved positions are released at the end of this second move 5, regardless of the destination.

[0161] The final move on this route to intermediate final position N15 requires only a length of three logical positions, less than the predetermined maximum reservation length for this embodiment. When crossing the boundary between Module 2 and Module 3, the reservation for this move at logical position N12 is released when the carrier reaches Module 3 at logical position N13.

[0162] After the carrier reaches its destination point N15, all reserved positions are released again except for destination position N15, which remains reserved for a theoretically infinite reservation length until the next move is performed, as indicated by the long diagonal line at N15.

[0163] In a further embodiment, the reservation may also be removed when the carrier 14 is removed from the transport surface 12 (not shown).

[0164] It should be emphasized that the squares in the reservation table along the time axis do not mean that time is discretized in these increments. For simplicity, time is shown as discrete increments, but the presented method uses a continuous time scale. Also, time is not illustrated to scale. For example, the last move with three logical positions takes more than three-quarters of the time required for a move along four logical positions, since the acceleration and deceleration portions of the time are more than in the case of a move along four logical positions.

[0165] In FIG. 5A, the carriers do not encounter cross traffic, so the maximum reservation length is used for the first two trips.

[0166] FIG. 5B shows an example of a second route 7, as a straight line, from logical location N4 to logical location N15, with a maximum reservation length of five logical locations. Another route 8 of another carrier intersects this route at logical location N10. For purposes of calculating the second route 7's movements, this intersection is marked by a black reservation at logical location N10 in the lower reservation table 15. A first movement 25 of the second route 7 starts at logical location N4 and stops at logical location N9. By passing from logical location N6 to logical location N7, the boundary between module 1 and module 3 is crossed, freeing up the reservation at logical locations N4-N6 for use in another movement. A second movement 26 of the route 7 starts at logical location N8. Because logical location N10 is already reserved for another carrier making movement 8, the carrier can only reserve up to logical location N9, at which point it must wait until location N10 becomes available again. After N10 becomes available again, a second move 26 is again planned for a maximum reservation length of five logical positions to logical position N13. Because the boundary between module 2 and module 3 is crossed between logical positions N12 and N13, the reservation of logical positions N9-N12 is released when the carrier reaches logical position N13. A third move 27 of the second route 7 starts from logical position N13 and stops at logical position N15, which is the intermediate or final destination of the carrier on the second route 7.

[0167] Again, in the second reservation table 15, time is illustrated by time blocks, but this is for ease of explanation only, however the algorithm uses a continuous time scale.

[0168] In a further embodiment, route selection system 50 adjusts the maximum reservation length that can be allocated to a planned trip depending on the density of traffic within a managed subarea of ​​the transport plane. For example, if more than 30% of the nodes within a subarea of ​​the transport plane are occupied by carriers, the maximum reservation length is set to three logical locations, and if less than 30% of the logical locations within the subarea are occupied by carriers, the maximum reservation length is set to six logical locations.

[0169] For carriers 14 whose final destination is unknown to the route selection system 50, the route selection system 50 assigns infinite reservation time to the node in the logical location of such carrier. Infinite reservation time means that the node cannot be used for further route selection until the carrier on that node either acquires a new final destination again or is removed from the transport plane 12.

[0170] The final destination position of carrier 14 may be determined by a command and control system 60 that receives commands from a higher-level command and control system 70 (FIG. 7 and see below), such as a laboratory information system in the case of a diagnostic laboratory transport system. This command and control system 60 then forwards the required final destination position for carrier 14 to control system 20 and / or route selection system 50. The final destination position on transport surface 12 is primarily determined by the needs of the objects 16 being transported by carrier 14.

[0171] In another embodiment, the command and control system 60 defines only an identifier, such as an analyzer, which the control system 20 converts into a location for the route selection system 50 .

[0172] It is also possible for a station 18 to request a particular object 16 or an empty carrier 14 for loading, and for the station control system to send a request to the command and control system 60 or laboratory information system to give the carrier 14 a new final destination.

[0173] Typically, the laboratory information system, according to information obtained from the upper command management 70, distributes the objects 16 evenly among the possible stations 18, for example, and realizes the workflow of the stations 18 to which the objects should be sent so that the objects can be processed by the correct stations. In a further embodiment, the upper command management 70 is an intermediate software layer, often called middleware. The command management system 60 converts this information into destination positions on the transfer surface 12 and provides assignments of final destination positions for the carriers 14, which represent their respective final destination positions. In a further embodiment, the distribution of empty carriers 14 for transferring the objects 16 from the stations 18 after processing is organized at the level of the control system 20 or at the level of the command management 60.

[0174] In a further embodiment, empty carriers are provided to stations 18 just in time as needed to their respective final destination locations.

[0175] The transport surface 12 is equipped with sensors for detecting status information of the transport system. For example, the position of carriers, unavailable or broken positions / nodes on the transport surface or dirt on the transport surface can be detected by a camera system 21 and appropriate image processing. Other position detection systems are also possible, such as Hall sensors, current sensors, conductivity sensors, capacitance sensors, inductive sensors or optical barriers.

[0176] In another embodiment, only the position of the carrier 14 is sensed by the transport surface 12. If the carrier 14 takes too long to reach the next logical position (timeout), the firmware attempts to move the carrier again. After several attempts, it stops and generates an error. The associated position is then marked as "broken" by the control system. For example, if the carrier 14 should move from position 1 to 2 to 3, but does not reach 3 and is sensed to have passed 2, the system concludes that the carrier 14 is stuck between 2 and 3. The control system 20 therefore marks positions 2, 3, and all positions in the line between them (and, e.g., positions around them) as unavailable.

[0177] This status information is forwarded to the route selection system 50, which stores it in a storage device, such as a semiconductor memory. The route selection system 50 compares the newly received status information with the most recent status information stored in the memory device. If a change in status is detected, such as damage or soiling of some areas on the transport surface, or a damaged carrier 14 blocking a logical location and, if necessary or meaningful, adjacent logical locations, the route selection system 50 calculates a new route for the affected carrier 14 by maintaining the carrier's 14's final destination location. Furthermore, a transfer or processing location, i.e., a final destination location, may no longer appear to be available, for example, because the corresponding station 18 has stopped operating. This station 18 then transmits this information to the dispatch and control system 60, the LIS, or a higher-level dispatch and control system 70. The route selection system 50 then blocks the respective final destination location as no longer available. This also constitutes a change in the status of the transport surface 12.

[0178] The route selection system 50 calculates subsequent movements of the carriers 14 before the carriers 14 reach their intermediate or final destinations, assuming that all carriers 14 will reach their intermediate or final destinations. The next movement is performed without waiting for the route selection system 50 to calculate the next route. This allows for faster route selection of the carriers 14.

[0179] Based on the planning model, the route selection system 50 checks whether the carriers reach their stopping positions after each movement. Examples of carrier movements that can lead to a stalemate are shown in Figures 6A, 6B, and 6C. On the left side of Figure 6A, a stalemate is shown due to the fact that two opposing movements of carriers block each other. On the right side of Figure 6A, a circular dependency occurs where the movements of carriers mutually block each other.

[0180] Figure 6B shows a stalemate resolution method using random moves, where the carriers are moved in random directions to break the stalemate dependency of Figure 6A, after which a normal routing plan is again generated that allows the carriers to move forward again, as shown in Figure 6C.

[0181] 1, the route selection system 50 uses some nodes only as the final destination of the next planned trip. For example, nodes N1 and N7 are not used as intermediate destinations in the cooperative search. Because these nodes are handover locations with the devices 18, locations N1 and N7 must not be blocked by carriers that do not need to be processed by these devices. Unnecessarily blocking these locations would be disadvantageous in terms of the processing capacity of these devices.

[0182] The dotted lines in FIG. 1 divide the transport surface 12 into four logical subareas 24. The route selection system 50 allows only a limited number of simultaneous carrier movements within a logical subarea 24 of the transport surface 12. For example, the route selection system 50 counts the start of a movement within a logical subarea 24 and does not plan another movement for that logical subarea 24 to limit the number of simultaneous movements. The maximum number of simultaneous movements may be fixed for the entire system or defined per region. For example, a maximum of eight simultaneous carrier movements may be allowed per transport tile or software-defined region. Furthermore, the maximum number of simultaneous movements allowed may also be related to the number of logical positions within that region. For example, the maximum allowable simultaneous movements within a logical subarea 24 may be within the range of 1% to 70%, 5% to 50%, 10% to 30%, or 15% to 25% of the number of available logical positions within the logical subarea 24. This limits the maximum power consumption peak for that region, thereby allowing the use of fewer or smaller power supplies or protecting electronic circuits from rapid degradation or damage.

[0183] To model the movement of carriers on the transfer surface 12, the route selection system 50 uses planning models. FIG. 2A shows a speed-time diagram of the movement of a carrier 14. Acceleration, constant velocity movement, and deceleration may not be constant or perfectly repeatable due to friction on the transfer surface 12 and disturbances in the drive system caused by different environmental conditions, dust, dirt, or variable material friction, transitions between separate modules (e.g., slight gaps or steps), or other influences. The carrier 14 arrives at its intermediate or final destination location at an actual arrival time 30. FIG. 2A shows the actual velocity profile of the carrier over time. Each carrier may have a slightly different profile for each move. FIG. 2B shows three planning models with different parameters for acceleration, deceleration, and maximum attained velocity. The different planning models result in different planned arrival times 32.

[0184] Dashed line 34 represents a realistic model in which planned arrival time 32 closely resembles actual arrival time 30 in FIG. 2A. Dashed-dotted line 36 represents a conservative planning model. This planned arrival time 32 is much later than actual arrival time 30 in FIG. 3A. The uneven dashed line 38 represents an overly optimistic planning model. This planned arrival time 32 is much earlier than actual arrival time 30 in FIG. 3B. The planning model includes parameters for acceleration, deceleration, and maximum velocity. More complex parameterizations are possible, for example, dividing acceleration and deceleration into time intervals with different acceleration and deceleration values ​​or considering displacement derivatives rather than acceleration. Specifically, FIG. 2B shows velocity profiles for different models. Model 36 is conservative (assuming lower velocity and acceleration), model 34 is an excellent match, and model 38 is overly optimistic. Plans generated using model 38 fail too frequently, requiring adjustments when new updates become available from the hardware. Plans generated using model 36 may be successful, but may lose some performance due to unnecessarily long reservation times.

[0185] 3A and 3B show two options for planning models in a speed-time diagram, as solid lines compared to the actual journey as dashed lines. In FIG. 3A, the conservative model assumes a planned arrival time 32 that is later than the actual arrival time 30 after the journey. In FIG. 3B, the overly optimistic model provides a planned arrival 32 that is earlier than the actual arrival time 30. Therefore, the plan in 3B is considered to fail because the actual arrival time of the journey will be later than planned.

[0186] A good choice of planning model represents actual travel as it actually is. The planned arrival time 32 needs to be later than the actual arrival time. However, a conservative model is needed to avoid too many failed plans due to stochastic variations in travel times. Otherwise, carriers 14 will not reach intermediate or final destination locations during the duration of the planned reservation. This would require changes in the time plan for carriers that did not arrive within the planned time window and replanning of other affected carriers, reducing the efficiency of the route selection algorithm. If replanning were required frequently, it would also mean much higher computational power consumption for the router.

[0187] In one embodiment, the route selection system 50 adjusts the planning model if the number of failed plans exceeds a predetermined number, also referred to as a threshold. The predetermined number may be an absolute number, such as 5, 10, or 100, within a defined time interval, or may be a ratio of failed plans to created plans, such as 5%-50% failures, 10%-25% failures, or 1%-10% failures within a time interval.

[0188] The time interval may be, for example, the time required to move the carrier over the maximum reservation length, or a predetermined fixed time on the order of minutes, hours, or days.

[0189] In another embodiment, adjusting the model to increase the time of a trip does not necessarily have to be done in one step. The actual trip time is measured, and the model plan is updated (time is increased) by the percentage of discrepancy between the old modeled trip time and the measured trip time. In particular, the model can increase the time of a trip in small, fixed, incremental percentage steps (checking the number of trips), and adjust the planning model to make further increases until the plan no longer fails for a given period or number of trips.

[0190] In a further embodiment, the route selection system 50 adjusts the planning model to faster trips in predetermined steps if 100% of the planned trips reach their intermediate or final destinations.

[0191] For example, in another embodiment, route selection system 50 adjusts parameters of the planning model depending on the percentage of carriers that reach intermediate or final destination locations in time.

[0192] This means that the route selection system 50 compares the planned trips to the reached stops of the carrier 14. For example, if 5%-10% or 10%-25% of all planned trips result in the carrier not reaching the planned stop or intermediate or final destination within the planned reservation time window, the route selection system 50 uses a reduced maximum speed or acceleration / deceleration in the planning model. If 90%-100% of all planned trips result in the planned destination, the route selection system 50 uses an increased reachable speed in the planning model.

[0193] Another possibility is that the route selection system 50 or the control system 20, respectively, are configured to measure or calculate the deviation between the modeled travel time and the actual travel time and use this information to adjust the planning model.

[0194] Ideally, there would be no failed plans. Ideally, therefore, the system would look at actual travel times and adjust the model so that the modeled travel time is equal to the actual travel time plus a small positive delta time. In another embodiment, the route selection system incrementally changes the planning model to be more optimistic until the plan begins to fail, and then adopts the last setting that was 100% successful before the plan began to fail. By measuring the delta time between the actual travel time and the modeled time for a large number of trips, the model or lookup table can be adjusted to provide more accurate, but not too early, arrival times for trips.

[0195] The route selection system 50 can assign and coordinate planning models to individual carriers 14 and also to logical or hardware-based subareas 24 of the transport plane 12 .

[0196] In another embodiment, the route selection system 50 counts and registers the number of failed plans for each carrier, and if the number of failed plans exceeds a predetermined number, e.g., between 20 and 100, the route selection system 50 marks the failed carrier as requiring maintenance and directs the failed carrier to an area where the failed carrier is accessible to an operator or where the failed carrier can be screened and removed from the transfer surface 12. Alternatively, for carriers that exceed a threshold number of failed plans, the route selection system 50 can flag the carrier to be used by the control system 20 or command and management system 60 with a lower priority compared to other available carriers 14. Instead of flagging after a certain threshold of failed plans is reached, the control system 20 or command and management system 60 for the transfer system can also prioritize carriers with fewer failed plans over carriers with a higher number of failed plans.

[0197] In a further embodiment, the route selection system 50 counts the number of failed planned trips within a logical subarea and flags the area as "requiring maintenance" if, for example, 20%-50%, or 5%-25%, or 1%-5% of the planned trips fail. In yet a further embodiment, this flagging can be indicated to the operator via a user interface to inform them that maintenance is required and which transport module or area needs to be repaired. Alternatively, the route selection model can be modified only for the area where the plan failed to increase the reserved time for trips in that area.

[0198] In another embodiment, route selection system 50 assigns a usage cost to a node depending on the frequency of carrier departures, stops, and passages at that node. For example, to indicate the impact of each action on surface wear, a node may be assigned a cost of 2 for departures, a cost of 1 for passages, and a cost of 3 for passages. In its planning, route selection system 50 considers not only distance costs, as is typical in search algorithms, but also pass, departure, and stop costs to minimize the overall cost of each planned trip. The effect of this cost function on wear is a more balanced wear load on the surface, resulting in a longer life and better reliability of the transfer surface.

[0199] In a further embodiment, the route selection system 50 uses a reserve length so that the departure and intermediate or final destination locations of the carriers 14 on the transport surface are distributed as evenly as possible among the logical locations on the transport surface 12. Figures 4A, 4B, and 4C show three different possibilities for how a carrier 14 can move from a departure location 44 to a final destination location 46 with several intermediate destination locations 48 for a transport system with a small number of carriers 14 so that the maximum reservation length can be used for each move. The locations used are represented as hatched, white, or black rectangles. The hatched locations are the locations where the carriers start or stop.

[0200] In Figure 4, the maximum reservation length 40 is always used until the final move to reach the final destination location 46. Therefore, the hatched locations in the system are statistically used for stopping and departing carriers 14 more frequently than other locations. As a result, these hatched locations wear out faster than the black locations. Moves with the maximum reservation length are very common in areas with little cross traffic. As a result, in these areas, a pattern of high wear occurs at logical locations located at a distance from each other equal to the maximum reservation length.

[0201] In one embodiment, the route selection system is configured to assign usage costs to logical locations according to the frequency of carrier passage, departure, or stop at this logical location, minimizing usage costs for each planned trip, thus avoiding increased wear in patterns and achieving more distributed wear.

[0202] 7 depicts one possible embodiment of the interaction of the various software modules and hardware components. A higher-level command management system 70 transfers commands to and receives confirmation of command execution from the command management system 60. The higher-level command management system 70 may be middleware software or a laboratory information system (LIS).

[0203] The command and control system 60 communicates the arrival of carriers at the stations 18 and obtains necessary information for the stations from the stations 18, such as payload placement and whether the carrier is ready / available for processing. The command and control system 60 assigns objects 16 to carriers and translates the middleware / LIS / higher command management task into a task of bringing the associated carrier to the location of the station 18 for processing the object 16 carried by the carrier 14.

[0204] The command and control system 60 communicates equipment destinations to the control system 20 and receives confirmation of arrival at the final destination from the control system 20. The route selection system 50, shown here for this embodiment as part of the control system, proposes moves and routes to the control system 20. The control system 20 forwards move plans to the drive system 22 and receives status updates from the drive system 22, such as position and error updates. The drive system 22 moves the carrier 14 on the transfer surface 12.

[0205] The component parts, command and control system 60 and control system 20, do not necessarily have to be different computers. In fact, the higher-level command and control system 70 and / or LIS can also run on the same mainframe. Furthermore, command and control system 60, which can include route selection system 50, and control system 20 can be a single piece of software. For example, these different software modules can be integrated together or can be separate services that communicate with each other via a software bus on the same computer.

[0206] In further embodiments, other configurations are possible. For example, the route selection system 50 may be incorporated into the control system 20, and in other embodiments, the control system 20 may also be part of the transport table 12.

[0207] FIG. 8 depicts a flow diagram of a system 50 for routing tube carriers in a diagnostic laboratory on a two-dimensional transport surface 12.

[0208] In a start step 100, the route selection system obtains the final destination location of the carrier 14, typically from the dispatch control system 60. In a new destination check step 101, the route selection system checks whether the destination is new. If the destination is new, using this information, in a route selection step 102, the route selection system 50 calculates n movements on logical locations with required free time windows, e.g., n=5, starting within a time T, e.g., 10 seconds, toward the final destination location on the logical location with free time windows using the A* algorithm.

[0209] In one embodiment, the heuristic is based on the shortest distance to the final destination or the shortest time to travel to the final destination ignoring other carriers on the transfer surface.

[0210] Each planned move of the n moves has a reservation length of k logical positions, which is less than or equal to the maximum reservation length, for example, six logical positions, and greater than or equal to one.

[0211] In a next transmitting step 103 , the route selection system 50 transmits the plan for the trip to a transport system, such as the drive system of the transport surface 12 .

[0212] While the transport system 11 of the distribution system 10 executes the movement of the carrier 14 on the transport surface 12 in an execute step 202, the route selection system waits to receive a transport system status update in a "wait" step 104. The transport system sends a status update to the route selection system in a status send step 204, for example, every 60 ms. The time interval between status messages need not be constant. This also includes the status of the carrier, for example, whether a planned stop or destination or intermediate destination has been reached. In a check step 105, the route selection system compares the transport system status with the latest stored state to check whether the plan is still valid. If in the "check" step 105 the route selection system detects that a move is not possible, i.e., the plan has failed, the route selection system returns to the route selection step 102 and recalculates the n moves for time T.

[0213] If travel is still possible, route selection system 50 again checks whether a new destination is available in destination check step 101. If a new destination is available, route selection system 50 proceeds to route selection step 102 and plans a new route.

[0214] If a new destination is not available, the route selection system checks whether the plan is older than a predetermined duration, e.g., 2 seconds, in plan age check step 106. If the plan is older than the predetermined duration, the route selection system calculates a new plan, and therefore a route, for the next time window, e.g., spanning 10 seconds, in route selection step 102. If the plan is not older, the route selection system waits for a status update in wait step 104, and so on.

[0215] 9A, 9B, and 9C show possible embodiments for releasing reserved logical positions. One or more already passed reserved logical positions are released. The boundaries of the areas are marked as black logical positions. In FIG. 9A, six modules are connected to each other, and when the carrier passes the boundary between the modules, the reserved logical positions are released. The carrier moves from the starting point to the final destination T. When the carrier leaves the black logical position from the first transport module and passes the black logical position of the next transport module, indicated by B!, the logical position reserved for this movement on the first transport module is released. The reserved logical positions can also be released after the end of the movement, indicated by M!, and when the final destination position, indicated by T!, is reached.

[0216] For example, all locations for a move may remain reserved as long as the complete move is made. For example, locations may be reserved as long as the move continues within a particular logical area. For example, locations may be reserved until the carrier passes the location. For example, locations may be released as soon as each logical location is passed. For example, locations may be released when a boundary of a block of logical locations is passed.

[0217] Figure 9B shows the same transport system layout, but instead of the areas for releasing reserved logical positions being defined by module boundaries, the area boundaries are defined by marking positions in software. Release is triggered by a message passing through the defined area. Passing the boundary of an area and entering another area, as indicated by A!, releases the reserved logical position for this move in the area to the left.

[0218] Figure 9C shows a direct release where a logical location release message is sent at each passing logical location, so each logical location on the route defines a unique release area.

[0219] Figures 10A and 10B show the effect of masking logic locations, for example, in the event of a field or module failure. Both Figures 10A and 10B show the same layout of a transfer surface 12 with 2x4 modules. The movement of each of the carriers is indicated by an arrow.

[0220] Figure 10A shows the situation where all logical locations are available. A route is shown from the starting point to the final destination consisting of six trips with different amounts of logical locations per trip.

[0221] On the other hand, Figure 10B shows a situation where, for example, the entire upper right module 80 needs to be replaced and the 3x8 logical location area 82 is also masked for transfers. Route selection system 50 takes into account the unavailability of the locations marked with an X, for example by setting infinite reservation times for those locations or by updating the location layout and attempting to calculate a route over the available logical locations. This results in a route consisting of eight moves from the same starting point to the same final destination as in Figure 10A.

[0222] 11 illustrates an embodiment of distribution system 10. Distribution system 10 comprises a route selection system 50 and an execution unit 51. While the route selection algorithms used to plan routes for carriers 14 on transfer surfaces 12 to transport payloads 19, for example, may be deployed in the same physical computer, this division between route selection system 50 and execution unit 51 allows route selection system 50 to only propose moves. Execution unit 51 attempts to execute the planned route, but due to stochastic effects, exceptions, and errors, it may not be able to execute all plans as prescribed.

[0223] FIG. 12 shows a Gantt diagram illustrating an example of travel (execution) time versus planned time for carriers numbered 1 through 5. If the execution time is within the planned time, the planned route can be executed. If not, the trip may exceed the planned time, interfering with other planned trips and causing the plan to fail. From top to bottom, carriers 1 and 2 illustrate an embodiment of successful plan execution. Carrier 3 shows that the start of a trip may be later than planned, for example, due to a delay in another carrier, but the carrier still arrives in time, thus successfully executing the plan. Carrier 4 shows two delayed arrivals, and carrier 5 shows that the final arrival is delayed.

[0224] Figure 13 illustrates the stochastic influence of movement on route selection. Transport methods in which the carrier 14 is driven by a combination of electromagnetic force and friction exhibit stochasticity. In general, friction forces between sliding surfaces can be a significant variable. Furthermore, the surface of the carrier and the transport sliding surface may not be uniform across the entire surface. For example, the roughness of the transport surface of the transport surface 12 may vary due to differences in wear, uneven materials, dirt or dust, or moisture. Also, some module surface plates may be more recently updated than others and exhibit different friction levels than older plates. The same is true for the carrier's contact surface with the transport surface. As a result of these varying friction forces, the time required for movement also has an unpredictable stochastic component. If friction is greater than the average friction, the carrier will accelerate more slowly and may not reach the same steady-state speed. The carrier will also decelerate more quickly. The execution unit controlling the movement attempts to accommodate this, but fluctuations in acceleration and velocity profiles are still observed for the movement and the carrier 14. An example of the effect of stochastic influences is shown in Figure 13. The top diagram shows a simplified representation of the effect of different carrier friction levels for Carrier 1 with low friction (dotted line), a model considering moderate friction (solid line), and Carrier 2 with high friction (dashed line). The lines represent the distance traveled by a carrier moving at a constant speed. In reality, these are not straight lines but rather may exhibit random fluctuations (whose average speed is simplified to resemble these lines). If Carrier 1 has lower friction than the deterministic model assumes, it will move faster and therefore faster than the more pessimistic model of Carrier 2, which exhibits more friction. The same can happen with different transfer surfaces, as shown in the middle diagram. The middle diagram shows the corresponding effects for a carrier on a low-friction transfer surface (dotted line), a model considering moderate friction (solid line), and a carrier on a high-friction transfer surface (dashed line).The bottom diagram shows an embodiment of localized friction differences on a transport surface, with a carrier moving across a transport surface with constant friction shown in dotted lines, a model assuming constant friction shown in solid lines, and a carrier moving across a transport surface with localized areas of higher friction, such as carrier 2, shown in dashed lines. In particular, the bottom diagram shows the effect of a carrier moving a distance with constant friction, a more pessimistic deterministic model assuming lower velocity, and a carrier movement situation where the beginning and end portions have the same friction, but the area between them has higher friction and therefore a lower average velocity of the carrier.

[0225] Figure 14 shows the results of stochastic influence on trips. Probability results in variability in trip times. Figure 14 shows an example of what can happen, where a deterministic model assumes various trip lengths. Here, a maximum trip length (maximum reservation length) of 6 is assumed. Over multiple measurements, the values ​​are as shown by the light gray bars, and the average trip time is shorter than the model assumes. However, for the two- and four-location trips, some trips took longer than the model predicted. Therefore, for these trips, the plan fails because the arrival time at the trip's end location is slower than planned.

[0226] FIG. 15 shows the discrete-time WHCA* algorithm for a game such as that described in B. Zahy et al., "Conflict-Oriented Windowed Hierarchical Cooperative A," 2014 IEEE International Conference on Robotics and Automation (ICRA), IEEE, May 31, 2014, pages 3743-3748, XP03265061 8, DO1:10.1 109 / ICRA.2014.6907401, compared with the continuous WHCA* algorithm proposed in accordance with the present invention.

[0227] While the duration of events in games is typically discretized, the present invention proposes using continuous time. In games, discrete time steps are typically related to the game's frame rate. In a route selection system according to the present invention, it is likely not necessary to use discrete time steps. Continuous time can enable more efficient route selection algorithms because movement can begin at any time. Figure 15 illustrates the difference between the discrete time and space in games and the present invention, which is discrete in space but continuous in time. Time is shown on the horizontal axis, and logical positions are shown on the vertical axis, with logical positions being reserved (gray) or free (white) at each time. When a carrier moves from position 1 to 2, a discrete-time route selection method may result in slower speeds because each reservation plan must be aligned with a discrete time step. In contrast, if any time is available, subsequent planning can be performed to fit the movement model without having to take discrete time steps into account.

[0228] In games, the game engine generates events. This means that all events are self-defined and therefore predictable. However, in the technical field of carrier distribution systems for transporting objects, the real world must be taken into account. In the real world, there are unpredictable factors such as stochastic effects on movement; new movement orders may be unexpectedly brought about; equipment may deploy or request carriers; and exceptions and errors may occur, such as fields being damaged, carriers no longer being able to move, areas being temporarily blocked for route selection, movement taking much longer than expected, or movement failing altogether. The route selection system cannot influence these events but must be able to react to them. Therefore, the route selection system in the present invention can be designed to perform route selection in a more robust manner so that it can find a solution to resolve any unexpected situation. In games, unpredictable events may arise from the player. However, although these events are unpredictable, they are all executed through the game engine, which generates route selection based on this "unpredictable" information. This means that the router in the game has access to all information before generating the route, and the route is generated so that the agent's movement can be executed according to the plan based on this completely known situation. The situation is different for a route selection system according to the present invention, which can only receive information about destinations to which the carrier should be moved, as well as updates about the success and failure of the moves, and ultimately other status information. The route selection system has no control over real-world events and can only wait to see which plans will succeed or not, and when the moves will end. Thus, in contrast to a game engine, the route selection system may be designed to be robust to handle the unexpected.

[0229] In the technical field of distribution systems for carriers to transport objects, the number of agents, i.e., carriers in a distribution system, can exceed 1,000. In games, this is typically much less. Furthermore, carrier density can be extremely high, to the point where many carriers are blocked by neighboring carriers and cannot move until traffic density decreases. In games, the space available for agents to move is typically unconstrained, and as mentioned above, the game engine has complete control and is aware of what is occurring. Due to this lower interaction between units, most games (e.g., StarCraft II) do not consider multi-agent pathfinding, but instead use a variant of A* for single-agent pathfinding combined with some kind of "steering behavior" heuristic that slightly adjusts trajectories if two units happen to cross paths. Thus, the technical fields of games and distribution systems for carriers to transport objects are quite different, have different requirements, and therefore require different specific solutions.

[0230] FIG. 16 illustrates possible trips. "S" may be the carrier's starting point where the carrier remains until tasked with moving to final destination "E." A route selection system may generate a plan of trips. Each trip may end at an intermediate destination, indicated by a circle. The route selection system may plan ahead a predetermined amount of time, such as 10 seconds. Thus, in a first plan, only the route portion from M1 to M3 may be planned. In subsequent plans, other trips may be planned. Each trip may have a length of a predetermined number of positions reserved in the time plan. During plan execution, planned positions for the next trip may be reserved, as indicated by the shaded position for trip M6. For example, the number of positions that can be reserved in a plan during plan execution depends on which positions are available (e.g., not currently occupied by another carrier) and the maximum reservation length. For example, in this example, the maximum reservation length may be six positions. Only the location reservation for movement M2 may be limited by this maximum reservation length, other movements are shorter and are limited by the availability of locations at this time due to, for example, other carriers (not shown in this figure). Other traffic can also be a reason for a carrier to change direction for movement M4, for example because another carrier blocks the possibility of continuing a linear movement.

[0231] The time it takes to complete the transfer from "S" to "E" may depend on the time taken for each move and the waiting time at the intermediate destinations indicated by the circles. The move time may depend on the move length (number of positions for each move) and the velocity profile (acceleration, steady velocity, deceleration). While the velocity, acceleration, and deceleration may always be the same (deterministic) for each move, the present invention is well suited to handle stochastic variations in velocity, acceleration, and deceleration as well. Thus, moves of the same move length may take different times. [Explanation of symbols]

[0232] 1 Module 1 2 Module 2 3 Module 3 4. First Move 5. Second Move 6. Third Move 7 Second Router 8 Intersection Routes 10 Distribution System 11 Transport System 12 Transfer surface 13. First Reserved Table 14 Career 15 Second Reserved Table 16 Object 18 Stations 19 Payload 20 Control System 21 Camera System 24 Logical Subareas 25 First Move of Second Route 26 Second Movement of Second Route 27 Third Movement of the Second Route 30 Actual arrival time 32 Planned Arrival Time 34 dashed line 36 Dotted Line 38 Uneven dashed lines 40 Maximum reserved length 42 Variable Reserve Length 44 Starting position 46 Final destination 48 Intermediate Destinations 50 Route Selection System 51 Execution Unit 60 Command and Control System 70 Upper Management System 80 Top right module 82 Area 100 Starting Steps 101 Destination Check Step 102 Route Selection Steps 103 Sending Step 104 Wait Step 105 Check Step 106 Planned Life Check Step 202 Execution Steps 204 Status Sending Step

Claims

1. 1. A distribution system comprising: a transfer surface having a plurality of logical locations for dispensing a plurality of objects; a plurality of carriers for transporting said objects; a drive system for moving the carrier between the logical positions on the transfer surface; a control system configured to control the carrier to move on the transport surface from a start location to a final destination location along a planned route; Equipped with the control system comprises a route selection system configured to calculate planned routes for at least two carriers on the transportation surface by modeling the transportation surface with a graph of nodes and using a knowledge-based windowed hierarchical cooperative search algorithm in a cooperative time window T, the knowledge-based windowed hierarchical cooperative search algorithm being a windowed hierarchical cooperative A* algorithm; the route selection system is configured to determine a reserved time window and an available time window for each logical location; the route selection system is configured to assign a separate reservation length to each carrier for a next movement for a logical location having an open time window; One logical position on the transfer surface is assigned an infinite reservation time; The route selection system comprises: The logical location is the location of a carrier that does not have a final destination location. A logical location is considered to be corrupted or blocked, and the logical location is considered corrupted when it can no longer be used due to a drive system failure at a particular location of the transport surface and / or a transport surface failure, etc.; no possibility of moving the carrier from its logical location is observed during the cooperation time window T; is configured to assign an infinite reservation time to the logical position when one or more of the following are realized: The control system comprises at least one execution unit configured to execute the planned routes to transport the carriers from their respective start locations to their respective final destination locations.

2. The distribution system of claim 1 , wherein the route selection system is configured to use the largest available reservation length for linear travel that is less than the determined maximum reservation length.

3. the route selection system receives transportation system status information; 3. A distribution system as claimed in claim 1 or 2, configured to compare the status information with latest status information stored in the route selection system and, if the comparison detects a change in status, calculate a new planned route with a new reservation length for the next journey.

4. 4. A distribution system according to any one of claims 1 to 3, wherein the route selection system is configured to determine subsequent planned routes after a current journey, assuming each journey is successful.

5. 5. The distribution system according to claim 1, wherein the route selection system is configured to check whether all carriers have arrived within their planned reservation lengths, and if a carrier is detected that has not arrived within the planned reservation lengths, to extend the next planned reservation length and recalculate the plans of other affected carriers.

6. 6. The distribution system of claim 1, wherein the route selection system is configured to allow at least one of the logical locations to be addressed by the carrier only as a final destination location and to prohibit its use as an intermediate destination location for transporting the carrier to the final destination location.

7. A distribution system according to any one of claims 1 to 6, wherein the transport surface is divided into logical sub-areas.

8. 8. The distribution system of claim 7, wherein the route selection system is configured to allow only a limited number of simultaneous movements of carriers within a logical subarea of ​​the transport surface.

9. 9. The distribution system of claim 7 or 8, wherein the route selection system is configured to release each logical position reserved by a current movement after the carrier passes a predetermined logical position on the transport surface.

10. the route selection system is configured to check the success or failure of the planned trips, and depending on the number of failed plans, parameters of a planning model are adjusted by the same value for the entire transport surface, or by different values ​​for each of the logical subareas, or by individual parameters for carriers of a particular subarea; 10. A distribution system according to any one of claims 7 to 9, wherein if the carrier reaches the destination in time, the movement speed of the carrier in the planning model is increased, or if the carrier does not reach the destination within the planned time, the maximum speed is reduced in the planning model, or the acceleration / deceleration values ​​are changed depending on whether only short movements fail, the speed value is kept the same, or if only plans for longer movements fail, only the speed value is adjusted.

11. 4. The distribution system of claim 3, wherein the route selection system is configured to adjust the maximum reservation length in response to a deviation between the maximum reservation length and the planned reservation length.

12. the route selection system is configured to count the number of failed plans for each carrier and / or subarea of ​​the transport plane; If the number of failed plans exceeds a predetermined number or the frequency of failure exceeds a predetermined number, the carrier is marked as needing maintenance; or A distribution system according to any one of claims 1 to 11, wherein the transport surface or the sub-area is marked as needing maintenance and / or is avoided or of low use for further routing.

13. 13. The distribution system of claim 1, wherein the route selection system is configured to assign usage costs to logical locations according to the frequency of carrier passage, departure, or stop at the logical locations, and to minimize the usage costs for each planned trip.

14. 14. The distribution system of claim 1, wherein the route selection system is configured to check whether a stalemate has occurred and, if a stalemate is detected, to randomly select the next n trips from among the possible trips for all carriers in the vicinity of the stalemate, disregarding metrics to the final destination location.

Images