Path planning method, apparatus, and system, and storage medium

Abstract

The present disclosure relates to the field of intelligent warehouse management, and provides a path planning method, apparatus, and system, and a storage medium. The path planning method comprises: constructing a route network model of a warehouse; and, on the basis of real-time environment information of the warehouse, dynamically setting a weight of each passage in the route network model of the warehouse, so as to obtain a dynamically weighted route network model, and, on the basis of requirement information of a task of an automated guided vehicle in the route network model, planning a path for the automated guided vehicle in the dynamically weighted route network model. Thus, in a warehouse environment changing in real time and under task requirements changing in real time, a good path can be found for the automated guided vehicle, so that the operating efficiency of the automated guided vehicle is improved.

Description

Path planning method, device, system and storage medium

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application is based on the application with CN application number 202311251376.5 and application date September 26, 2023, and claims its priority. The disclosed content of the CN application is hereby introduced as a whole into this application. Technical Field

[0003] The present disclosure relates to the field of intelligent warehouse management, and in particular to a path planning method, device, system, and storage medium for an automated guided vehicle. Background Art

[0004] Automated Guided Vehicles (AGVs) are automated transport devices used in warehouses, production lines, and other environments to handle and transport goods. AGVs play a crucial role in the operation of smart warehouses, offering significant advantages in handling goods within warehouses and assisting with warehouse management.

[0005] Automated guided vehicles (AGVs) need to follow a planned path and perform their tasks. However, in some related technologies, the path planning algorithms for AGVs typically only consider static warehouse environments and task requirements. While they can find good paths in these environments, they struggle to adapt to real-time, changing warehouse environments and task requirements.

[0006] Summary of the Invention

[0007] Some embodiments of the present disclosure provide a path planning method, including:

[0008] Obtain the warehouse's road network model;

[0009] Determine the demand information of the automated guided vehicle's tasks in the road network model;

[0010] Dynamically setting the weight of each channel in the road network model according to the real-time environmental information of the warehouse to obtain a dynamic weighted road network model;

[0011] A path is planned for the automatic guided vehicle in the dynamic weighted road network model according to the demand information of the automatic guided vehicle's task in the road network model.

[0012] In some embodiments, obtaining a road network model of a warehouse includes: representing channels in the warehouse as edges according to the geographical layout and channel distribution of the warehouse, representing intersections between different channels as nodes, and connecting the edges and nodes of the warehouse to obtain a road network model of the warehouse.

[0013] In some embodiments, determining the requirement information of the automated guided vehicle's task in the road network model includes: determining the starting point, end point, and waypoint of the automated guided vehicle's task in the road network model.

[0014] In some embodiments, the weights of each channel in the road network model are dynamically set according to the real-time environmental information of the warehouse to obtain a dynamic weight road network model, including: dynamically setting the weight of any channel in the road network model according to at least one of the real-time congestion information and shelf utilization information of any channel in the warehouse to obtain a dynamic weight road network model.

[0015] In some embodiments, the congestion information and shelf usage information of any channel are weighted and summed to obtain the weight of any channel.

[0016] In some embodiments, the more congested any channel is, the greater the weight of the channel is; the higher the shelf utilization rate of any channel is, the smaller the weight of the channel is.

[0017] In some embodiments, planning a path for the automatic guided vehicle in the dynamic weighted road network model includes: using a heuristic algorithm, a reinforcement learning algorithm, or a genetic algorithm to plan a path for the automatic guided vehicle in the dynamic weighted road network model based on the demand information of the automatic guided vehicle's task in the road network model.

[0018] In some embodiments, the heuristic algorithm comprises an A* algorithm.

[0019] In some embodiments, the reinforcement learning algorithm comprises a Q-learning algorithm or a deep Q-network algorithm.

[0020] In some embodiments, the genetic algorithm comprises a non-dominated sorting genetic algorithm.

[0021] In some embodiments, planning a path for the automatic guided vehicle in the dynamic weighted road network model further includes: adjusting the path using a local search algorithm.

[0022] In some embodiments, using a local search algorithm to adjust the path includes: using a local search algorithm to adjust adjacent nodes in the path, comparing the paths before and after the adjustment, and if the adjusted path has a smaller cost than the path before the adjustment, selecting the adjusted path.

[0023] In some embodiments, the local search algorithm comprises a 2-Opt algorithm.

[0024] Some embodiments of the present disclosure provide a path planning device, comprising: a memory; and a processor coupled to the memory, wherein the processor is configured to execute a path planning method based on instructions stored in the memory.

[0025] Some embodiments of the present disclosure provide a path planning device, including: a module for executing a path planning method.

[0026] Some embodiments of the present disclosure provide a path planning system, including:

[0027] A path planning device configured to execute a path planning method; and

[0028] The automatic guided vehicle is configured to run and perform tasks according to the path planned by the path planning device.

[0029] In some embodiments, it also includes: a monitoring device configured to obtain real-time environmental information of the warehouse and send it to the path planning device.

[0030] Some embodiments of the present disclosure provide a non-transitory computer-readable storage medium having a computer program stored thereon, which implements the steps of a path planning method when executed by a processor.

[0031] Some embodiments of the present disclosure provide a computer program, comprising: instructions, which, when executed by a processor, cause the processor to execute a path planning method. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The following briefly introduces the drawings required for describing the embodiments or related technologies. The present disclosure can be more clearly understood based on the following detailed description with reference to the drawings.

[0033] Obviously, the drawings described below are only some embodiments of the present disclosure. A person skilled in the art can derive other drawings based on these drawings without inventive effort.

[0034] FIG1 is a flow chart showing a path planning method according to some embodiments of the present disclosure.

[0035] FIG2 shows a schematic diagram of a path planning system according to some embodiments of the present disclosure.

[0036] FIG3 shows a schematic diagram of a path planning device according to some embodiments of the present disclosure.

[0037] FIG4 shows a schematic diagram of a path planning device according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present disclosure.

[0039] In this disclosed embodiment, the weights of each channel in the warehouse's road network model are dynamically set based on the warehouse's real-time environmental information to create a dynamic weighted road network model. Based on the requirements of the AGV's mission within this model, a path is planned for the AGV within this dynamic weighted road network model. This allows the AGV to find optimal paths even under constantly changing warehouse environments and mission requirements, improving its operational efficiency.

[0040] FIG1 is a flow chart showing a path planning method according to some embodiments of the present disclosure.

[0041] As shown in FIG1 , the path planning method of this embodiment may be executed by a path planning device, for example. The method includes the following steps.

[0042] In step 110 , a road network model of the warehouse is obtained.

[0043] In some embodiments, a pre-built warehouse road network model can be obtained, or a warehouse road network model can be constructed to obtain a warehouse road network model. Constructing a warehouse road network model includes: according to the geographical layout and channel distribution of the warehouse, representing the channels in the warehouse as edges, representing the intersections between different channels as nodes, and also representing key locations in the warehouse (such as shelves, workstations, etc.) as nodes, connecting the various edges and nodes of the warehouse to construct a warehouse road network model. In the road network model, a node represents a specific location in the warehouse, such as but not limited to an intersection, a shelf location, a turning point, a starting point or an end point. Each node has an associated navigation marker or ground code for navigation and positioning of the automatic guided vehicle.

[0044] In step 120 , requirement information of the AGV task in the road network model is determined.

[0045] In some embodiments, determining the requirement information of the automated guided vehicle's task in the road network model includes determining the starting point, end point, and waypoint of the automated guided vehicle's task in the road network model to facilitate subsequent path planning.

[0046] In some embodiments, the task of the automated guided vehicle can be an initial task or a task dynamically adjusted based on real-time information. For the initial task and each dynamically adjusted task, steps 120-140 can be executed / re-executed to dynamically plan a path for the automated guided vehicle.

[0047] In step 130, the weights of the various channels in the road network model are dynamically set according to the real-time environmental information of the warehouse to obtain a dynamic weight road network model.

[0048] Real-time warehouse environment information is acquired through monitoring equipment such as cameras and sensors and sent to the path planning device. Real-time warehouse environment information includes, but is not limited to, aisle congestion information and shelf utilization information.

[0049] The following describes congestion information. Warehouse aisles can become congested for various reasons (such as employees walking, other vehicles, etc.). If the automated guided vehicle still uses a statically planned path, it may lead to low work efficiency and even safety accidents such as collisions. By monitoring the congestion information of the channels in real time and dynamically adjusting the weights, the automated guided vehicle can choose a smoother and safer path. When the congestion information of a certain channel indicates that the channel is very busy, the weight of this channel should be increased, so that the automated guided vehicle is more likely to choose other channels during path planning. The situation of each channel can be monitored in real time by cameras installed in the warehouse. For example, computer vision technology can be used to detect and count moving objects (such as people, vehicles, etc.) in the channel to determine the congestion level of the channel.

[0050] The following describes shelf utilization. Shelf utilization may affect the tasks of automated guided vehicles (AGVs). For example, a high-utilization shelf area means that there are more inbound or outbound tasks in that area, and AGVs need to travel to that area frequently. Therefore, shelf utilization can be used to predict the future workload of certain areas and adjust path planning accordingly. When the utilization rate of a particular shelf is very high, the weight of the channels near that shelf may be reduced to make it easier for AGVs to travel to this area. Shelf utilization can be obtained using various methods. One method is to use RFID (Radio Frequency Identification) technology. Each item is attached with an RFID tag. The RFID reader on the shelf can read the number of items on the shelf in real time, thereby determining the shelf utilization rate. In addition, cameras can also be used to determine or assist in determining shelf utilization. For example, computer vision technology can be used to analyze the number of items on the shelf to determine shelf utilization.

[0051] According to the aforementioned weight setting method, under the same conditions, the greater the weight of a channel, the less likely the corresponding channel will be selected.

[0052] In some embodiments, the weight of any channel in the road network model is dynamically set based on at least one of real-time congestion information and shelf utilization information for any channel in the warehouse to obtain a dynamic weighted road network model. In other words, a road network model in which channels (or "edges") have weights and the weights change dynamically is called a dynamic weighted road network model.

[0053] The weight of any channel is obtained by weighting and summing the congestion information and shelf utilization information of any channel. The calculation formula of the channel weight can be expressed as: W_ij = α × C_ij + β × S_ij

[0054] Among them, Wij represents the weight of the channel from node i to j; Cij represents the congestion level of the channel from node i to j; Sij represents the shelf utilization rate of the channel from node i to j; α and β are weight coefficients, which can be adjusted according to actual conditions.

[0055] In step 140 , a path is planned for the automated guided vehicle in the dynamic weighted road network model based on the demand information of the automated guided vehicle's task in the road network model.

[0056] Step 140 includes step 140-1 and, depending on planning requirements, may further include step 140-2. The path planned in step 140-1 is also called the initial path, and the path adjusted in step 140-2 is the final path to be output.

[0057] In step 140-1, based on the demand information of the automated guided vehicle's task in the road network model, a heuristic algorithm, a reinforcement learning algorithm, or a genetic algorithm is used to plan a path for the automated guided vehicle in the dynamic weighted road network model. The heuristic algorithm includes, but is not limited to, the A* algorithm. The reinforcement learning algorithm includes, but is not limited to, the Q-learning algorithm or the Deep Q-Network (DQN) algorithm. The genetic algorithm includes, but is not limited to, the Non-Dominated Sorting Genetic Algorithm (NSGA-II) algorithm. The Non-Dominated Sorting Genetic Algorithm is a multi-objective optimization algorithm that uses methods such as non-dominated sorting and congestion distance to find the Pareto optimal solution for multi-objective problems. Q-learning algorithm: A reinforcement learning algorithm that learns to take different actions in different states to obtain the maximum cumulative reward by interacting with the environment. Deep Q-Network algorithm: A reinforcement learning algorithm that combines deep learning and Q-learning, uses a neural network as a function approximator to represent the Q value, and is suitable for processing problems in high-dimensional, continuous state spaces.

[0058] The following describes a method for planning a path using a heuristic algorithm. The heuristic algorithm uses heuristic information (such as distance, time, etc.) to guide the search direction during the search process to reduce the search space and computational complexity.

[0059] (a) Define a heuristic function: Choose or design a heuristic function, such as Euclidean distance, to estimate the cost from the current node to the target node. This function takes into account the impact of dynamic weights.

[0060] (b) Initialization: Select a starting node, put it into the open list, and set the initial cost to 0.

[0061] (c) Iteration: Find the node with the lowest cost in the open list, expand this node, and calculate the costs to its neighboring nodes, including the effects of dynamic weights. Add the neighboring node to the open list and update the cost and parent node information. Repeat this step until the target node is found or the open list is empty.

[0062] (d) Path reconstruction: trace back from the target node to the start node and reconstruct the path along the parent node pointer.

[0063] By combining weights with the A* algorithm, it is possible to respond to dynamic environments, such as dynamically adjusting paths as the congestion level of aisles and the utilization rate of shelves change.

[0064] The following describes a method for planning a path using a reinforcement learning algorithm.

[0065] (a) Define the environment and state space: In the context of AGV path planning, the state space can include the AGV's location, destination, warehouse layout, and other dynamic factors (e.g., obstacles). Weights can be part of the state space, representing the priority or cost of various paths or directions.

[0066] (b) Select a suitable reinforcement learning algorithm, such as Q-learning or Deep Q-Network, which can be used to train the AGV to choose the optimal path in a given environment.

[0067] (c) Training and real-time updating: By interacting with the environment, the AGV learns which path is optimal for different situations. Weights can be used to adjust the learning process to reflect the importance or priority of different paths. Real-time updating allows the AGV to adapt to changes in the environment.

[0068] (d) Applying the trained reinforcement learning method to the path planning system.

[0069] Reinforcement learning methods can adaptively learn optimal paths under different circumstances and task requirements through interaction with the environment. In the ever-changing warehouse environment and task scheduling scenarios, reinforcement learning methods can update path planning strategies in real time, improving the accuracy and practicality of path planning.

[0070] The following describes a method for planning a path using a genetic algorithm.

[0071] (a) Define optimization objectives and constraints: Identify and quantify multiple optimization objectives, such as minimizing aisle congestion and shelf utilization. Weights can be introduced to balance different objectives. Constraints are set based on the warehouse's geographic layout and aisle distribution, as well as the deployment information of the AGVs.

[0072] (b) Select a multi-objective optimization algorithm: For example, use the non-dominated sorting genetic algorithm (NSGA-II) to solve this multi-objective problem.

[0073] (c) Set and adjust algorithm parameters: Set and adjust algorithm parameters such as population size, crossover and mutation rate according to the specific problem.

[0074] (d) Optimize and select a solution: Run the multi-objective optimization algorithm and select a set of near-optimal solutions. Various techniques, such as the Pareto optimal front, can be used to select and present the optimal solution. The near-optimal solution is also the planned path.

[0075] The automatic guided vehicle path planning algorithm based on multi-objective optimization incorporates real-time changing warehouse environment information (such as channel congestion level and shelf utilization rate) into the optimization objectives to find a trade-off solution between multiple objectives and find a set of approximate optimal solutions under limited computing resources, thereby improving the efficiency and adaptability of path planning.

[0076] In step 140 - 2 , a local search algorithm is used to adjust the planned path.

[0077] Local search optimization: Based on the initial path, a local search method is used to optimize the initial path. This method attempts to find a more optimal path by making local adjustments to adjacent nodes in the initial path. Examples of local search algorithms include, but are not limited to, the 2-Opt algorithm. The 2-Opt algorithm is a local search algorithm that optimizes a path by swapping two adjacent nodes. The 2-Opt algorithm can find a near-optimal solution using limited computing resources.

[0078] In a path, any two consecutive nodes are considered adjacent nodes. In other words, adjacent nodes are continuous in the path sequence. For example, if the path consists of nodes 1, 2, 3, 4, and 5, then nodes 1 / 2, nodes 2 / 3, nodes 3 / 4, nodes 4 / 5, and so on are all adjacent nodes.

[0079] In some embodiments, a local search algorithm is used to adjust the adjacent nodes in the path (i.e., the planned initial path), and the paths before and after the adjustment are compared. If the adjusted path has a lower cost than the path before the adjustment, the adjusted path is selected. The adjustment and selection process is repeated until a preset number of iterations is reached or other iteration stop conditions are met. The adjustment of adjacent nodes, for example, includes changing the connection method between these nodes or replacing some of the nodes. For example, the initial path consists of nodes 1, 2, 3, 4, and 5 in sequence. By adjusting the order of nodes 3 and 4, a path with a lower overall cost is obtained, which consists of nodes 1, 2, 4, 3, and 5 in sequence.

[0080] Local adjustments can be made by changing the connection between these nodes or replacing some of the nodes to find a better path, thereby fine-tuning the existing path.

[0081] In this disclosed embodiment, the weights of each channel in the warehouse's road network model are dynamically set based on the warehouse's real-time environmental information to create a dynamic weighted road network model. Based on the requirements of the AGV's mission within this model, a path is planned for the AGV within this dynamic weighted road network model. This allows the AGV to find optimal paths even under constantly changing warehouse environments and mission requirements, improving its operational efficiency.

[0082] FIG2 shows a schematic diagram of a path planning system according to some embodiments of the present disclosure.

[0083] As shown in FIG2 , the path planning system of this embodiment includes the following devices.

[0084] The path planning device 210 is configured to execute the path planning method in each embodiment.

[0085] The automatic guided vehicle 220 is configured to run and perform tasks according to the path planned by the path planning device 210.

[0086] The monitoring device 230 is configured to obtain real-time environment information of the warehouse and send it to the path planning device 210 so that the path planning device 210 can plan a path for the automatic guided vehicle 220 based on the real-time environment information of the warehouse.

[0087] FIG3 shows a schematic diagram of a path planning device according to some embodiments of the present disclosure.

[0088] As shown in FIG3 , the path planning device 300 of this embodiment includes: a memory 310 and a processor 320 coupled to the memory 310 , and the processor 320 is configured to execute the path planning method in any of the aforementioned embodiments based on instructions stored in the memory 310 .

[0089] The path planning device 300 may further include an input / output interface 330 , a network interface 340 , a storage interface 350 , etc. These interfaces 330 , 340 , 350 , the memory 310 , and the processor 320 may be connected via a bus 360 , for example.

[0090] The memory 310 may include, for example, a system memory, a fixed non-volatile storage medium, etc. The system memory may store, for example, an operating system, application programs, a boot loader, and other programs.

[0091] The processor 320 may be implemented using a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, or discrete hardware components such as discrete gates or transistors.

[0092] Among them, the input and output interface 330 provides a connection interface for input and output devices such as a display, mouse, keyboard, and touch screen. The network interface 340 provides a connection interface for various networked devices. The storage interface 350 provides a connection interface for external storage devices such as SD cards and USB flash drives. The bus 360 can use any of a variety of bus structures. For example, bus structures include but are not limited to the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MCA) bus, and the Peripheral Component Interconnect (PCI) bus.

[0093] FIG4 shows a schematic diagram of a path planning device according to some embodiments of the present disclosure.

[0094] As shown in FIG4 , the path planning device 400 of this embodiment includes a module for executing a path planning method.

[0095] An acquisition module 410 is configured to acquire a road network model of a warehouse;

[0096] A determination module 420 is configured to determine demand information of the automatic guided vehicle task in the road network model;

[0097] A setting module 430 is configured to dynamically set the weight of each channel in the road network model according to real-time environmental information of the warehouse to obtain a dynamic weight road network model;

[0098] The planning module 440 is configured to plan a path for the automated guided vehicle in the dynamic weighted road network model according to the demand information of the automated guided vehicle's task in the road network model.

[0099] The acquisition module 410 is configured to represent the channels in the warehouse as edges and the intersection points between different channels as nodes according to the geographical layout and channel distribution of the warehouse, and connect the edges and nodes of the warehouse to obtain the road network model of the warehouse.

[0100] The determination module 420 is configured to determine the starting point, end point, and waypoint of the task of the automated guided vehicle in the road network model.

[0101] The setting module 430 is configured to dynamically set the weight of any channel in the road network model based on at least one of the real-time congestion information of any channel and the shelf utilization rate information in the warehouse to obtain a dynamic weight road network model.

[0102] The setting module 430 is configured to perform weighted summation on the congestion information and shelf usage rate information of any channel to obtain the weight of any channel.

[0103] The planning module 440 is configured to plan a path for the automatic guided vehicle in the dynamic weighted road network model using a heuristic algorithm, a reinforcement learning algorithm, or a genetic algorithm based on the demand information of the automatic guided vehicle's task in the road network model.

[0104] Wherein: the heuristic algorithm includes the A* algorithm; or the reinforcement learning algorithm includes the Q learning algorithm or the deep Q network algorithm; or the genetic algorithm includes the non-dominated sorting genetic algorithm.

[0105] The planning module 440 is configured to adjust the path using a local search algorithm.

[0106] The planning module 440 is configured to use a local search algorithm to adjust adjacent nodes in the path, compare the paths before and after the adjustment, and select the adjusted path if the adjusted path has a lower cost than the path before the adjustment.

[0107] Wherein: the local search algorithm includes the 2-Opt algorithm.

[0108] (1) A path planning method comprising:

[0109] Obtain the warehouse's road network model;

[0110] Determine the demand information of the automated guided vehicle's tasks in the road network model;

[0111] Dynamically setting the weight of each channel in the road network model according to the real-time environmental information of the warehouse to obtain a dynamic weighted road network model;

[0112] A path is planned for the automatic guided vehicle in the dynamic weighted road network model according to the demand information of the automatic guided vehicle's task in the road network model.

[0113] (2) According to (1), obtaining the warehouse road network model includes: according to the geographical layout and channel distribution of the warehouse, representing the channels in the warehouse as edges, representing the intersection points between different channels as nodes, and connecting the edges and nodes of the warehouse to obtain the warehouse road network model.

[0114] (3) According to any one of (1-2), determining the demand information of the automatic guided vehicle's task in the road network model includes: determining the starting point, end point and waypoint of the automatic guided vehicle's task in the road network model.

[0115] (4) According to any one of items (1-3), the weight of each channel in the road network model is dynamically set according to the real-time environmental information of the warehouse to obtain a dynamic weight road network model, including: according to at least one of the real-time congestion information of any channel in the warehouse and the shelf utilization rate information, the weight of any channel in the road network model is dynamically set to obtain a dynamic weight road network model.

[0116] (5) According to (4), the congestion information and shelf utilization rate information of any channel are weighted and summed to obtain the weight of any channel.

[0117] (6) According to any one of (1-5), planning a path for the automatic guided vehicle in the dynamic weighted road network model includes: using a heuristic algorithm, a reinforcement learning algorithm, or a genetic algorithm to plan a path for the automatic guided vehicle in the dynamic weighted road network model based on the demand information of the automatic guided vehicle's task in the road network model.

[0118] (7) According to (6), wherein: the heuristic algorithm includes the A* algorithm; or, the reinforcement learning algorithm includes the Q learning algorithm or the deep Q network algorithm; or, the genetic algorithm includes the non-dominated sorting genetic algorithm.

[0119] (8) According to any one of (1-7), the path is adjusted using a local search algorithm.

[0120] (9) According to any one of items (1-8), using a local search algorithm to adjust the path includes: using a local search algorithm to adjust adjacent nodes in the path, comparing the paths before and after the adjustment, and if the adjusted path has a smaller cost than the path before the adjustment, selecting the adjusted path.

[0121] (10) According to any one of (8-9), the local search algorithm includes a 2-Opt algorithm.

[0122] Unless specifically stated otherwise, the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure.

[0123] It should be understood that in the embodiments of the present disclosure, “plurality” may refer to two or more than two, and “at least one / item” may refer to one, two or more than two.

[0124] It should also be understood that any component, data or structure mentioned in the embodiments of the present disclosure can generally be understood as one or more, unless explicitly limited or otherwise indicated in the context.

[0125] In addition, the term "and / or" in this disclosure is merely a description of the association relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this disclosure generally indicates that the related objects are in an "or" relationship.

[0126] It should also be understood that the description of the various embodiments in this disclosure focuses on the differences between the various embodiments, and the same or similar aspects thereof can be referenced with each other. For the sake of brevity, they will not be described one by one.

[0127] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0128] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the present disclosure, its application, or uses.

[0129] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0130] It should be noted that like reference numerals and letters refer to like items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0131] In addition, in order to avoid obscuring the present disclosure with unnecessary details, only the processing steps and / or device structures that are closely related to at least the solution according to the present disclosure are shown in the drawings, while other details that are not closely related to the present disclosure are omitted. It should also be noted that similar reference numerals and letters in the drawings indicate similar items, and therefore once an item is defined in one drawing, it does not need to be discussed again for subsequent drawings.

[0132] The embodiments of the present disclosure can be applied to electronic devices such as terminal devices, computer systems, and servers, and can operate in conjunction with numerous other general-purpose or specialized computing system environments or configurations. Examples of well-known terminal devices, computing systems, environments, and / or configurations suitable for use with terminal devices, computer systems, servers, and other electronic devices include, but are not limited to, personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, minicomputer systems, mainframe computer systems, and distributed cloud computing technology environments including any of the above systems, among others.

[0133] Electronic devices such as terminal devices, computer systems, and servers can be described in the general context of computer system-executable instructions (such as program modules) executed by a computer system. Generally, program modules can include routines, programs, object programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types. Computer systems / servers can be implemented in a distributed cloud computing environment, where tasks are performed by remote processing devices linked via a communication network. In a distributed cloud computing environment, program modules can be located on local or remote computing system storage media, including storage devices.

[0134] Those skilled in the art will appreciate that embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present disclosure may take the form of a computer program product implemented on one or more non-transitory computer-readable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer program code.

[0135] The present disclosure is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0136] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0138] The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A path planning method, comprising: Get the road network model of the warehouse; Determine the demand information of the task of the automated guided vehicle in the road network model; According to the real-time environment information of the warehouse, the weight of each channel in the road network model is dynamically set to obtain a dynamic weight road network model; According to the demand information of the task of the automatic guided vehicle in the road network model, a path is planned for the automatic guided vehicle in the dynamic weighted road network model.

2. The path planning method according to claim 1, wherein: Obtaining the warehouse's road network model includes: According to the geographical layout and channel distribution of the warehouse, the channels in the warehouse are represented as edges, the intersections between different channels are represented as nodes, and the edges and nodes of the warehouse are connected to obtain the road network model of the warehouse.

3. The path planning method according to claim 1, wherein: Determine the required information of the automatic guided vehicle task in the road network model including: The starting point, the end point and the waypoint of the task of the automated guided vehicle in the road network model are determined.

4. The path planning method according to claim 1, wherein: According to the real-time environment information of the warehouse, the weight of each channel in the road network model is dynamically set to obtain a dynamic weight road network model including: According to at least one of the real-time congestion information of any channel in the warehouse and the shelf utilization rate information, the weight of any channel in the road network model is dynamically set to obtain a dynamic weight road network model.

5. The path planning method according to claim 4, wherein: The congestion information and shelf utilization rate information of any channel are weighted and summed to obtain the weight of any channel.

6. The path planning method according to any one of claims 1 to 5, wherein: Planning a path for the automatic guided vehicle in the dynamic weighted road network model includes: According to the demand information of the task of the automatic guided vehicle in the road network model, a heuristic algorithm, a reinforcement learning algorithm, or a genetic algorithm is used to plan a path for the automatic guided vehicle in the dynamic weighted road network model.

7. The path planning method according to claim 6, wherein: The heuristic algorithm includes an A* algorithm; or, The reinforcement learning algorithm includes a Q learning algorithm or a deep Q network algorithm; or, The genetic algorithm comprises a non-dominated sorting genetic algorithm.

8. The path planning method according to claim 6, wherein: Planning a path for the automatic guided vehicle in the dynamic weighted road network model further includes: using a local search algorithm to adjust the path.

9. The path planning method according to claim 8, wherein: Adopting a local search algorithm, adjusting the path includes: Adopting a local search algorithm, the adjacent nodes in the path are adjusted, and the paths before and after the adjustment are compared. If the adjusted path has a smaller cost than the path before the adjustment, the adjusted path is selected.

10. The path planning method according to claim 9, wherein: The local search algorithm includes a 2-Opt algorithm.

11. The path planning method according to claim 4 or 5, wherein: The more congested any channel is, the greater the weight of the channel is; the higher the shelf utilization rate of any channel is, the smaller the weight of the channel is.

12. A path planning device, comprising: Memory; as well as A processor coupled to the memory, wherein the processor is configured to execute the path planning method according to any one of claims 1 to 11 based on instructions stored in the memory.

13. A path planning device, comprising: A module for executing the path planning method according to any one of claims 1 to 11.

14. A path planning system, comprising: A path planning device, configured to execute the path planning method according to any one of claims 1 to 11; as well as The automatic guided vehicle is configured to run and perform tasks according to the path planned by the path planning device.

15. The system of claim 14, further comprising: The monitoring device is configured to obtain real-time environmental information of the warehouse and send it to the path planning device.

16. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the steps of the path planning method according to any one of claims 1 to 11 are implemented.

17. A computer program comprising: Instructions, when executed by a processor, cause the processor to execute the path planning method according to any one of claims 1-11.

Images