A multi-robot dynamic task allocation planning method and related device

Abstract

The application provides a multi-robot dynamic task allocation planning method and related device, and relates to the technical field of robot control. The application carries out robot work zoning on all task points in the target cooperative work scene by taking the actual spatial position and actual residual energy of all idle robots in the target cooperative work scene at the current time as the basis, so that the actual residual energy of any idle robot in the robot zoning result of a single task point meets the task execution energy consumption requirement of the robot near the task point, to realize the robot adaptive dynamic zoning effect considering the global robot real-time state information in the dynamic and variable scene. Then, the application allocates the to-be-executed task associated with each task point to the multiple robots in the robot zoning result of the task point, to realize the efficient and reasonable multi-robot dynamic task coordination allocation effect in the dynamic and variable scene, and avoid the task allocation conflict phenomenon.

Description

Technical Field

[0001] This application relates to the field of robot control technology, and more specifically, to a method and apparatus for dynamic task allocation and planning for multiple robots. Background Technology

[0002] With the continuous development of science and technology, the application of robotics technology in various fields is becoming more and more widespread. Typically, it is necessary to deploy multiple mobile robots of various types to work collaboratively in designated narrow work scenarios such as automated production lines in factories, large logistics and warehousing centers, and smart venues (for example, deploying multiple unmanned transport vehicles to different locations to perform transport tasks, deploying multiple robot-arm composite robots to different locations to perform loading and unloading tasks, deploying multiple drones to perform dynamic inspection tasks in different areas, etc.) in order to achieve the intelligent collaborative effect of multiple robots in the corresponding designated narrow work scenarios.

[0003] For multi-robot collaborative operation scenarios, the core technical challenge lies in how to quickly and reasonably allocate various tasks to be executed by each robot in the corresponding scenario. It is necessary to solve for the globally optimal task allocation action or the suboptimal task allocation action with high efficiency under the comprehensive consideration of multiple capability constraints of each robot in the corresponding scenario (such as remaining energy constraints, robot maximum movement distance constraints, operation space domain constraints, operation time limits, etc.) in order to achieve the effect of reasonable task allocation planning in dynamic and ever-changing scenarios.

[0004] However, it is worth noting that existing task allocation planning schemes for multi-robot collaborative operation scenarios mainly adopt centralized or distributed task allocation mechanisms. Centralized task allocation mechanisms require a static scenario (including static robot positions, static task composition, static robot composition, and static scenario boundary range). When facing dynamic and changing scenarios (involving task additions, deletions, and changes in robot composition), frequent task allocation replanning for all robots within the scenario is necessary, resulting in excessive waste of computational resources and low task allocation efficiency. Therefore, it is not practically suitable for solving the dynamic task allocation problem of multi-robots in dynamic and changing scenarios. While distributed task allocation mechanisms can be implemented through local interactions and independent decision-making among multiple robots, this mechanism is prone to task allocation conflicts due to a lack of global information exchange. Furthermore, this mechanism is suitable for coordinating task allocation among multiple homogeneous robots with the same task type (e.g., multiple unmanned transport vehicles performing transport tasks), but it is difficult to achieve dynamic task coordination and allocation between multiple heterogeneous robots with different task types in dynamic and changing scenarios. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a multi-robot dynamic task allocation planning method, robot job scheduling device and readable storage medium, which can comprehensively consider the real-time status information of the global robots in a specified dynamic collaborative operation scenario, adaptively and dynamically partition multiple heterogeneous robots with different task types in the specified dynamic collaborative operation scenario, and allocate the tasks to be executed associated with the task point to multiple robots divided into the same task point partition, so as to achieve efficient and reasonable multi-robot dynamic task coordination and allocation effect in dynamic and changing scenarios, and avoid task allocation conflict.

[0006] To achieve the above objectives, the technical solutions adopted in the embodiments of this application are as follows:

[0007] In a first aspect, this application provides a multi-robot dynamic task allocation planning method, the method comprising:

[0008] Obtain the actual spatial position and actual remaining energy of each idle robot in the target collaborative operation scenario at the current moment;

[0009] The robot operation partitioning model adapted to the target collaborative operation scenario is invoked. Based on the actual spatial position and actual remaining energy of each of the idle robots, robot operation partitioning is performed on all task points in the target collaborative operation scenario to obtain the robot partitioning results of all task points at the current time. The actual remaining energy of any idle robot in the robot partitioning result of a single task point meets its task execution energy consumption requirements near the corresponding task point.

[0010] For each task location, all tasks to be executed at that task location in the target collaborative operation scenario are assigned to the respective idle robots in the corresponding robot partitioning results.

[0011] In an optional implementation, the method further includes:

[0012] The actual spatial position and actual remaining energy of each mobile robot in the target collaborative operation scenario are obtained at preset time intervals.

[0013] For each mobile robot, based on the task execution performance parameters of the mobile robot, calculate the overall energy consumption of the mobile robot from the corresponding real space location to different task points, and determine the theoretical partition label of the mobile robot at different task points based on the energy support status of the mobile robot's actual remaining energy for the overall energy consumption of task execution at different task points.

[0014] For each task location, a robot partition sample set is constructed based on the actual spatial position and actual remaining energy of all mobile robots, as well as the theoretical partition label of each mobile robot at that task location. The robot partition sample set includes robot partition samples of each mobile robot at that task location. Each robot partition sample includes a sample input feature vector composed of the actual spatial position and actual remaining energy of the corresponding mobile robot, and the theoretical partition label of the corresponding mobile robot at that task location.

[0015] A multi-class support vector machine model is trained based on the robot partition sample set for each of the task locations, and the trained multi-class support vector machine model is used as the robot task partition model.

[0016] In an optional implementation, for each mobile robot, the step of determining the theoretical partition label of the mobile robot at different task locations based on the energy support status of the overall energy consumption of task execution at different task locations according to the actual remaining energy of the mobile robot includes:

[0017] The test was conducted to determine whether the overall energy consumption of the mobile robot at different task locations was less than the robot's actual remaining energy.

[0018] When the overall energy consumption of task execution at at least one first task point is not less than the actual remaining energy, the theoretical partition label of the mobile robot at the at least one first task point is directly assigned a label value using the first preset label value; wherein the first preset label value is used to characterize that the remaining energy of the mobile robot does not meet its task execution energy consumption requirements near the corresponding task point.

[0019] When the overall energy consumption of task execution at at least one second task point is less than the actual remaining energy, a second preset label value is used to assign a theoretical partition label to the mobile robot at the second task point with the lowest overall energy consumption of task execution, and a first preset label value is used to assign a theoretical partition label to the mobile robot at other second task points; wherein the second preset label value is used to characterize that the remaining energy of the mobile robot meets its task execution energy consumption requirements near the corresponding task point.

[0020] In an optional implementation, the robot task partitioning model includes a multi-classifier model corresponding to all task locations simultaneously. The step of calling the robot task partitioning model adapted to the target collaborative task scenario, and performing robot task partitioning processing on all task locations in the target collaborative task scenario based on the actual spatial positions and actual remaining energy of each of the idle robots, to obtain the robot partitioning results for all task locations at the current time, includes:

[0021] For each idle robot, the actual spatial position and actual remaining energy of the idle robot are vector synthesized to obtain the actual input feature vector of the idle robot at the robot operation partition model;

[0022] The multi-classifier model included in the robot task partitioning model is invoked to predict the partition labels for all task locations based on the actual input feature vector, thereby obtaining the actual partition label values ​​of the idle robot at each task location;

[0023] The idle robot is added to the robot partitioning result of the target task point whose actual partition label value is the second preset label value.

[0024] In an optional implementation, for each task location, the step of assigning all unexecuted tasks related to that task location in the target collaborative operation scenario to the respective idle robots in the corresponding robot partitioning result includes:

[0025] Based on the actual task type of each of the tasks to be executed involved in the task location, the tasks to be executed are grouped to obtain at least one task group, wherein the actual task type of each task group is consistent.

[0026] Based on the pre-assigned task type of all idle robots in the robot partitioning result of the task location, all idle robots in the robot partitioning result are grouped to obtain at least one robot group, wherein the pre-assigned task type of each idle robot in each robot group is consistent.

[0027] For each task group, all tasks to be executed within that task group are individually assigned to an idle robot in the target robot group that matches the task type of that task group.

[0028] In an optional implementation, the method further includes:

[0029] For each task point, based on the task execution positions of all tasks to be executed involved in that task point, and the actual spatial positions of all target robots assigned to the tasks to be executed at that task point, with the aim of maximizing the global path cumulative utility function of that task point, collaborative planning of the movement trajectory of all target robots is performed to obtain the robot movement trajectory of each target robot when the assigned task execution constraints are met.

[0030] All target robots at the task location are controlled to move to the task execution position of the assigned task according to their corresponding robot movement trajectories, so as to drive all target robots to execute the assigned task.

[0031] In an optional implementation, the first target collaborative operation scenario The global path cumulative utility function for each task point is represented by the following equation:

[0032] ;

[0033] in, Used to indicate the first Global path cumulative utility function for each task location Used to indicate the first The cumulative length of the robot's straight path at each task point. Used to indicate the first Total number of target robots at each task location. Used to indicate the first The conversion coefficient between the robot's energy consumption and the robot's straight-line travel distance at each task location. Used to indicate the first The first task point The energy consumption of a target robot moving from its current actual spatial location to its assigned task execution location. Used to indicate the first The first task point The time-related aspects of the robot's trajectory planning process for a target robot. The function of change of acceleration during movement, Used to indicate the first The first task point The target robot at the current moment The actual spatial location, Used to indicate the first The task execution location of the assigned task for each target robot. Used to indicate the first The first task point The estimated time when the target robot arrives at the task execution position assigned to it. Used to represent the L2 norm.

[0034] In an optional implementation, when the target collaborative operation scenario is... The first task point When the assigned task of a target robot is a fixed-point static task, the execution constraints of the assigned task of the target robot are expressed by the following equation:

[0035] ;

[0036] And when the target collaborative operation scenario is the first The first task point When the assigned task of a target robot is a dynamic cruise task, the execution constraints of the assigned task of the target robot are expressed by the following equation:

[0037] ;

[0038] in, Used to indicate the first The first task point The target robot at the current moment The state of the robot performing the assigned task. Used to indicate the first The first task point The target robot at the current moment The actual spatial location, Used to indicate the first The first task point The target robot at the current moment Actual moving speed Used to indicate the first The task execution location of the assigned task for each target robot. Used to indicate the first The first task point The estimated time when the target robot arrives at the task execution position assigned to it. Used to indicate the first The expected movement speed of a target robot at its assigned task execution location. Used to indicate the first The task status of each target robot when its assigned task is executed. Used to indicate the first The first task point The target robot at the estimated time The state of the robot performing the assigned task. Used to indicate the first The first task point The target robot at the estimated time The estimated spatial location, Used to indicate the first The first task point The target robot at the estimated time The estimated speed of movement, Used to indicate the first The effective cruising radius of each target robot when its assigned task is executed. Used to represent the L2 norm.

[0039] Secondly, this application provides a robot job scheduling device, including a processor and a memory, wherein the memory stores a computer program that can be executed by the processor, and the processor can execute the computer program to implement the multi-robot dynamic task allocation planning method described in any of the foregoing embodiments.

[0040] Thirdly, this application provides a readable storage medium storing a computer program thereon, which, when executed by a computer device, implements the multi-robot dynamic task allocation and planning method described in any of the foregoing embodiments.

[0041] In this case, the beneficial effects of the embodiments of this application may include the following:

[0042] After obtaining the actual spatial position and remaining energy of all idle robots in the target collaborative operation scenario at the current moment, this application calls a robot operation partitioning model adapted to the target collaborative operation scenario. Based on the actual spatial position and remaining energy of each idle robot, robot operation partitioning is performed on all task points in the target collaborative operation scenario. This ensures that the remaining energy of any idle robot in the robot partitioning result of a single task point meets the energy consumption requirements for task execution near that task point. Then, for each task point, all tasks to be executed related to that task point in the target collaborative operation scenario are assigned to each idle robot in the corresponding robot partitioning result. Thus, in the specified dynamic collaborative operation scenario, the real-time status information of the global robots (including actual spatial position and remaining energy) is comprehensively considered. Adaptive dynamic partitioning is performed on multiple heterogeneous robots with different task types in the specified dynamic collaborative operation scenario. The tasks to be executed associated with the task point are assigned to multiple robots in the same task point partition. This achieves efficient and reasonable multi-robot dynamic task coordination and allocation in dynamic and changing scenarios, avoiding task allocation conflicts.

[0043] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a schematic diagram of the robot job scheduling device provided in the embodiments of this application;

[0046] Figure 2 This is one of the flowcharts illustrating the multi-robot dynamic task allocation and planning method provided in the embodiments of this application;

[0047] Figure 3 for Figure 2 A flowchart illustrating the sub-steps included in step S220;

[0048] Figure 4 for Figure 2 A flowchart illustrating the sub-steps included in step S230;

[0049] Figure 5 A second flowchart illustrating the multi-robot dynamic task allocation and planning method provided in this application embodiment;

[0050] Figure 6 This is the third flowchart illustrating the multi-robot dynamic task allocation and planning method provided in the embodiments of this application.

[0051] Icons: 10-Robot job scheduling equipment; 11-Memory; 12-Processor; 13-Communication unit. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0053] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0054] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0055] In the description of this application, it should be understood that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product is in use, or the orientation or positional relationship commonly understood by those skilled in the art. They are used only for the convenience of describing this application and simplifying the description, and are not intended to indicate or imply that the equipment or component referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0056] In the description of this application, it should also be noted that, unless otherwise expressly specified and limited, the terms "set up," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0057] Furthermore, it is understood in the description of this application that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. Those skilled in the art will understand the specific meaning of the above terms in this application based on the specific circumstances.

[0058] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0059] Please refer to Figure 1 , Figure 1 This is a schematic diagram of the robot task scheduling device 10 provided in this application embodiment. In this application embodiment, the robot task scheduling device 10 can communicate with each mobile robot in any specified multi-robot collaborative operation scenario (i.e., target collaborative operation scenario) to obtain the robot status information of each mobile robot in the target collaborative operation scenario in real time or at preset time intervals (including the spatial position and remaining energy of the corresponding mobile robot in the target collaborative operation scenario). Based on the obtained robot status information of each mobile robot and the actual position information of each of the multiple task points in the target collaborative operation scenario, the device adaptively and dynamically partitions the multiple mobile robots with different task types in the target collaborative operation scenario. Then, it assigns the tasks to be executed associated with the task point to each of the multiple mobile robots divided into the same task point partition (wherein, the actual task type of the task to be executed assigned to any mobile robot is consistent with the pre-assigned task type that the mobile robot is responsible for) for execution, thereby achieving an efficient and reasonable multi-robot dynamic task coordination and allocation effect in dynamic and changing scenarios, and avoiding task allocation conflicts. The robot operation scheduling device 10 can be a computer device independent of each mobile robot. The computer device can be, but is not limited to, a personal computer, a laptop, a tablet computer, a server, etc.

[0060] In this embodiment, each task to be executed in the target collaborative operation scenario corresponds to a task execution location. When the task execution location of a task to be executed is near the actual location information of a task point (including when the task execution location of the corresponding task to be executed overlaps with the actual location of the task point), the task to be executed will be included in the list of assignable tasks for that task point (at this time, the task to be executed is associated with that task point). The target collaborative operation scenario involves various types of robots, including but not limited to unmanned transport vehicles, vehicle-arm hybrid robots, and drones, wherein each type of robot corresponds to at least one mobile robot in the target collaborative operation scenario. In one implementation of this embodiment, when the actual location information of each task point in the target collaborative operation scenario is represented by coordinates on a two-dimensional scene map, the spatial position of each mobile robot and the task execution location of each task to be executed will also be represented by coordinates on a two-dimensional scene map.

[0061] In this embodiment, the robot job scheduling device 10 may include a memory 11, a processor 12, and a communication unit 13. The memory 11, the processor 12, and the communication unit 13 are electrically connected to each other directly or indirectly to achieve data transmission or interaction. For example, these components can be electrically connected to each other via one or more communication buses or signal lines.

[0062] In this embodiment, the memory 11 may be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory 11 is used to store computer programs, and the processor 12 can execute the computer programs accordingly after receiving execution instructions.

[0063] Meanwhile, the memory 11 is also used to store a robot operation partitioning model adapted to the target collaborative operation scenario. The robot operation partitioning model is used to perform adaptive dynamic partitioning of multiple task points based on the robot state information of each mobile robot in the target collaborative operation scenario, so as to ensure that the remaining energy of the mobile robot assigned to a single task point can meet its (i.e., the mobile robot's) task execution energy consumption requirements near the corresponding task point (which can be represented by the overall task execution energy consumption of the mobile robot from its own spatial position to the corresponding task point, the overall task execution energy consumption consisting of the initial movement energy consumption of the mobile robot to reach the vicinity of the corresponding task point, the body operation energy consumption required for the mobile robot to perform the task of the type adapted to the task, and the later movement energy consumption of the mobile robot leaving the corresponding task point and returning to the specified parking position (e.g., the aforementioned own spatial position).

[0064] In this embodiment, the robot task partitioning model can be a multi-class support vector machine model trained using robot partitioning sample sets for each task location in the target collaborative task scenario. The robot task partitioning model is periodically updated to adapt to changes in robot state within the target collaborative task scenario. The robot partitioning sample set for each task location can include robot partitioning samples for each mobile robot in the target collaborative task scenario at that task location. Each robot partitioning sample includes an input feature vector composed of the corresponding mobile robot's true spatial position and true remaining energy at the model update time, and a partitioning label for the mobile robot at the corresponding task location. The partitioning label can be represented by a first preset label value or a second preset label value. The first preset label value indicates that the remaining energy of the corresponding mobile robot does not meet its task execution energy consumption requirements near the corresponding task location (in which case the mobile robot cannot actually be assigned to the robot partitioning result for that task location), while the second preset label value indicates that the remaining energy of the corresponding mobile robot meets its task execution energy consumption requirements near the corresponding task location (in which case the mobile robot will actually be assigned to the robot partitioning result for that task location).

[0065] Optionally, in one embodiment of this example, all task points in the target collaborative operation scenario are represented by "-1" in the partition label to indicate a first preset label value, and all task points are represented by "1" in the partition label to indicate a second preset label value.

[0066] Optionally, in another embodiment of this example, all task points in the target collaborative operation scenario use "0" to represent the first preset label value in the partition label, but each task point uses its own task point number to represent the second preset label value in the partition label. In this case, the second preset label values ​​used by different task points are different from each other.

[0067] In this embodiment, the processor 12 can be an integrated circuit chip with signal processing capabilities. The processor 12 can be a general-purpose processor, including at least one of a Central Processing Unit (CPU), Graphics Processing Unit (GPU), Network Processor (NP), Digital Signal Processor (DSP), Application-Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0068] In this embodiment, the communication unit 13 is used to establish a communication connection between the robot job scheduling device 10 and other electronic devices through a network, and to send and receive data through the network, wherein the network includes wired communication networks and wireless communication networks. For example, the robot job scheduling device 10 can obtain information on newly added tasks to be executed, newly added task locations, and / or modified existing task locations configured by the user for the target collaborative operation scenario through the communication unit 13.

[0069] In this embodiment, the robot job scheduling device 10 can pre-store a specific computer program related to the multi-robot dynamic task allocation planning function in the memory 11. By driving the processor 12 to execute the specific computer program, the device comprehensively considers the real-time status information of the global robots in a specified dynamic collaborative operation scenario, adaptively and dynamically partitions multiple heterogeneous robots with different task types in the specified dynamic collaborative operation scenario, and assigns the tasks to be executed associated with the task point to each of the multiple robots divided into the same task point partition. This achieves an efficient and reasonable multi-robot dynamic task coordination and allocation effect in dynamic and ever-changing scenarios, avoiding task allocation conflicts.

[0070] Understandable Figure 1 The block diagram shown is only a schematic diagram of one composition of the robot job scheduling device 10. The robot job scheduling device 10 may also include more than Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown. Figure 1 The components shown can be implemented using hardware, software, or a combination thereof.

[0071] In this application, to ensure that the robot job scheduling device 10 can comprehensively consider the real-time status information of the global robots in a specified dynamic collaborative operation scenario, adaptively and dynamically partition multiple heterogeneous robots with different task types in the specified dynamic collaborative operation scenario, and assign the tasks to be executed associated with the task point to each of the multiple robots divided into the same task point partition, so as to achieve an efficient and reasonable multi-robot dynamic task coordination and allocation effect in a dynamic and ever-changing scenario, this application provides a multi-robot dynamic task allocation planning method to achieve the aforementioned objective. The multi-robot dynamic task allocation planning method provided in this application will be described in detail below.

[0072] Please refer to Figure 2 , Figure 2 This is one of the flowcharts illustrating the multi-robot dynamic task allocation and planning method provided in this application embodiment. In this application embodiment, Figure 2 The multi-robot dynamic task allocation planning method shown may include steps S210 to S230.

[0073] Step S210: Obtain the actual spatial position and actual remaining energy of each idle robot in the target collaborative operation scenario at the current moment.

[0074] In this embodiment, the idle robot is a mobile robot that has not been assigned a task in the target collaborative operation scenario.

[0075] Step S220: Call the robot task partitioning model adapted to the target collaborative operation scenario. Based on the actual spatial position and actual remaining energy of each idle robot, perform robot task partitioning processing on all task points in the target collaborative operation scenario to obtain the robot partitioning results of all task points at the current time.

[0076] In this embodiment, the robot partitioning result for each task location consists of at least one idle robot in the target collaborative operation scenario at the current time. The actual remaining energy of any idle robot in a single robot partitioning result meets the energy consumption requirements for its task execution when performing a task of its own type adaptation (i.e., any task matching the pre-assigned task type of the idle robot) near the corresponding task location. When the robot task partitioning model adapted to the target collaborative operation scenario is implemented using a multi-class support vector machine model, the robot task partitioning model may include a multi-classifier model corresponding to all existing task points in the target collaborative operation scenario. This multi-classifier model can be equivalently composed of binary classification models matched by each of the task points. The binary classification model matched by each task point is used to detect whether any idle robot is suitable to be assigned to the corresponding task point partition (i.e., the robot partitioning result of the corresponding task point) based on its current actual spatial position and actual remaining energy. Each binary classification model can be implemented using a linear support vector machine model with slack variables (which can be converted into a convex quadratic programming problem), and the dual problem of the binary classification model is solved using a linear kernel function. The classification decision function of the binary classification model is constructed using a Gaussian kernel function to ensure that the finally trained binary classification model has the robot partitioning recognition function for the matched task points. Each binary classification model is trained based on the robot partitioning sample set of the corresponding task point.

[0077] Based on this, please refer to Figure 3 , Figure 3 yes Figure 2 A flowchart illustrating the sub-steps included in step S220. In this embodiment, step S220 may include sub-steps S221 to S223 to achieve adaptive dynamic partitioning of multiple idle robots in a dynamic collaborative operation scenario.

[0078] Sub-step S221: For each idle robot, perform vector synthesis processing on the actual spatial position and actual remaining energy of the idle robot to obtain the actual input feature vector of the idle robot at the robot operation partition model.

[0079] Sub-step S222: Call the multi-classifier model included in the robot task partitioning model to predict the partition labels for all task points based on the actual input feature vector, and obtain the actual partition label values ​​of the idle robot at each task point.

[0080] Sub-step S223: Add the idle robot to the robot partitioning result of the target task point whose actual partition label value is the second preset label value.

[0081] Therefore, this application can achieve adaptive dynamic partitioning of multiple idle robots in a dynamic collaborative operation scenario by executing the above sub-steps S221 to S223.

[0082] Step S230: For each task location, assign all tasks to be executed at that task location in the target collaborative operation scenario to the respective idle robots in the corresponding robot partitioning results.

[0083] In this embodiment, after completing the robot partitioning process for all idle robots in the target collaborative operation scenario, for each task point in the target collaborative operation scenario, based on the actual task types of all tasks to be executed recorded in the allocable task list for that task point in the target collaborative operation scenario, and the pre-assigned task types of each idle robot in the robot partitioning result of that task point, each task to be executed involving that task point is assigned to each idle robot assigned to that task point for execution according to the task type matching principle. This achieves an efficient and reasonable multi-robot dynamic task coordination and allocation effect in dynamic and ever-changing scenarios, avoiding task allocation conflicts.

[0084] Alternatively, please refer to Figure 4 , Figure 4 yes Figure 2 A flowchart illustrating the sub-steps included in step S230. In this embodiment of the application, for each task location in the target collaborative operation scenario, step S230 may include sub-steps S231 to S233 to ensure that multiple mobile robots assigned to the same task location partition can each be assigned to a task to be executed that is associated with that task location and matches the task type, thus avoiding task allocation conflicts across the entire dynamic and changing scenario.

[0085] Sub-step S231: Based on the actual task type of each of the tasks to be executed involved in the task location, perform task grouping processing on all tasks to be executed to obtain at least one task group.

[0086] In this embodiment, the actual task type of all tasks to be executed included in each task group at a single task location remains consistent.

[0087] Sub-step S232: Based on the pre-assigned task type of all idle robots in the robot partitioning result of the task location, perform robot grouping processing on all idle robots in the robot partitioning result to obtain at least one robot group.

[0088] In this embodiment, the pre-assigned task types of each idle robot in each robot group at a single task location remain consistent.

[0089] Sub-step S233: For each task group, assign each of the tasks to be executed in the task group to an idle robot in the target robot group that matches the task type of the task group.

[0090] In one embodiment of this invention, the robot job scheduling device 10 can randomly assign all the tasks to be executed in a single task group to an idle robot in the corresponding target robot group.

[0091] In another embodiment of this example, during the execution of sub-step S233, the robot task scheduling device 10 may consider the task execution position of each task to be executed in the corresponding task group, the actual spatial position of each idle robot in the corresponding target robot group, and task execution performance parameters (which may include information such as the radius of motion of the unmanned transport vehicle, maximum travel speed, maximum travel acceleration, maximum travel angular velocity, average travel energy consumption speed, and average energy consumption speed of the transport operation, etc., and the radius of motion of the vehicle-arm composite robot, maximum seat travel speed, maximum seat travel acceleration, maximum seat travel angular velocity, robotic arm extension length, maximum robotic arm travel speed, maximum robotic arm travel acceleration, and maximum robotic arm closing length). Information such as pitch angular velocity, average energy consumption speed of the vehicle seat during travel, average energy consumption speed of the vehicle seat during standby, average energy consumption speed of the robotic arm during operation, and average energy consumption speed of the robotic arm during standby, as well as information such as the maximum altitude range, maximum ascent speed, maximum ascent acceleration, maximum travel speed, maximum travel acceleration, maximum descent speed, maximum descent acceleration, average energy consumption speed of the body during ascent and descent, average energy consumption speed of the body during travel, and average energy consumption speed of the body during hovering, etc., are used to perform dynamic task allocation processing with the aim of "minimizing the sum of the overall energy consumption of all idle robots in the target robot group for the tasks assigned in the task group", so that each task to be executed in the task group can be assigned to an idle robot that meets the requirement of minimizing the overall energy consumption of task execution.

[0092] Therefore, by executing the above sub-steps S231 to S233, this application ensures that for each task point in the target collaborative operation scenario, multiple mobile robots divided into the same task point partition can be assigned to an execution task that is associated with the task point and matches the task type, thereby avoiding task allocation conflicts in the entire scope of the dynamic and ever-changing scenario.

[0093] This application can perform the above steps S210 to S230 to comprehensively consider the real-time status information of the global robots in a specified dynamic collaborative operation scenario, adaptively and dynamically partition multiple heterogeneous robots with different task types in the specified dynamic collaborative operation scenario, and assign the tasks to be executed associated with the task point to each of the multiple robots that are assigned to the same task point partition, so as to achieve an efficient and reasonable multi-robot dynamic task coordination and allocation effect in dynamic and ever-changing scenarios.

[0094] Alternatively, please refer to Figure 5 , Figure 5 This is the second flowchart illustrating the multi-robot dynamic task allocation and planning method provided in this application embodiment. In this application embodiment, [the method is related to...]. Figure 2 Compared to the multi-robot dynamic task allocation planning method shown, Figure 5 The multi-robot dynamic task allocation planning method shown may also include steps S310 to S340 to realize the periodic update function of the robot operation partitioning model, ensuring that the updated robot operation partitioning model can adapt to the robot state changes in the target collaborative operation scenario, and can provide a more accurate robot adaptive dynamic partitioning function.

[0095] Step S310: Obtain the actual spatial position and actual remaining energy of all mobile robots in the target collaborative operation scenario according to a preset time interval.

[0096] In this embodiment, all mobile robots in step S310 include idle robots without assigned tasks and task-assigned robots.

[0097] Step S320: For each mobile robot, calculate the overall energy consumption of the mobile robot from the corresponding real space location to different task points based on the task execution performance parameters of the mobile robot, and determine the theoretical partition label of the mobile robot at different task points based on the energy support status of the mobile robot's actual remaining energy for the overall energy consumption of task execution at different task points.

[0098] In this embodiment, for each mobile robot, the travel time of the mobile robot from its real spatial location to different task locations can be predicted using a trapezoidal velocity curve planning algorithm, based on the maximum travel speed and maximum travel acceleration included in the task execution performance parameters of the mobile robot, as well as the actual spatial location of the mobile robot and the actual location information of different task locations. Combined with the average execution time of the mobile robot for type-adaptive tasks and the various average energy consumption speeds included in the corresponding task execution performance parameters, the overall energy consumption of the mobile robot from its corresponding real spatial location to different task locations can be predicted to obtain the overall energy consumption of the task execution of the mobile robot from its corresponding real spatial location to different task locations.

[0099] Then, the overall energy consumption of the mobile robot at different task locations can be compared with the actual remaining energy of the mobile robot to determine the energy support of the actual remaining energy of the mobile robot for the overall energy consumption of task execution at different task locations, and then determine whether the theoretical partition label of the mobile robot at different task locations should be represented by the first preset label value or the second preset label value.

[0100] It is understandable that the above step "determine the theoretical partition label of the mobile robot at different task locations based on the energy support status of the mobile robot's actual remaining energy for the overall energy consumption of task execution at different task locations" may include sub-steps A to C to ensure that no mobile robot is assigned to multiple task location partitions at the same time.

[0101] Sub-step A: Detect whether the overall energy consumption of the mobile robot in performing tasks at different task locations is less than the actual remaining energy of the mobile robot.

[0102] Sub-step B: When the overall energy consumption of task execution at at least one first task point is not less than the actual remaining energy, directly assign a label value to the theoretical partition label of the mobile robot at the at least one first task point using the first preset label value. The first preset label value is used to characterize that the remaining energy of the mobile robot does not meet its task execution energy consumption requirements near the corresponding task point.

[0103] Sub-step C: When the overall energy consumption of task execution at at least one second task point is less than the actual remaining energy, a second preset label value is used to assign a theoretical partition label to the mobile robot at the second task point with the lowest overall energy consumption, and a first preset label value is used to assign a theoretical partition label to the mobile robot at other second task points. The second preset label value indicates that the mobile robot's remaining energy meets its task execution energy consumption requirements near the corresponding task point.

[0104] Step S330: For each task location, construct a robot partition sample set for that task location based on the actual spatial position and actual remaining energy of all mobile robots, as well as the theoretical partition labels of all mobile robots at that task location.

[0105] In this embodiment, the robot partition sample set for each task location includes robot partition samples of all mobile robots at that task location; each robot partition sample includes a sample input feature vector consisting of the actual spatial position and actual remaining energy of the corresponding mobile robot, and the theoretical partition label of the corresponding mobile robot at that task location.

[0106] Step S340: Train a multi-class support vector machine model based on the robot partition sample set for each task location, and use the trained multi-class support vector machine model as the robot task partition model.

[0107] Therefore, by executing the above steps S310 to S340, this application can realize the function of periodically updating the robot operation partitioning model, ensuring that the updated robot operation partitioning model can adapt to the robot state changes in the target collaborative operation scenario, and can provide a more accurate robot adaptive dynamic partitioning function.

[0108] Alternatively, please refer to Figure 6 , Figure 6 This is the third flowchart illustrating the multi-robot dynamic task allocation and planning method provided in this application embodiment. In this application embodiment, with... Figure 2 or Figure 5 Compared to the multi-robot dynamic task allocation planning method shown, Figure 6 The multi-robot dynamic task allocation planning method shown may also include steps S240 to S250, to plan the movement trajectory for each target robot assigned to the task at each task point, so as to ensure that the planned movement trajectory can maximize the global path cumulative utility function of the task point (i.e. minimize the global total energy consumption) while enabling each target robot to achieve the effect of coordinated execution of dynamic and static tasks.

[0109] Step S240: For each task point, based on the task execution positions of all tasks to be executed involved in the task point, and the actual spatial positions of all target robots assigned to the tasks to be executed at the task point, with the aim of maximizing the global path cumulative utility function of the task point, perform collaborative motion trajectory planning for all target robots to obtain the robot motion trajectory of each target robot when satisfying the execution constraints of the assigned tasks.

[0110] In this embodiment, the first target collaborative operation scenario The global path cumulative utility function for each task point is represented by the following equation:

[0111] ;

[0112] in, Used to indicate the first Global path cumulative utility function for each task location Used to indicate the first The cumulative length of the robot's straight path at each task point. Used to indicate the first Total number of target robots at each task location. Used to indicate the first The conversion coefficient between the robot's energy consumption and the robot's straight-line travel distance at each task location. Used to indicate the first The first task point The energy consumption of a target robot moving from its current actual spatial location to its assigned task execution location. Used to indicate the first The first task point The time-related aspects of the robot's trajectory planning process for a target robot. The function of change of acceleration during movement, Used to indicate the first The first task point The target robot at the current moment The actual spatial location, Used to indicate the first The task execution location of the assigned task for each target robot. Used to indicate the first The first task point The estimated time when the target robot arrives at the task execution position assigned to it. Used to represent the L2 norm. Therefore, the effect of maximizing the global path cumulative utility function can be regarded as the total global travel energy cost (i.e., Minimize effect.

[0113] Furthermore, the tasks to be executed in the target collaborative operation scenario can be either fixed-point static tasks (i.e., tasks that require the corresponding robot to perform at a fixed task execution position, such as handling tasks or loading / unloading tasks) or dynamic patrol tasks (i.e., tasks that require the corresponding robot to patrol and inspect around a certain task execution position). In other words, the target collaborative operation scenario can be a scenario of coordinated execution of static and dynamic tasks. Therefore, when the first... The first task point When the assigned task of a target robot is a fixed-point static task, the execution constraints of the assigned task of the target robot are expressed by the following equation:

[0114] ;

[0115] And when the target collaborative operation scenario is the first The first task point When the assigned task of a target robot is a dynamic cruise task, the execution constraints of the assigned task of the target robot are expressed by the following equation:

[0116] ;

[0117] in, Used to indicate the first The first task point The target robot at the current moment The state of the robot performing the assigned task. Used to indicate the first The first task point The target robot at the current moment The actual spatial location, Used to indicate the first The first task point The target robot at the current moment Actual moving speed Used to indicate the first The task execution location of the assigned task for each target robot. Used to indicate the first The first task point The estimated time when the target robot arrives at the task execution position assigned to it. Used to indicate the first The expected movement speed of a target robot at its assigned task execution location. Used to indicate the first The task status of each target robot when its assigned task is executed. Used to indicate the first The first task point The target robot at the estimated time The state of the robot performing the assigned task. Used to indicate the first The first task point The target robot at the estimated time The estimated spatial location, Used to indicate the first The first task point The target robot at the estimated time The estimated speed of movement, Used to indicate the first The effective cruising radius of each target robot when its assigned task is executed. Used to represent the L2 norm.

[0118] Among them, the The estimated time when a target robot arrives at the task execution position of its assigned task can be obtained by referring to the travel time calculation process in step S320 above (that is, the estimated time can be the time point at which half of the travel time is added to the current time). For a target robot assigned to a dynamic cruise task, it can be required to arrive at the edge of the dynamic cruise area at a certain time point before the estimated time (this time point can be a preset time interval from the estimated time) during the corresponding robot movement trajectory planning process, and continuously perform roving movement along the edge of the dynamic cruise area during the time period between this time point and the estimated time.

[0119] Step S250: Control all target robots at the task location to move to the task execution position of the assigned task according to the corresponding robot movement trajectory, so as to drive all target robots to execute the assigned task.

[0120] Therefore, by executing the above steps S240 to S250, this application can plan the movement trajectory for all target robots assigned to the task at each task point, ensuring that the planned movement trajectory can maximize the global path cumulative utility function of the task point (i.e. minimize the total global travel energy consumption) while enabling each target robot to achieve the effect of coordinated execution of dynamic and static tasks.

[0121] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0122] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part. If the various functions provided in this application are implemented in the form of software functional modules and sold or used as independent products, they can be stored in a storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause any computer device to execute all or part of the steps of the methods described in the various embodiments of this application as the robot job scheduling device 10. The aforementioned readable storage medium includes: USB flash drive, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk, and other media capable of storing program code.

[0123] The above descriptions are merely various embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A multi-robot dynamic task allocation and planning method, characterized in that, The method includes: Obtain the actual spatial position and actual remaining energy of each idle robot in the target collaborative operation scenario at the current moment; The robot operation partitioning model adapted to the target collaborative operation scenario is invoked. Based on the actual spatial position and actual remaining energy of each of the idle robots, robot operation partitioning is performed on all task points in the target collaborative operation scenario to obtain the robot partitioning results of all task points at the current time. The actual remaining energy of any idle robot in the robot partitioning result of a single task point meets its task execution energy consumption requirements near the corresponding task point. For each task location, all tasks to be executed at that task location in the target collaborative operation scenario are assigned to the respective idle robots in the corresponding robot partitioning results; The robot task partitioning model includes a multi-classifier model corresponding to all task locations. The step of calling the robot task partitioning model adapted to the target collaborative task scenario, and performing robot task partitioning processing on all task locations in the target collaborative task scenario based on the actual spatial positions and remaining energy of all idle robots, to obtain the robot partitioning results for all task locations at the current time, includes: For each idle robot, the actual spatial position and actual remaining energy of the idle robot are vector synthesized to obtain the actual input feature vector of the idle robot at the robot operation partition model; The multi-classifier model included in the robot task partitioning model is invoked to predict the partition labels for all task locations based on the actual input feature vector, thereby obtaining the actual partition label values ​​of the idle robot at each task location; The idle robot is added to the robot partitioning result of the target task point whose actual partition label value is the second preset label value.

2. The method according to claim 1, characterized in that, The method further includes: The actual spatial position and actual remaining energy of each mobile robot in the target collaborative operation scenario are obtained at preset time intervals. For each mobile robot, based on the task execution performance parameters of the mobile robot, calculate the overall energy consumption of the mobile robot from the corresponding real space location to different task points, and determine the theoretical partition label of the mobile robot at different task points based on the energy support status of the mobile robot's actual remaining energy for the overall energy consumption of task execution at different task points. For each task location, a robot partition sample set is constructed based on the actual spatial position and actual remaining energy of all mobile robots, as well as the theoretical partition label of each mobile robot at that task location. The robot partition sample set includes robot partition samples of each mobile robot at that task location. Each robot partition sample includes a sample input feature vector composed of the actual spatial position and actual remaining energy of the corresponding mobile robot, and the theoretical partition label of the corresponding mobile robot at that task location. A multi-class support vector machine model is trained based on the robot partition sample set for each of the task locations, and the trained multi-class support vector machine model is used as the robot task partition model.

3. The method according to claim 2, characterized in that, For each mobile robot, the step of determining the theoretical partition label of the mobile robot at different task locations based on the energy support status of the overall energy consumption of task execution at different task locations according to the actual remaining energy of the mobile robot includes: The test was conducted to determine whether the overall energy consumption of the mobile robot at different task locations was less than the robot's actual remaining energy. When the overall energy consumption of task execution at at least one first task point is not less than the actual remaining energy, the theoretical partition label of the mobile robot at the at least one first task point is directly assigned a label value using the first preset label value; wherein the first preset label value is used to characterize that the remaining energy of the mobile robot does not meet its task execution energy consumption requirements near the corresponding task point. When the overall energy consumption of task execution at at least one second task point is less than the actual remaining energy, a second preset label value is used to assign a theoretical partition label to the mobile robot at the second task point with the lowest overall energy consumption of task execution, and a first preset label value is used to assign a theoretical partition label to the mobile robot at other second task points; wherein the second preset label value is used to characterize that the remaining energy of the mobile robot meets its task execution energy consumption requirements near the corresponding task point.

4. The method according to any one of claims 1-3, characterized in that, For each task location, the step of assigning all pending tasks related to that task location in the target collaborative operation scenario to the respective idle robots in the corresponding robot partitioning result includes: Based on the actual task type of each of the tasks to be executed involved in the task location, the tasks to be executed are grouped to obtain at least one task group, wherein the actual task type of each task group is consistent. Based on the pre-assigned task type of all idle robots in the robot partitioning result of the task location, all idle robots in the robot partitioning result are grouped to obtain at least one robot group, wherein the pre-assigned task type of each idle robot in each robot group is consistent. For each task group, all tasks to be executed within that task group are individually assigned to an idle robot in the target robot group that matches the task type of that task group.

5. The method according to any one of claims 1-3, characterized in that, The method further includes: For each task point, based on the task execution positions of all tasks to be executed involved in that task point, and the actual spatial positions of all target robots assigned to the tasks to be executed at that task point, with the aim of maximizing the global path cumulative utility function of that task point, collaborative planning of the movement trajectory of all target robots is performed to obtain the robot movement trajectory of each target robot when the assigned task execution constraints are met. All target robots at the task location are controlled to move to the task execution position of the assigned task according to their corresponding robot movement trajectories, so as to drive all target robots to execute the assigned task.

6. The method according to claim 5, characterized in that, The first in the target collaborative operation scenario The global path cumulative utility function for each task point is represented by the following equation: ； in, Used to indicate the first Global path cumulative utility function for each task location Used to indicate the first The cumulative length of the robot's straight path at each task point. Used to indicate the first Total number of target robots at each task location. Used to indicate the first The conversion coefficient between the robot's energy consumption and the robot's straight-line travel distance at each task location. Used to indicate the first The first task point The energy consumption of a target robot moving from its current actual spatial location to its assigned task execution location. Used to indicate the first The first task point The time-related aspects of the robot's trajectory planning process for a target robot. The function of change of acceleration during movement, Used to indicate the first The first task point The target robot at the current moment The actual spatial location, Used to indicate the first The task execution location of the assigned task for each target robot. Used to indicate the first The first task point The estimated time when the target robot arrives at the task execution position assigned to it. Used to represent the L2 norm.

7. The method according to claim 5, characterized in that, In the target collaborative operation scenario, the first The first task point When the assigned task of a target robot is a fixed-point static task, the execution constraints of the assigned task of the target robot are expressed by the following equation: ； And when the target collaborative operation scenario is the first The first task point When the assigned task of a target robot is a dynamic cruise task, the execution constraints of the assigned task of the target robot are expressed by the following equation: ； in, Used to indicate the first The first task point The target robot at the current moment The state of the robot performing the assigned task. Used to indicate the first The first task point The target robot at the current moment The actual spatial location, Used to indicate the first The first task point The target robot at the current moment Actual moving speed Used to indicate the first The task execution location of the assigned task for each target robot. Used to indicate the first The first task point The estimated time when the target robot arrives at the task execution position assigned to it. Used to indicate the first The expected movement speed of a target robot at its assigned task execution location. Used to indicate the first The task status of each target robot when its assigned task is executed. Used to indicate the first The first task point The target robot at the estimated time The state of the robot performing the assigned task. Used to indicate the first The first task point The target robot at the estimated time The estimated spatial location, Used to indicate the first The first task point The target robot at the estimated time The estimated speed of movement, Used to indicate the first The effective cruising radius of each target robot when its assigned task is executed. Used to represent the L2 norm.

8. A robot job scheduling device, characterized in that, It includes a processor and a memory, the memory storing a computer program that can be executed by the processor, the processor executing the computer program to implement the multi-robot dynamic task allocation planning method according to any one of claims 1-7.

9. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a computer device, it implements the multi-robot dynamic task allocation and planning method according to any one of claims 1-7.

Images