Dynamic scheduling method for palletizing robot

Abstract

The invention provides a palletizing robot dynamic scheduling method. The palletizing robot dynamic scheduling method comprises the steps that order information of palletizing materials and a palletizing robot list are obtained; according to the order information and the palletizing robot list, a distributed auction algorithm is adopted to determine a palletizing task, wherein the palletizing task is used for representing the palletizing robot to execute palletizing processing of corresponding palletizing materials; according to the stacking task, an A * algorithm and an energy consumption model are adopted to determine the optimal carrying strategy for the stacked materials to reach the stacking robot, and the optimal carrying strategy is used for representing the minimum ratio of the energy consumption value to the carrying path; and according to the optimal carrying strategy, distribution of the stacked materials is executed. The multi-palletizing robot has the beneficial effects that the overall energy consumption of the multi-palletizing robot is reduced, and the palletizing accuracy and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent warehousing technology, and in particular to a dynamic scheduling method for a palletizing robot. Background Art

[0002] With the rapid development of industrial automation, palletizing robots have been widely used in logistics, manufacturing and other fields. At present, traditional palletizing robot scheduling methods are mostly based on preset rules or static optimization models. These methods usually rely on fixed scheduling strategies and task allocation rules, and lack the ability to adapt to complex dynamic production environments. For example, when the task volume of the production line fluctuates greatly or the equipment status changes, the scheduling method with fixed rules is often difficult to adjust in time, resulting in uncoordinated resource allocation, reduced palletizing efficiency, increased energy consumption, and increased production costs.

[0003] In addition, when multiple palletizing robots palletize different palletizing materials, this will cause a significant increase in overall energy consumption, and the scheduling efficiency will directly affect the overall operating efficiency of the production line. Traditional scheduling methods usually rely on pre-set rules and paths, lack flexibility, and are difficult to cope with dynamically changing production environments and complex task requirements. Summary of the invention

[0004] Aiming to solve at least one of the technical problems existing in the prior art, the present invention provides a dynamic scheduling method for a palletizing robot, which reduces the overall energy consumption of multiple palletizing robots and improves the palletizing accuracy and efficiency.

[0005] One aspect of the present invention provides a dynamic scheduling method for a palletizing robot, characterized by comprising: Obtain order information of palletizing materials and list of palletizing robots; Determine the palletizing task by using a distributed auction algorithm according to the order information and the palletizing robot list, wherein the palletizing task is used to represent the palletizing process of the corresponding palletizing material performed by the palletizing robot; According to the palletizing task, the A* algorithm and the energy consumption model are used to determine the optimal handling strategy for the palletizing materials to reach the palletizing robot, wherein the optimal handling strategy is used to characterize the minimum ratio of the energy consumption value to the handling path; According to the optimal handling strategy, the distribution of the palletized materials is performed.

[0006] According to the dynamic scheduling method for palletizing robots, obtaining order information of palletizing materials and a list of palletizing robots includes: Using one of RFID, barcode and graphic code to read the label information of the incoming palletizing material to obtain the order information, wherein the order information includes the volume, weight, material type and order priority of the material; The operating power of the palletizing robot is collected through sensors, as well as the current task load and remaining palletizing space of the palletizing robot. The list of palletizing robots is determined based on the operating power, current task load and remaining palletizing space.

[0007] According to the dynamic scheduling method for a palletizing robot, the method further comprises: Multiple palletizing robots communicate using the ROS2 communication protocol, and a central database is used to dynamically maintain the list of palletizing robots based on operating power, current task load, and remaining palletizing space.

[0008] According to the dynamic scheduling method for a palletizing robot, the method further includes: The palletizing robots that have stopped working or malfunctioned are removed from the palletizing robot list through the central database, and the remaining task load of the removed palletizing robots is obtained. The remaining task load is used as the auction object and redistributed using a distributed auction algorithm.

[0009] According to the dynamic scheduling method for palletizing robots, the order information and the list of palletizing robots use a distributed auction algorithm to determine the palletizing tasks, including: According to the operating power, current task load and remaining palletizing space, a distributed auction algorithm is used to determine the bid value of each palletizing robot in the palletizing robot list, where the distributed auction algorithm is: in is the bid value, Indicates the distance between the palletizing robot and the palletizing material. is the current task load, is the operating power, is the ratio of the remaining palletizing space of the palletizing robot to the total palletizing space, , and is the weight parameter, i is the serial number of the palletizing robot; The bidding values ​​of all palletizing robots in the robot list are sorted by size to determine a candidate list of palletizing robots corresponding to a preset number of bidding values.

[0010] According to the dynamic scheduling method for a palletizing robot, the optimal transport strategy for the palletizing materials to reach the palletizing robot is determined by using the A* algorithm and the energy consumption model according to the palletizing task, including: The A* algorithm is used to determine the handling path of the palletizing robot in the candidate list; According to the transport path, the energy consumption model is used to calculate the total energy consumption value of the palletizing materials arriving at the palletizing robot. for: in, is the total number of palletizing robots, The power of the palletizing robot to move the palletized materials. is the moving time of the palletizing robot to move the palletizing materials, It is the lifting power of the palletizing robot to grab the palletized materials. The lifting height of the palletizing robot to grab the palletized materials. The time it takes for the palletized material to reach the palletizing machine. is the conveyor belt power; Calculate the ratio of the energy consumption value and the handling path of each palletizing robot in the candidate list, and take the minimum ratio as the optimal handling strategy of the palletizing robot.

[0011] According to the dynamic scheduling method for a palletizing robot, the method further comprises: The input layer of the deep reinforcement learning model uses material information, stacking status and order priority as input feature vectors, and the attention mechanism is used to process order priority and stacking status through the hidden layer, and the action space of the palletizing robot is output through the output layer; The action space of the palletizing robot is evaluated through the reward function to obtain the evaluation results. The palletizing strategy of the palletizing robot is optimized according to the evaluation results. The reward function is: in, To ensure the stacking stability of the palletizing robot, For the stacking efficiency of the palletizing robot, is the stacking failure penalty factor, , and is the weight parameter.

[0012] According to the dynamic scheduling method of the palletizing robot, the deep reinforcement learning model adopts a deep Q learning algorithm or a proximal strategy optimization algorithm, wherein the loss function of the deep Q learning algorithm is for in is the target Q value, and , is the current network parameter, is the target network parameter, is the discount factor, is the current state, is the action space, is the action space at t+1, is the reward value, is the number of time steps; Among them, the loss function of the proximal strategy optimization algorithm is for: Among them is Strategy probability ratio, is the advantage function, which is used to measure the advantages and disadvantages of the palletizing strategy of the palletizing robot. is the clipping range, function To limit the update amplitude of palletizing strategy.

[0013] The beneficial effects of the present invention are as follows: a distributed auction algorithm is adopted to allow each robot to bid for tasks autonomously, thereby reducing the burden of central control and improving the efficiency of palletizing allocation; by combining real-time data, the palletizing robot can dynamically allocate tasks according to its own status (such as operating power, operating load) and order priority, thereby optimizing resource utilization; by performing optimal strategy planning (path length and energy consumption ratio) from the conveyor belt to the palletizing robot for palletizing materials, energy consumption is reduced and efficiency is improved; dynamic optimization of the palletizing strategy is achieved through deep reinforcement learning, and the palletizing sequence and stacking rules can be flexibly adjusted according to real-time material information, order requirements and stacking status, thereby improving the accuracy and efficiency of palletizing. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a schematic flow chart of a dynamic scheduling method for a palletizing robot according to an embodiment of the present invention.

[0015] Figure 2 It is a schematic diagram of the palletizing task determination process of an embodiment of the present invention.

[0016] Figure 3 It is a schematic diagram of the optimal transport strategy generation process of an embodiment of the present invention.

[0017] Figure 4 It is a schematic diagram of the dynamic optimization process of the palletizing strategy of an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. In this subsequent description, the continuous numbering of the method steps is for the convenience of review and understanding. In combination with the overall technical solution of the present invention and the logical relationship between the various steps, adjusting the implementation order between the steps will not affect the technical effect achieved by the technical solution of the present invention. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0019] refer to Figure 1 , Figure 11 is a schematic flow chart of a dynamic scheduling method for a palletizing robot according to an embodiment of the present invention. It includes but is not limited to steps S100 to S400: S100, obtaining order information of palletizing materials and a list of palletizing robots.

[0020] In some embodiments, one of RFID, barcode and graphic code is used to read the label information of the incoming palletized material to obtain order information, wherein the order information includes the volume, weight, material type and order priority of the material.

[0021] It is understandable that in logistics warehousing, order information also includes deadlines. For example, when the courier is at the sorting center, sorting / palletizing must be completed within the agreed time. Therefore, the order information for palletized materials also changes dynamically, such as temporary expedited orders.

[0022] In some embodiments, the operating power of the palletizing robot is collected by sensors, as well as the current task load and remaining palletizing space of the palletizing robot, and a list of palletizing robots is determined based on the operating power, current task load and remaining palletizing space.

[0023] It is understandable that the operating power, current task load and remaining palletizing space are used to analyze whether the palletizing robot can receive the palletizing material. If the operating power is high, the task load is high (the processing list is full), and there is insufficient space at the palletizing stacking position, performing the receiving task again will lead to increased operating power and low efficiency. Therefore, each palletizing material must be reasonably allocated.

[0024] In some embodiments, multiple palletizing robots communicate using the ROS2 communication protocol, and a central database is used to dynamically maintain a list of palletizing robots based on operating power, current task load, and remaining palletizing space.

[0025] In some embodiments, a stopped or faulty palletizing robot is removed from a list of palletizing robots through a central database, the remaining task load of the removed palletizing robot is obtained, and the remaining task load is used as an auction object and redistributed using a distributed auction algorithm.

[0026] In some embodiments, ROS2 can also be used to broadcast and resolve conflicts in the palletizing task allocation results to avoid the palletizing robot from repeatedly accepting tasks.

[0027] S200, determining a palletizing task by using a distributed auction algorithm according to the order information and the list of palletizing robots, wherein the palletizing task is used to represent the palletizing process of the corresponding palletizing material performed by the palletizing robot.

[0028] In some embodiments, reference Figure 2The schematic diagram of the process flow for determining the palletizing task shown in FIG. 1 includes but is not limited to steps S210 to S220: S210, determining the bid value of each palletizing robot in the palletizing robot list by using a distributed auction algorithm according to the operating power, the current task load and the remaining palletizing space.

[0029] The distributed auction algorithm is: in is the bid value, Indicates the distance between the palletizing robot and the palletizing material. is the current task load, is the operating power, is the ratio of the remaining palletizing space of the palletizing robot to the total palletizing space, , and is the weight parameter, and i is the serial number of the palletizing robot.

[0030] S220, sorting the bidding values ​​of all palletizing robots in the robot list, and determining a candidate list of palletizing robots corresponding to a preset number of bidding values.

[0031] In some embodiments, the preset number of bidding values ​​refers to the first few bidding values ​​with the largest ranking. Preferably, the preset number is set to 3.

[0032] S300, according to the palletizing task, an optimal transport strategy for the palletizing material to reach the palletizing robot is determined using an A* algorithm and an energy consumption model, wherein the optimal transport strategy is used to characterize a minimum ratio of an energy consumption value to a transport path.

[0033] In some embodiments, reference Figure 3 The schematic diagram of the optimal transport strategy generation process shown in FIG. 1 includes but is not limited to steps S310 to S330: S310, using an A* algorithm to determine a transport path for the palletizing robot in the candidate list; S320, according to the transport path, using an energy consumption model to calculate the total energy consumption value of the palletizing materials reaching the palletizing robot.

[0034] The energy consumption model for: in, is the total number of palletizing robots, The power of the palletizing robot to move the palletized materials. is the moving time of the palletizing robot to move the palletizing materials, It is the lifting power of the palletizing robot to grab the palletized materials. The lifting height of the palletizing robot to grab the palletized materials. The time it takes for the palletized material to reach the palletizing machine. is the conveyor belt power.

[0035] S330, calculating the ratio of the energy consumption value to the transport path of each palletizing robot in the candidate list, and taking the minimum ratio as the optimal transport strategy of the palletizing robot.

[0036] In some embodiments, reference Figure 4 The schematic diagram of the dynamic optimization process of the palletizing strategy shown in the figure, that is, the palletizing strategy optimization of the palletizing robot is performed to improve the efficiency and accuracy of dynamic changes during palletizing, which includes but is not limited to steps S340-350: S340, using the input layer of the deep reinforcement learning model to take material information, stacking status and order priority as input feature vectors, using the attention mechanism through the hidden layer to process the order priority and stacking status, and outputting the action space of the palletizing robot through the output layer; S350, evaluating the action space of the palletizing robot by using a reward function to obtain an evaluation result, and optimizing the palletizing strategy of the palletizing robot according to the evaluation result, wherein the reward function is: in, The stacking stability of the palletizing robot, such as whether the stacking is tilted and whether it meets the center of gravity requirements; The palletizing efficiency of the palletizing robot, such as the speed of completing orders and the number of pallets per unit time; is the stacking failure penalty factor, i.e., the penalty for collision risk or stacking failure, , and is the weight parameter.

[0037] It is understandable that the above-mentioned material information, stacking status and specific values ​​of order priority can be obtained through data collection and analysis by pre-set sensors, such as setting an image detection device to determine whether the stacking robot is tilted during stacking.

[0038] In some embodiments, the deep reinforcement learning model adopts a deep Q-learning algorithm (DQN), which estimates the value of an action by calculating a Q value through a Q function; and a proximal policy optimization algorithm (PPO), which optimizes action selection through a policy gradient.

[0039] Specifically, the loss function of the deep Q learning algorithm is for in is the target Q value, and , is the current network parameter, is the target network parameter, is the discount factor, is the current state, is the action space, is the action space at t+1, is the reward value, is the number of time steps; Among them, the loss function of the proximal strategy optimization algorithm is for: Among them is Strategy probability ratio, is the advantage function, which is used to measure the advantages and disadvantages of the palletizing strategy of the palletizing robot. is the clipping range, function To limit the update amplitude of palletizing strategy.

[0040] In some embodiments, the deep reinforcement learning model training process includes: The data set collects a large amount of material information, order requirements and stacking status of the logistics center; uses the simulation tool PyBullet to build a virtual stacking environment and record the interaction data between the robot and the environment.

[0041] During the training process, material information and stacking status are obtained from the environment, and the intelligent agent selects actions (stacking position, order, etc.) according to the current strategy. After the action is executed, the environment updates the status, and the reward is calculated and returned through the reward value. During the strategy update, the policy network is updated through the DRL (deep reinforcement learning) algorithm.

[0042] Among them, the training parameters are learning rate = 0.001, discount factor = 0.99, batch size 64, and number of training rounds 5000.

[0043] The experimental data is set to 50 material types, with a weight range of 1kg-20kg and a size range of 10cm-50cm; the order demand is generated by randomly generating 100 orders, each order contains 5-20 materials; the stacking robots are 6 fixed-position robots, and the layout is a regular matrix.

[0044] After the optimization of the above embodiments, the experimental results are as follows: palletizing efficiency, the average time to dynamically complete 100 orders is 120 seconds, which is 22% less than the traditional regular palletizing method; stacking stability, the stacking inclination rate is reduced to 3%, which is 18% less than the traditional method. Dynamic adjustment capability, in the case of real-time order changes, the intelligent palletizing algorithm can re-plan task allocation to ensure that the completion time of high-priority orders is shortened by 15%.

[0045] S400, executing the distribution of the palletized materials according to the optimal handling strategy.

[0046] In some embodiments, a dynamic scheduling system for palletizing includes a control center, a detection point, a distribution point and a palletizing robot, wherein the detection point is provided with an RFID identification device or an image code (or a two-dimensional code) identification device; the detection point and the distribution point are connected through a conveyor belt, wherein the control center is respectively connected to the detection point, the distribution point and the palletizing robot, and control instructions are issued between the palletizing robots through the ROS2 communication protocol, such as adjusting the corresponding conveyor belt according to the palletizing task accepted by the palletizing robot to make it flow to the target palletizing robot, and the control center collects the palletizing status and monitors the palletizing materials through the central database.

[0047] The beneficial effects of the embodiments of the present invention are as follows: a distributed auction algorithm is adopted to allow each robot to bid for tasks autonomously, thereby reducing the burden of central control and improving the efficiency of palletizing allocation; by combining real-time data, the palletizing robot can dynamically allocate tasks according to its own status (such as operating power, operating load) and order priority, thereby optimizing resource utilization; by performing optimal strategy planning (path length and energy consumption ratio) from the conveyor belt to the palletizing robot for palletizing materials, energy consumption is reduced and efficiency is improved; dynamic optimization of the palletizing strategy is achieved through deep reinforcement learning, and the palletizing sequence and stacking rules can be flexibly adjusted according to real-time material information, order requirements and stacking status, thereby improving the accuracy and efficiency of palletizing.

[0048] An embodiment of the present invention further provides an electronic device, the electronic device comprising a processor and a memory; The memory stores a program; The processor executes the program to execute the aforementioned dynamic scheduling method for the palletizing robot; the electronic device has the function of carrying and running the software system for collaborative palletizing equipment provided by the embodiment of the present invention, for example, a personal computer, a minicomputer, a main frame, a workstation, a network or distributed computing environment, a separate or integrated computer platform, or communicates with a charged particle tool or other imaging device, etc.

[0049] An embodiment of the present invention further provides a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement the dynamic scheduling method for a palletizing robot as described above.

[0050] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part for which is described as a larger operation is performed independently.

[0051] The embodiment of the present invention also discloses a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the aforementioned dynamic scheduling method for a palletizing robot.

[0052] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features described may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions, and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional skills of the engineer. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0053] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0054] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0055] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0056] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0057] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0058] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0059] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the described embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A dynamic scheduling method for a palletizing robot, characterized in that: include: Obtain order information of palletizing materials and list of palletizing robots; Determine the palletizing task by using a distributed auction algorithm according to the order information and the palletizing robot list, wherein the palletizing task is used to represent the palletizing process of the corresponding palletizing material performed by the palletizing robot; According to the palletizing task, the A* algorithm and the energy consumption model are used to determine the optimal handling strategy for the palletizing materials to reach the palletizing robot, wherein the optimal handling strategy is used to characterize the minimum ratio of the energy consumption value to the handling path; According to the optimal handling strategy, the distribution of the palletized materials is performed.

2. The dynamic scheduling method for a palletizing robot according to claim 1, characterized in that: The method of obtaining the order information of the palletizing materials and the list of the palletizing robots includes: Using one of RFID, barcode and graphic code to read the label information of the incoming palletizing material to obtain the order information, wherein the order information includes the volume, weight, material type and order priority of the material; The operating power of the palletizing robot is collected through sensors, as well as the current task load and remaining palletizing space of the palletizing robot. The list of palletizing robots is determined based on the operating power, current task load and remaining palletizing space.

3. The dynamic scheduling method for a palletizing robot according to claim 2, characterized in that: The method further comprises: Multiple palletizing robots communicate using the ROS2 communication protocol, and a central database is used to dynamically maintain the list of palletizing robots based on operating power, current task load, and remaining palletizing space.

4. The dynamic scheduling method for a palletizing robot according to claim 3, characterized in that: The method further comprises: The palletizing robots that have stopped working or malfunctioned are removed from the palletizing robot list through the central database, and the remaining task load of the removed palletizing robots is obtained. The remaining task load is used as the auction object and redistributed using a distributed auction algorithm.

5. The dynamic scheduling method for a palletizing robot according to claim 2, characterized in that: The method of determining the palletizing task by using a distributed auction algorithm according to the order information and the palletizing robot list includes: According to the operating power, current task load and remaining palletizing space, a distributed auction algorithm is used to determine the bid value of each palletizing robot in the palletizing robot list, where the distributed auction algorithm is: in is the bid value, Indicates the distance between the palletizing robot and the palletizing material. is the current task load, is the operating power, is the ratio of the remaining palletizing space of the palletizing robot to the total palletizing space, , and is the weight parameter, i is the serial number of the palletizing robot; The bidding values ​​of all palletizing robots in the robot list are sorted by size to determine a candidate list of palletizing robots corresponding to a preset number of bidding values.

6. The dynamic scheduling method for a palletizing robot according to claim 5, characterized in that: According to the palletizing task, the optimal transport strategy for the palletizing materials to reach the palletizing robot is determined by using the A* algorithm and the energy consumption model, including: The A* algorithm is used to determine the handling path of the palletizing robot in the candidate list; According to the transport path, the energy consumption model is used to calculate the total energy consumption value of the palletizing materials arriving at the palletizing robot. for: in, is the total number of palletizing robots, The power of the palletizing robot to move the palletized materials. is the moving time of the palletizing robot to move the palletizing materials, It is the lifting power of the palletizing robot to grab the palletized materials. The lifting height of the palletizing robot to grab the palletized materials. The time it takes for the palletized material to reach the palletizing machine. is the conveyor belt power; Calculate the ratio of the energy consumption value and the handling path of each palletizing robot in the candidate list, and take the minimum ratio as the optimal handling strategy of the palletizing robot.

7. The dynamic scheduling method for a palletizing robot according to claim 6, characterized in that: The method further comprises: The input layer of the deep reinforcement learning model uses material information, stacking status and order priority as input feature vectors, and the attention mechanism is used to process order priority and stacking status through the hidden layer, and the action space of the palletizing robot is output through the output layer; The action space of the palletizing robot is evaluated through the reward function to obtain the evaluation results. The palletizing strategy of the palletizing robot is optimized according to the evaluation results. The reward function is: in, To ensure the stacking stability of the palletizing robot, For the stacking efficiency of the palletizing robot, is the stacking failure penalty factor, , and is the weight parameter.

8. The dynamic scheduling method for a palletizing robot according to claim 7, characterized in that: The deep reinforcement learning model adopts a deep Q learning algorithm or a proximal strategy optimization algorithm, where the loss function of the deep Q learning algorithm is for in is the target Q value, and , is the current network parameter, is the target network parameter, is the discount factor, is the current state, is the action space, is the action space at t+1, is the reward value, is the number of time steps; Among them, the loss function of the proximal strategy optimization algorithm is for: Among them is Strategy probability ratio, is the advantage function, which is used to measure the advantages and disadvantages of the palletizing strategy of the palletizing robot. is the clipping range, function To limit the update amplitude of palletizing strategy.

Images