Optimal fleet composition and deployment tool for autonomous mobile robots

Optimizing AMR path and task allocation through branch bounding and dual graph methods, combining multi-processors and stagnant incremental algorithms to optimize fleet composition, solve the problem of low material flow throughput in automation facilities and improve delivery efficiency.

CN119065364APending Publication Date: 2024-12-03FORD GLOBAL TECH LLC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410670387.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-06-02
Filing Date
2024-05-28
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

In automation facilities, autonomous mobile robot (AMR) path planning, fleet composition and task allocation are complex, resulting in low material flow throughput and difficult to effectively optimize the existing technology.

Method used

The branch bounding (B&B) method is used to determine the waypoint sequence and path, and the path cost is optimized through the dual graph method. Multiple processors are used to divide the search space to determine task allocation and fleet composition, and the fleet size and composition are optimized in combination with the stagnant incremental algorithm.

Benefits of technology

Improve material flow throughput, and reduces robot travel, turn, wait and charge times by optimizing paths and task allocation, achieving more efficient material delivery.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119065364A_ABST
    Figure CN119065364A_ABST
Patent Text Reader

Abstract

The invention provides an optimal fleet composition and deployment tool for autonomous mobile robots. A method of managing a robotic fleet for delivering material in a facility is provided. The method comprises: delimiting by a branch delimiting (Bamp; b) determining a waypoint sequence by a method; determining a path passing through the waypoint sequence through a dual graph method; the method comprises the following steps: passing through Bamp; the method B is used for determining task allocation among the robots; and through the Bamp; and B, determining the composition of the motorcade.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to systems and methods for managing the delivery of materials in an automated facility, and more particularly to systems and methods for managing the delivery of materials by a fleet of robotic vehicles in an automated facility. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.

[0003] Autonomous mobile robots (AMRs) have been used in production lines to deliver materials to different task stations. Robot fleets are often used to pick up and drop materials between different locations in a factory or warehouse. The robots can be multi-load robots that can pick up and drop multiple items at multiple locations simultaneously in a single task. Once a task is assigned an ordered list of pick-up and drop-off locations, the robot fleet autonomously navigates through the pick-up and drop-off locations in a prescribed order.

[0004] The material flow throughput of a facility depends on the path of the robots navigating through these pick-up locations and drop-off locations. Material flow throughput can be enhanced by reducing the time spent by the robots traveling, turning, waiting, and downtime. Path planning for a fleet of robots becomes more complex when multi-load robots are used. In addition, the material flow throughput of a facility also depends on the fleet composition, which includes the size of the fleet and the types of AMRs in the fleet.

[0005] Issues associated with using AMRs to handle material delivery in automated facilities are addressed in the present disclosure, particularly path planning for AMRs, fleet formation, and allocation of tasks between AMRs. Summary of the invention

[0006] This section provides a general summary of the disclosure and is not a comprehensive disclosure of its full scope or all of its features.

[0007] In one form, a method of managing a fleet of robots for delivering materials in a facility is provided. The method includes: determining a waypoint sequence by a branch and bound (B&B) method; determining a path through the waypoint sequence by a dual graph method; determining task allocation among robots by the B&B method; and determining fleet composition by the B&B method.

[0008] In other features, the method also includes determining a small candidate fleet composition having the most efficient fleet composition by using a stagnant incremental method. Determining a path through a sequence of waypoints includes determining a total path cost for the path. The total path cost includes a traversal cost and a pivot cost. The traversal cost is a cost associated with translational movement of the robot fleet. The pivot cost is a cost associated with changing the direction of the robot and the time spent changing the direction. Determining a path through a sequence of waypoints includes selecting a path with a lowest total path cost. Determining a sequence of waypoints includes determining movement constraints. Determining a task allocation among robots includes partitioning a search space for task allocation by using a plurality of processors. The fleet composition includes the number of robots in the fleet and the type of robots. The robot fleet includes a plurality of autonomous mobile robots (AMRs).

[0009] In other features, the method further includes determining a waypoint sequence by considering three waypoints staggered at a time. The waypoints include a pick-up location and a drop-off location where the robot picks up and drops an item, respectively.

[0010] In another form, a method for managing a fleet of robots for delivering materials in a facility is provided. The method includes: determining a waypoint sequence by a branch and bound (B&B) method based on movement constraints; determining a path through the waypoint sequence by a dual graph method based on a total path cost; determining task allocation between robots by the B&B method; and determining a fleet composition by the B&B method.

[0011] In other features, the method also includes determining a waypoint sequence by considering three staggered waypoints at a time, and partitioning the search space for task allocation by using multiple processors. The fleet composition includes the number of robots in the fleet and the type of robots. The robot fleet includes a plurality of autonomous mobile robots (AMRs). Determining a path through the waypoints based on the dual graph also includes selecting a path with a minimum total path cost. The waypoints include a pick-up location and a drop-off location where the robot picks up and drops an item, respectively.

[0012] Further areas of applicability will become apparent from the description provided herein.It should be understood that the description and specific examples are intended for purposes of illustration only and are not intended to limit the scope of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order that the present disclosure may be better understood, various forms of the present disclosure will now be described by way of example with reference to the accompanying drawings, in which:

[0014] Figure 1 is a schematic diagram of a fleet management system according to the teachings of the present disclosure;

[0015] Figure 2 is a schematic diagram of the layout of the automation facility;

[0016] Figure 3 is a diagram of two possible paths for a robot tasked with moving an object from one waypoint to another;

[0017] Figure 4 yes Figure 3 The dual graph representation of the path;

[0018] Figure 5 is a schematic diagram of the optimal path for a robot to navigate through ten waypoints when considering two consecutive waypoints at a time for path planning;

[0019] Figure 6 is a schematic diagram of the optimal path for a robot to navigate through ten waypoints when the entire waypoint sequence or three waypoints interleaved are considered at once for path planning;

[0020] Figure 7 is a schematic diagram of four waypoints with constraints on the waypoint sequence, where points (a) and (c) are pick-up locations, and points (b) and (d) are corresponding drop locations;

[0021] Figure 8 It shows that the robot navigation based on the branch and bound algorithm is Figure 7 A tree diagram of all possible paths to the four waypoints.

[0022] Fig. 9 is a schematic diagram showing how to divide the tasks and assign them to each robot;

[0023] Fig.10 is a tree diagram showing all possible task assignments to multiple robots based on a branch and bound algorithm;

[0024] Fig.11 is similar to Fig.10 A tree diagram of a tree diagram showing a parallel implementation of a branch and bound algorithm for assigning tasks to multiple robots;

[0025] Fig.12 is a diagram showing how the Stagnant Increment Algorithm is used to determine the optimal fleet size and composition; and

[0026] Fig.13 A method of managing a fleet of robots navigating through multiple waypoints in a nested optimization structure in accordance with the teachings of the present disclosure.

[0027] The drawings described herein are for illustration purposes only and are not intended to limit the scope of the present disclosure in any way. DETAILED DESCRIPTION

[0028] The following description is merely exemplary in nature and is not intended to limit the present disclosure, application, or uses. It should be understood that throughout the drawings, corresponding reference numerals indicate the same or corresponding parts and features.

[0029] refer to Figure 1 A fleet management system 20 for managing material delivery in a facility according to the teachings of the present disclosure includes a robotic fleet 24 configured to autonomously deliver materials to and from various locations in the facility and a fleet management module 26 configured to manage and control the delivery of materials by the robotic fleet 24 in the facility.

[0030] The fleet management module 26 includes a memory 30, a path planning module 32, a sequence determination module 34, a task allocation module 36, a fleet planner module 38, and a fleet selection and operation module 40. The memory 40 is configured to store information about the layout of the facility, the constraints on the movement of the robots, and the types of robots in the robot fleet 24. Movement constraints refer to factors that affect the movement planning of the robots in the facility, such as the time window of each pick-up or delivery location, the state of charge (SOC) of the robot's battery, and the required order between the pick-up and delivery locations. The robot fleet 24 includes autonomous mobile robots (AMRs) that can autonomously travel to various locations in the facility according to a specified path. The robots can be of any available model and can be of the same type, different types, or a combination of any available models. For example, a robot fleet can include two robots of one type and four robots of another type. Figure 1 The robots shown are for illustration purposes only and the solutions proposed in this application are not limited to Figure 1 Each robot can be configured to pick up multiple items from multiple pick-up locations and drop multiple items at multiple drop-off locations.

[0031] refer to Figure 2 , the facility may be a factory 22 that manufactures products such as vehicles. The shaded areas represent various features of the factory layout, which may include inventory holding locations, dedicated workstations, and robot charging or idling bays. The robots are responsible for the movement of materials between the different areas. The robots are allowed to travel along paths connecting the different shaded areas.

[0032] The paths available to the robotic fleet 24 can be represented by a node graph that discretizes the paths of the robotic fleet 24 into a plurality of nodes. The nodes represent the physical locations of intersections of the paths and material handling locations of interest, i.e., the pick-up locations and drop-off locations of items to be transported by the robotic fleet 24.

[0033] The path planning module 32 is configured to determine the total path cost of all possible paths by using a dual graph approach in order to identify the path with the lowest cost (i.e., the most efficient path) that the robot that must navigate through a defined sequence of waypoints will take. Waypoints are material handling nodes and are restricted to pick-up locations and drop-off locations where the fleet of robots 24 pick up and drop off items, respectively. The total path cost includes a traversal cost and a pivot cost (or turn cost, vertex cost). The traversal cost refers to the cost associated with translational movement between consecutive waypoints. The pivot cost refers to the cost associated with the robot turning to change the direction of travel of the robot.

[0034] refer to Figure 3 , we now explain the traversal costs and pivot costs associated with navigating a robot through consecutive waypoints. Figure 3 In the illustrative example of , Path A and Path B are two possible paths for a robot to travel from point (a) to point (b). The robot is oriented to head north in the direction of travel for Path A. The distance traveled for either Path A or Path B is the same, and therefore the traversal costs for Path A and Path B are the same. However, the robot needs to make one 90° turn for Path A and four 90° turns for Path B to travel from point (a) to point (b). Path B uses more energy and takes more time when changing the direction of the robot. Therefore, Path B has a higher pivot cost than Path A.

[0035] The material flow throughput of a facility can be enhanced by reducing the time a robot spends traveling, turning, waiting, and charging while navigating through an assigned sequence of waypoints. Thus, the optimal path is the most efficient path with the lowest total path cost (including traversal cost and pivoting cost).

[0036] refer to Figure 4 , in order to consider pivot costs more effectively, a dual graph method is used. Dual graph is a method in space syntax that considers edges as nodes and nodes as edges. In a road network, large avenues consisting of several road segments become signal nodes, and intersections with other avenues or streets become links (edges). Edges are weighted by the total cost of traversal and orientation changes. This method is useful for revealing hierarchical structures in planar networks. Based on planar networks such as city streets (A), space syntax proposes to consider line segments in a different way from traditional graph theory, where links (edges) are streets and intersections are nodes (vertices). The optimal path with the lowest total cost can be obtained by numerous polynomial time algorithms (such as Dijkstra's algorithm) or the A* algorithm (if there is an acceptable heuristic).

[0037] exist Figure 3In an illustrative example with only two waypoints, Path A and Path B each have a travel distance of 140 m. Path A requires one 90° turn, while Path B requires four 90° turns. Path A consumes 214 kJ of power, while Path B consumes 244 kJ of power, 14% more than Path A. The time required for Path A is 87.2 seconds, while the time required for Path B is 89.6 seconds, 3% more than Path A. Therefore, Path B has a higher total path cost than Path A. These values ​​are for illustrative purposes only and are not intended to limit the scope of the present disclosure.

[0038] refer to Figure 5 and Figure 6 For facilities with more than two waypoints (such as ten waypoints in the illustrative example), the optimal path (i.e., the path with the lowest total path cost) based on the dual graph method can depend on the number of waypoints considered at one time. Figure 5 The optimal path is shown when only two consecutive waypoints are considered at a time. Figure 6 The optimal path is shown when all ten waypoints are considered at once. Based on dual graph analysis, Figure 6 The path in can achieve a 10% reduction in total path cost. It should be understood that the values ​​are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Therefore, in order to find the optimal path, all sequences of ten waypoints should be considered at once.

[0039] For multi-payload tasks, assuming that the path between consecutive waypoints is independent of other waypoints creates unnecessary constraints on the robot orientation. This produces suboptimal solutions because the optimal path between waypoints depends on the robot orientation at each waypoint, which is unconstrained in real problems.

[0040] The dual graph approach can be used to find an optimal path for a multi-payload robot through a sequence of waypoints in space. The waypoint sequence represents a method that can take into account time, energy, or any relevant cost associated with moving the robot through the environment. This includes vertex costs such as turning costs, waiting times at intersections or when merging into higher speed traffic. Depending on the environment waypoint sequence and the cost function, a substantial reduction in path cost can be achieved when compared to methods that only use two waypoints at a time.

[0041] However, when the number of nodes in the waypoint sequence representation is large, it may cause computational problems if the sequence of all waypoints is considered at once. Case studies have shown that the staggered three waypoint method can be used instead. The staggered three waypoint method is an approximation method that can obtain a near-optimal path by decomposing the waypoint sequence problem into sub-problems containing three consecutive waypoints. It has been determined that considering a sequence of waypoints instead of just considering multiple pairs of consecutive waypoints results in fewer and smaller turns. This is because the two waypoint method does not have the look-ahead capability of the staggered three waypoint or global method. The staggered three waypoint method scales better than the global method. Case studies have shown that path planning using three consecutive waypoints at a time is computationally more efficient and has no optimality gap with the solution found by considering all waypoints at once, as shown in Table 1 below. The values ​​are for illustrative purposes only and are not intended to limit the scope of the present disclosure.

[0042]

[0043]

[0044] Table 1

[0045] refer to Figure 7 After the path planning module 32 determines the total path cost of all possible paths, this information is sent to the sequence determination module 34, which is configured to determine the waypoint sequence while taking into account the movement constraints. Figure 7 In the illustrative example of FIG. 1 , the path of the robot may include pick-up points (a), (c) for picking up materials and drop-off points (b), (d) for dropping materials. One or more pick-up points must be visited before the robot fleet 24 travels to the corresponding one or more drop-off points. Therefore, the sequence of some of the pick-up locations and drop-off locations imposes constraints on the path planning.

[0046] The sequence determination module 34 is configured to determine the path of the robot by using a branch and bound (B&B) algorithm while taking into account these constraints. B&B is a method for solving optimization problems by breaking the optimization problem into smaller sub-problems and using bounding functions to eliminate sub-problems that cannot contain the optimal solution. The B&B algorithm systematically divides the solution search space into subsets arranged in a tree structure. The root of the tree is the original problem, and the leaves of the tree are individual candidate solutions to the original problem. Between the root and the leaves are intermediate nodes, which represent sub-problems obtained by recursively dividing the original problem through a process called branching. The order in which these sub-problems are examined is determined by a best-first selection criterion, which first explores the sub-problems with the lowest cost (i.e., utilization).

[0047] For the minimization problem, the upper limit is the current solution defined as the most efficient candidate solution to the original problem found at the leaf node. The upper limit is constantly updated as the tree is explored, and is used to prune suboptimal branches without recursively evaluating their solutions until the leaf node. Therefore, when the algorithm searches from the root to the leaf, it branches only when the cost at the node is lower than the current solution, and the branch can potentially find a better solution than the current solution. After this process, the B&B algorithm recursively decomposes the original problem until further branching is invalid when the solution cannot be improved, or until the original problem has been solved when each feasible branch has been evaluated.

[0048] Combination Figure 7 refer to Figure 8 , as shown in the illustrative example, points (a) and (c) are pick-up locations, and points (b) and (d) are drop-off locations, and therefore the robot must first travel to point (a) or point (c) to pick up an item before traveling to point (b) or point (d) to drop an item. Given this constraint, the B&B algorithm is used to analyze multiple possible sequences. The B&B algorithm explores all possible sequences in the form of branches of a tree. For example, six possible sequences are shown. The first possible sequence is (a)—>(c)—>(d)—>(b), which has a total path cost of 227. The second possible sequence is (a)—>(c)—>(b)—>(d), which has a total path cost of 214. During the exploration of the sequence, the estimated upper and lower bounds of the optimal solution are checked. Therefore, during the exploration of the third possible sequence (a)—>(b)—>(c), before exploring the fourth node (d), the path cost 215 is checked against the estimated lower bound 214 of the second possible sequence. Since the third possible sequence cannot produce a better solution than the second possible sequence due to the higher total path cost, the third possible sequence is discarded. It should be understood that if the initial path evaluated (i.e., the first possible sequence) is well selected (i.e., has a relatively low total path cost), more sequence branches can be cut off earlier, thereby reducing the number of calculations of the B&B algorithm.

[0049] refer to Fig. 9 After determining the sequence of waypoints to be visited by the robot fleet 24, this information is sent to the task assignment module 36, which is configured to assign tasks to the robot fleet 24. In the illustrative example, twelve tasks will be assigned to a fleet of four robots.

[0050] Consider a known layout with various spatial constraints and a set of material handling tasks Each task involves picking up certain items of goods at a stock location and dropping them off at their designated drop-off location within a limited time window. The goal of this step is to find the optimal solution that completes all defined tasks. The optimal fleet of multi-load capacitive AMRs that minimizes both fixed and operating costs is described in the following equation:

[0051]

[0052] Among them J 0 (k) is the operating cost, is the minimum operating cost, k is the number of robots in the fleet, R k is a set of robots of type, k, r is a set R k The robot in is the set of tasks assigned to robot r.

[0053] The minimum operating cost Jr(r,Tr) for each robot in fleet k depends on the AMR type and is also affected by the sequence in which the task locations are visited. This is because a robot can pick up multiple items and then drop them off, as long as each pickup point is visited before the corresponding drop.

[0054] refer to Fig.10 The task allocation module 36 is configured to use the B&B algorithm to determine how to allocate tasks to the robots. Determining the fleet composition and the tasks assigned to each robot is an NP-hard problem.

[0055] Fleet composition refers to the composition of the robots in the fleet, including the number of robots (i.e., the size of the fleet) and the types of robots included in the fleet 24. Fleet composition optimization involves determining the fleet size and the type of AMRs to be used. The B&B algorithm is used to obtain the optimal solution to this NP-hard problem.

[0056] refer to Fig.11 , in order to better track the subproblems, the task allocation search space is divided among multiple computational processors including processor 1, processor 2, ... processor p. The pooled optimal cost allows faster branch pruning across the search space. The B&B algorithm for the NP-hard resource allocation problem is implemented in a parallel framework using p processing cores, where the robots in the fleet are identified by subscripts r∈R:={1,2,...,m}. Therefore, several subproblems are explored simultaneously by splitting the tree graph at a certain task allocation level and assigning the emitted subtrees to the available processors. During the exploration of each processor, the updated current solution found is instantly available to each processor in an asynchronous information sharing method using a shared worker pool.

[0057] For each processor, the B&B algorithm is implemented by a recursive function to minimize memory and computation requirements while exploring the tree. In addition, since the computation time of the B&B algorithm increases with the number of feasible branches at each node, the fleet is initiated with a candidate fleet that is smaller than the maximum fleet. After evaluating the total cost of this candidate fleet, the number of robots is increased incrementally until further increments do not reduce the total cost or the additional robots remain idle. For each fleet increment, only the task allocation subproblem that includes at least one of the newly added robots is evaluated, because it is guaranteed that other solutions have already been evaluated.

[0058] refer to Fig.12 , after the task assignment module 36 determines the number of tasks to be performed by the robots and the number of robots to be included in the fleet, the fleet planner module 38 then determines the fleet composition and the tasks to be performed by the fleet of robots. Instead of initializing the problem with the maximum possible fleet size, the fleet planner module 38 starts with a smaller fleet and finds its cost J by solving a nested problem, and then incrementally adds to the fleet until further additions do not reduce the total cost J. The optimal fleet size and composition problem is a nested optimization problem and can be solved by using a stagnant incremental algorithm. The goal of the overall algorithm is to find the smallest possible fleet that can complete all tasks while minimizing the operating cost. If adding a robot to the fleet does not increase the total cost, then any larger fleet will result in a suboptimal solution. There is no need to explore the total cost further and the algorithm can be stopped.

[0059] After determining the optimal fleet composition, the optimal task allocation between robots, the optimal waypoint sequence, and the optimal path through the waypoint sequence, the fleet selection and operation module 40 manages and controls the fleet of robots to move in the automated facility accordingly.

[0060] refer to Fig.13, a method 80 for managing a fleet of robots through a plurality of waypoints in a facility begins by initializing a small candidate fleet composition. After initializing with the small candidate fleet composition, the most efficient fleet composition (i.e., the fleet composition with the lowest cost) for completing the task is determined in step 82 by using a stuck incremental algorithm. For each candidate fleet composition explored in step 82, the most efficient task allocation between robots (i.e., the task allocation with the lowest cost) is determined in step 84 by using a branch and bound algorithm. For each candidate task allocation explored in step 84, the most efficient waypoint sequence (i.e., the waypoint sequence with the lowest cost) is determined in step 86 by using a branch and bound algorithm. For each candidate position sequence explored in step 86, the most efficient path through the factory layout is determined in step 88 by using a dual graph transformation method. After determining the most efficient task allocation, the most efficient waypoint sequence, and the most efficient path in steps 84, 86, and 88, respectively, the sequence cost, the task allocation cost, and the fleet cost may be determined.

[0061] In a fleet management system and method according to the teachings of the present disclosure, multiple tools are used to manage and control the deployment of AMRs by: 1. selecting the optimal fleet composition; 2. determining the optimal task allocation within the selected fleet; 3. once the tasks have been allocated, determining the optimal sequence to visit waypoints; and 4. determining the optimal path through the waypoint sequence. Using the fleet management and method of the present disclosure, a fleet of robots can handle material delivery more efficiently at the lowest operating cost.

[0062] Unless otherwise expressly indicated herein, all numerical values ​​indicating mechanical / thermal properties, composition percentages, dimensions and / or tolerances or other characteristics when describing the scope of the present disclosure should be understood to be modified by the word "about" or "approximately". Such modifications are desirable for a variety of reasons, including: industrial practice; material, manufacturing and assembly tolerances; and testing capabilities.

[0063] As used herein, the phrase at least one of A, B, and C should be construed to mean a logical (A or B or C), using a non-exclusive logical "or", and should not be construed to mean "at least one of A, at least one of B, and at least one of C."

[0064] In this application, the terms "controller" and / or "module" may refer to, be part of, or include: an application-specific integrated circuit (ASIC); a digital, analog, or mixed analog / digital discrete circuit; a digital, analog, or mixed analog / digital integrated circuit; a combinatorial logic circuit; a field programmable gate array (FPGA); a processor circuit (shared, dedicated, or group) that executes code; a memory circuit (shared, dedicated, or group) that stores code executed by the processor circuit; other suitable hardware components that provide the described functionality (e.g., an operational amplifier circuit integrator as part of a heat flux data module); or a combination of some or all of the above, such as in a system on a chip.

[0065] The term memory is a subset of the term computer-readable medium. The term computer-readable medium as used herein does not encompass transient electrical or electromagnetic signals propagated through a medium (such as on a carrier wave); therefore, the term computer-readable medium may be considered to be tangible and non-transitory. Non-limiting examples of non-transitory tangible computer-readable media are non-volatile memory circuits (such as flash memory circuits, erasable programmable read-only memory circuits, or mask read-only circuits), volatile memory circuits (such as static random access memory circuits or dynamic random access memory circuits), magnetic storage media (such as analog or digital tape or hard drives), and optical storage media (such as CDs, DVDs, or Blu-ray discs).

[0066] The apparatus and methods described in this application may be implemented partially or completely by a special-purpose computer, which is created by configuring a general-purpose computer to perform one or more specific functions embodied in a computer program. The functional blocks, flow chart components, and other elements described above are used as software specifications, which can be translated into a computer program by routine work of a technician or programmer.

[0067] The description of the present disclosure is merely exemplary in nature, and thus, variations that do not depart from the essence of the present disclosure are intended to be within the scope of the present disclosure. Such variations should not be regarded as a departure from the spirit and scope of the present disclosure.

[0068] According to the present invention, a method for managing a fleet of robots for delivering materials in a facility includes: determining a waypoint sequence by a branch and bound (B&B) method based on movement constraints; determining a path through the waypoint sequence by a dual graph method based on a total path cost; determining task allocation among robots by the B&B method; and determining fleet composition by the B&B method.

[0069] In one aspect of the invention, the method comprises determining a waypoint sequence by considering three staggered waypoints at a time.

[0070] In one aspect of the invention, the method includes partitioning a search space for task allocation by using a plurality of processors.

[0071] In one aspect of the present invention, the fleet composition includes the number of robots and the types of robots in the fleet.

[0072] In one aspect of the invention, a robotic fleet includes a plurality of autonomous mobile robots (AMRs).

[0073] In one aspect of the invention, determining a path through the waypoint based on the dual graph further comprises selecting a path with a lowest total path cost.

[0074] In one aspect of the invention, waypoints include a pick-up location and a drop-off location where the robot picks up and drops an item, respectively.

Claims

1. A method of managing a fleet of robots for delivering materials in a facility, the method comprising: Determine the waypoint sequence by the branch and bound (B&B) method; determining a path through the sequence of waypoints by a dual graph method; Determine the task allocation between the robots by the B&B method; as well as The fleet composition is determined by the B&B method.

2. The method of claim 1 further comprising determining a small candidate fleet composition having a most efficient fleet composition by using a stagnant incremental method. 3 . The method of claim 1 , wherein said determining a path through said sequence of waypoints comprises determining a total path cost of said path. The method according to claim 3 , wherein the total path cost comprises a traversal cost and a pivot cost. The method of claim 4 , wherein the traversal cost is a cost associated with a translational movement of the robotic fleet.

6. The method of claim 4, wherein the pivot cost is a cost associated with changing the orientation of the robot and the time taken to change orientation.

7. The method of claim 1 or 3, wherein said determining a path through said sequence of waypoints comprises selecting said path having a lowest total path cost. The method of claim 1 , wherein determining a waypoint sequence comprises determining movement constraints.

9. The method of claim 8, wherein said determining the task allocation among said robots comprises partitioning a search space for task allocation by using a plurality of processors.

10. The method of claim 1, wherein the fleet composition includes the number of robots and robot types in the fleet.

11. The method of claim 1 , wherein the robotic fleet comprises a plurality of autonomous mobile robots (AMRs).

12. The method of claim 1, further comprising determining the waypoint sequence by considering three waypoints staggered at a time.

13. The method of claim 1, wherein the waypoints include a pick-up location and a drop-off location where the robot picks up and drops an item, respectively.