Task decision-making method, device and equipment for robot and computer program product

Abstract

The invention is suitable for the technical field of robots, and provides a task decision-making method, device and equipment for a robot and a computer program product. The task decision-making method of the robot comprises the following steps: constructing an environment model of an environment where the robot is located; the environment model and historical operation data of the robot are input into a behavior planning model, the behavior planning model is used for outputting a target behavior scheme enabling a target function to be maximized, and the target function is used for representing safety when the robot executes corresponding behaviors; and controlling the robot to execute a target task according to the target behavior scheme. According to the embodiment of the invention, the task decision can adapt to the environment and the actual operation state of the robot, and the operation safety and reliability of the robot are improved.

Description

Technical Field

[0001] The present application belongs to the field of robotics technology, and in particular, relates to a robot task decision method, device, equipment and computer program product. Background Art

[0002] With the rapid development of artificial intelligence technology, service robots have gradually entered people's daily lives, especially showing great potential in the fields of education, medical care, family companionship, etc. Existing service robots mostly rely on preset programs or remote control, lack sufficient autonomous decision-making capabilities, and cannot flexibly respond to complex and changing actual environments and the special limitations of their own structures. Summary of the invention

[0003] The embodiments of the present application provide a robot task decision method, apparatus, device and computer program product, which can make the task decision adapt to the environment and the actual operating state of the robot, and improve the safety and reliability of the robot operation.

[0004] A first aspect of an embodiment of the present application provides a robot task decision-making method, comprising: constructing an environmental model of the robot's environment; inputting the environmental model and the robot's historical operating data into a behavior planning model, the behavior planning model being used to output a target behavior plan that maximizes an objective function, the objective function being used to characterize the safety of the robot when performing a corresponding behavior; and controlling the robot to perform a target task according to the target behavior plan.

[0005] In some embodiments of the first aspect, constructing an environmental model of the robot's environment includes: acquiring multimodal data collected by the robot; performing semantic recognition on the multimodal data to obtain associations between different objects in the environment; and generating the environmental model based on the associations.

[0006] In some embodiments of the first aspect, performing semantic recognition on the multimodal data to obtain the association relationship between different objects in the environment includes: preprocessing the multimodal data; performing data fusion on the preprocessed data to obtain fused data; and performing semantic recognition on the fused data to obtain the association relationship.

[0007] In some embodiments of the first aspect, the process of controlling the robot to perform the target task according to the target behavior plan also includes: acquiring real-time feedback data of the robot; and adjusting the target behavior plan according to the real-time feedback data.

[0008] In some embodiments of the first aspect, when the real-time feedback data is obstacle data, adjusting the target behavior plan according to the real-time feedback data includes: classifying obstacles according to the obstacle data; and controlling the robot to wait or adjust the movement path of the robot according to the obstacle type.

[0009] In some embodiments of the first aspect, when the real-time feedback data is task interruption information, adjusting the target behavior plan according to the real-time feedback data includes: saving the current task progress of the target task; when the target task is re-triggered, executing the target task according to the current task progress.

[0010] In some embodiments of the first aspect, after controlling the robot to perform the target task according to the target behavior plan, it also includes: obtaining task performance information of the target task; determining parameters to be adjusted during the execution of the target task based on the task performance information; and adjusting the parameters to be adjusted.

[0011] A second aspect of an embodiment of the present application provides a robot task decision-making device, including: an environment model construction unit, used to construct an environment model of the robot's environment; a behavior planning unit, used to input the environment model and the robot's historical operation data into the behavior planning model, the behavior planning model is used to output a target behavior plan that maximizes an objective function, and the objective function is used to characterize the safety of the robot when performing a corresponding behavior; a task execution unit, used to control the robot to perform a target task according to the target behavior plan.

[0012] A third aspect of an embodiment of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned robot task decision method when executing the computer program.

[0013] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the task decision method of the robot are implemented.

[0014] A fifth aspect of an embodiment of the present application provides a computer program product, which, when executed on an electronic device, enables the electronic device to execute the steps of the above-mentioned robot task decision method.

[0015] In an implementation manner of the present application, an environmental model of the robot's environment is constructed, and the environmental model and the robot's historical operating data are input into a behavior planning model, so that the behavior planning model outputs a target behavior plan that maximizes the objective function, and the robot is controlled to perform the target task according to the target behavior plan. Since the objective function is used to characterize the safety of the robot when performing the corresponding behavior, the task decision can be adapted to the environment and the actual operating state of the robot, thereby improving the safety and reliability of the robot's operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0017] Figure 1 It is a schematic diagram of the implementation flow of a robot task decision method provided in an embodiment of the present application;

[0018] Figure 2 It is a schematic diagram of the architecture of the pushing robot provided in the embodiment of the present application;

[0019] Figure 3 A flowchart of a task execution provided by an embodiment of the present application;

[0020] Figure 4 It is a structural schematic diagram of a task decision device of a robot provided in an embodiment of the present application;

[0021] Figure 5 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are protected by the present application.

[0023] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0024] In the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0025] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0026] With the rapid development of artificial intelligence technology, service robots have gradually entered people's daily lives, especially showing great potential in the fields of education, medical care, family companionship, etc. Existing service robots mostly rely on preset programs or remote control, lack sufficient autonomous decision-making capabilities, and cannot flexibly respond to complex and changing actual environments and the special limitations of their own structures.

[0027] In view of this, the present application proposes a robot task decision method, which can make the task decision adapt to the environment and the actual operating state of the robot, and improve the safety and reliability of the robot operation.

[0028] In order to illustrate the technical solution of the present application, a specific embodiment is provided below for illustration.

[0029] Figure 1 The present invention provides a schematic diagram of a process flow of implementing a robot task decision method provided by an embodiment of the present invention, which can be applied to an electronic device. The electronic device can be a computer, a mobile phone, or other intelligent device for controlling or communicating with the robot, or it can be the robot itself.

[0030] Specifically, the task decision method of the robot may include the following steps S101 to S103.

[0031] Step S101, constructing an environment model of the robot's environment.

[0032] In an embodiment of the present application, the environmental model is used to describe the situation of the robot's environment. For example, it can characterize the state parameters of the environment (such as temperature, humidity, etc.) or the information of each object in the environment (such as the density of obstacles, the type of obstacles, the relationship between different objects, etc.).

[0033] Step S102, inputting the environment model and the robot's historical operation data into the behavior planning model.

[0034] Among them, historical operation data refers to the data recorded by the robot during its past operation, which can reflect the robot’s past task execution experience, and then reflect some structural characteristics and program functions of the robot itself.

[0035] The behavior planning model can be used to output a target behavior plan that maximizes the objective function. The objective function is used to characterize the safety of the robot when performing the corresponding behavior. Maximizing the objective function means that the robot can achieve the highest safety when performing the task according to the target behavior plan, such as the maximum sum of the distances to various obstacles, the minimum speed difference between the robot's average running speed and the optimal speed, etc. In practical applications, the objective function can integrate different factors related to safety, such as the distance to obstacles, running speed, acceleration, number of turns, etc., to take into account the entire process of executing the task.

[0036] Using the environmental model and the robot's historical operating data as the input of the behavior planning model can ensure that the output target behavior plan can adapt to the robot's environment and conform to the robot's own operating experience and operating status.

[0037] Step S103, controlling the robot to perform the target task according to the target behavior plan.

[0038] In the implementation manner of the present application, the target behavior plan may be a sequence of actions to be performed by the robot, and the robot may complete each action to be performed one by one according to the target behavior plan to perform the target task.

[0039] In an implementation manner of the present application, an environmental model of the robot's environment is constructed, and the environmental model and the robot's historical operating data are input into a behavior planning model, so that the behavior planning model outputs a target behavior plan that maximizes the objective function, and the robot is controlled to perform the target task according to the target behavior plan. Since the objective function is used to characterize the safety of the robot when performing the corresponding behavior, the task decision can be adapted to the environment and the actual operating state of the robot, thereby improving the safety and reliability of the robot's operation.

[0040] In some embodiments of the present application, constructing an environmental model of the robot's environment may include: acquiring multimodal data collected by the robot; performing semantic recognition on the multimodal data to obtain the association relationship between different objects in the environment; and generating an environmental model based on the association relationship.

[0041] Specifically, the robot can be equipped with various types of sensors, such as visual cameras, infrared sensors, ultrasonic sensors, laser radars, etc. These sensors can capture environmental information from different dimensions, including but not limited to the position, shape, color of objects, and the layout and lighting conditions of the environment, and obtain multimodal data.

[0042] Based on multimodal data, semantic recognition of objects in the environment can be performed from different dimensions, identifying objects, people, actions in the environment and the relationships between them, and ultimately building a dynamic and interpretable environmental model.

[0043] In order to improve the reliability of the environmental model, in some embodiments of the present application, semantic recognition is performed on multimodal data to obtain the association relationship between different objects in the environment, which may include: preprocessing the multimodal data; performing data fusion on the preprocessed data to obtain fused data; and performing semantic recognition on the fused data to obtain the association relationship.

[0044] The preprocessing may include but is not limited to data cleaning, data denoising, and normalization. Data fusion may adopt the existing multimodal data fusion method, aiming to integrate information from different sensors into a more comprehensive and accurate fused data.

[0045] In some embodiments of the present application, the environment model and historical operation data are input into the behavior planning model as a decision basis, and the objective function can be optimized, different behavior options can be evaluated, and the target behavior plan that can maximize the objective function in the current environment can be found. At this time, according to the target behavior plan, specific action instructions for executing the target task can be generated.

[0046] In some embodiments of the present application, before constructing an environment model of the robot's environment, task allocation may be performed to determine a target task.

[0047] Specifically, the task information of each task to be performed may be obtained, and the task information may include but is not limited to the task urgency, the task target location, etc. According to the task information, the environment information and the current state of the robot, the target task may be determined in each task to be performed to improve the service efficiency.

[0048] In some embodiments of the present application, the process of controlling the robot to perform a target task according to a target behavior plan may also include: obtaining real-time feedback data of the robot; and adjusting the target behavior plan according to the real-time feedback data.

[0049] Specifically, the real-time feedback data can be the execution results fed back by the robot's actuator, or it can be the information collected by the robot in real time through sensors. According to the real-time feedback data, the target behavior plan can be dynamically adjusted to ensure the smooth execution of the target task.

[0050] In some embodiments of the present application, the real-time feedback data may be obstacle data. In this case, adjusting the target behavior plan according to the real-time feedback data may include: classifying obstacles; and controlling the robot to wait or adjusting the robot's moving path according to the obstacle type.

[0051] Specifically, the obstacle data may be point cloud data, and point cloud classification may be performed based on the point cloud data to divide the obstacles into static obstacles and dynamic obstacles. For dynamic obstacles, considering that they will leave the path of the robot, the robot may be controlled to wait for a preset time and move again along the original path after the preset time. For static obstacles, since they do not have the ability to move, the path may be re-planned according to the target position and the robot's moving path may be adjusted to change the robot's actions.

[0052] In some other embodiments of the present application, the real-time feedback data may be task interruption information. At this time, adjusting the target behavior plan according to the real-time feedback data may include: saving the current task progress of the target task; when the target task is re-triggered, executing the target task according to the current task progress.

[0053] Specifically, when an error occurs in the robot's actuator during the execution of an action, such as failure to move to a predetermined position or failure to reset, task interruption information may be generated. In response to the task interruption information, the current task progress of the target task may be saved. Subsequently, the robot may be controlled to return to the waiting point, or wait in place, until the target task is re-triggered, and the target task is re-executed according to the current task progress.

[0054] In some embodiments of the present application, after controlling the robot to perform the target task according to the target behavior plan, the process may also include: recording the task performance information of the target task in real time. The task performance information may include but is not limited to the robot's actions, task execution results, completion speed, completion accuracy, etc. during the execution of the target task.

[0055] Accordingly, in some embodiments of the present application, after controlling the robot to perform the target task according to the target behavior plan, it may also include: obtaining task performance information of the target task, determining the parameters to be adjusted during the execution of the target task based on the task performance information, and adjusting the parameters to be adjusted.

[0056] Specifically, based on the task performance information, the completion quality of each task node of the target task can be checked. For target task nodes whose completion quality is lower than the quality threshold, the various parameters used in the process of executing the target task node can be used as parameters to be adjusted to adjust the parameters to be adjusted.

[0057] The parameters to be adjusted here may include but are not limited to: navigation parameters, motion parameters, and the position and posture control parameters of the actuators, etc. In this way, according to the execution of the target task, it is possible to determine which dimension has an object that does not meet expectations, and dynamically adjust the parameters of the corresponding dimension to achieve better completion results in future tasks.

[0058] In other implementations, the real-time feedback data may also be the real-time operating status of the robot, task execution results, etc.

[0059] In some embodiments of this application, please refer to Figure 2 Schematic diagram of the robot architecture shown. The robot can be configured with functional modules, which may include but are not limited to: an interaction and data acquisition center, a state strategy center, a task strategy center, a mobile control center, a motor and power module, and a sensor module. The interaction and data acquisition center controls the sensor module to collect data and obtain real-time feedback data. The sensor module can be provided with a variety of sensors to collect multimodal data. The state strategy center is used for data processing, such as preprocessing and fusing sensor data. The task strategy center is used to determine the target action plan. The mobile control center is used to control the robot to move in order to perform actions. The motor and power module is used to supply power and provide kinetic energy.

[0060] Data can be transmitted within each functional module and between each other through a specific communication mechanism. Specifically, EventBusManager can be used to publish events within the process of each functional module, and RobotEventBusManager can be used to publish events between different processes or functional modules. Each functional module can use Robot Skill to describe the life cycle status. Data transmission between functional modules can be achieved through Robot Skill Service, and data can be stored persistently across platforms through Robot Provider.

[0061] In some embodiments of this application, please refer to Figure 3The robot's task execution process is shown. The robot can use multimodal information for processing and output a target action plan, which is executed by the execution processing center. In the process of executing the target task, the real-time feedback data such as the real-time operation status and task operation results are input into the decision tree to adjust the target action plan for execution by the execution agency. At the same time, the decision tree can also adjust the parameters to be adjusted according to the task performance information, so as to provide reference for the execution agency when it executes the task next time.

[0062] It should be noted that, for the sake of simplicity of description, the aforementioned method embodiments are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited to the described order of actions, because according to the present application, certain steps can be performed in other orders.

[0063] like Figure 4 Shown is a schematic structural diagram of a robot task decision device 400 provided in an embodiment of the present application, wherein the robot task decision device 400 is configured on an electronic device.

[0064] Specifically, the robot's task decision device 400 may include:

[0065] An environment model building unit 401 is used to build an environment model of the environment where the robot is located;

[0066] A behavior planning unit 402, configured to input the environment model and the historical operation data of the robot into a behavior planning model, wherein the behavior planning model is configured to output a target behavior plan that maximizes an objective function, wherein the objective function is configured to characterize the safety of the robot when executing a corresponding behavior;

[0067] The task execution unit 403 is used to control the robot to execute the target task according to the target behavior plan.

[0068] In some embodiments of the present application, the environment model construction unit 401 is specifically used to: obtain multimodal data collected by the robot; perform semantic recognition on the multimodal data to obtain the association relationship between different objects in the environment; and generate the environment model based on the association relationship.

[0069] In some implementations of the present application, the environment model building unit 401 is specifically used to: preprocess the multimodal data; perform data fusion on the preprocessed data to obtain fused data; perform semantic recognition on the fused data to obtain the association relationship.

[0070] In some implementations of the present application, the task execution unit 403 may also be used to: obtain real-time feedback data of the robot; and adjust the target behavior plan according to the real-time feedback data.

[0071] In some embodiments of the present application, when the real-time feedback data is obstacle data, the task execution unit 403 can also be used to: classify obstacles according to the obstacle data; and control the robot to wait or adjust the movement path of the robot according to the obstacle type.

[0072] In some embodiments of the present application, when the real-time feedback data is task interruption information, the task execution unit 403 can also be used to: save the current task progress of the target task; when the target task is re-triggered, execute the target task according to the current task progress.

[0073] In some implementations of the present application, the task execution unit 403 may also be used to: obtain task performance information of the target task; determine parameters to be adjusted during the execution of the target task according to the task performance information; and adjust the parameters to be adjusted.

[0074] It should be noted that, for the convenience and simplicity of description, the specific working process of the task decision device 400 of the robot can be referred to Figures 1 to 3 The corresponding process of the method will not be repeated here.

[0075] like Figure 5 , which is a schematic diagram of an electronic device provided in an embodiment of the present application. Specifically, the electronic device 5 may include: a processor 50, a memory 51, and a computer program 52 stored in the memory 51 and executable on the processor 50, such as a robot task decision program. When the processor 50 executes the computer program 52, the steps in the above-mentioned robot task decision method embodiments are implemented, such as Figure 1 Alternatively, when the processor 50 executes the computer program 52, the functions of each module / unit in the above-mentioned device embodiments are realized, for example Figure 4 The functions of the environment model building unit 401, the behavior planning unit 402 and the task execution unit 403 are shown.

[0076] The computer program may be divided into one or more modules / units, which are stored in the memory 51 and executed by the processor 50 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0077] For example, the computer program can be divided into: an environment model construction unit, a behavior planning unit, and a task execution unit. The specific functions of each unit are as follows: an environment model construction unit, used to construct an environment model of the robot's environment; a behavior planning unit, used to input the environment model and the robot's historical operation data into a behavior planning model, and the behavior planning model is used to output a target behavior plan that maximizes an objective function, and the objective function is used to characterize the safety of the robot when performing the corresponding behavior; a task execution unit, used to control the robot to perform the target task according to the target behavior plan.

[0078] The electronic device may include, but is not limited to, a processor 50 and a memory 51. Those skilled in the art will appreciate that Figure 5 These are merely examples of electronic devices and do not constitute a limitation of the electronic device. The electronic device may include more or fewer components than those shown in the figure, or a combination of certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0079] The processor 50 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), off-the-shelf programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0080] The memory 51 may be an internal storage unit of the electronic device, such as a hard disk or memory of the electronic device. The memory 51 may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the memory 51 may also include both an internal storage unit of the electronic device and an external storage device. The memory 51 is used to store the computer program and other programs and data required by the electronic device. The memory 51 may also be used to temporarily store data that has been output or is to be output.

[0081] It should be noted that, for the convenience and brevity of description, the structure of the above electronic device can also refer to the specific description of the structure in the method embodiment, which will not be repeated here.

[0082] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0083] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0084] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0085] In the embodiments provided in the present application, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are merely schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0086] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0087] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0088] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0089] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A robot task decision method, characterized in that: include: Build an environmental model of the robot’s environment; Inputting the environment model and the historical operation data of the robot into a behavior planning model, wherein the behavior planning model is used to output a target behavior plan that maximizes an objective function, wherein the objective function is used to characterize the safety of the robot when performing a corresponding behavior; Control the robot to perform the target task according to the target behavior plan.

2. The robot task decision method according to claim 1, characterized in that: The step of constructing an environment model of the robot's environment includes: Obtain multimodal data collected by the robot; Performing semantic recognition on the multimodal data to obtain association relationships between different objects in the environment; The environment model is generated according to the association relationship.

3. The robot task decision method according to claim 2, characterized in that: The performing semantic recognition on the multimodal data to obtain association relationships between different objects in the environment includes: Preprocessing the multimodal data; Perform data fusion on the preprocessed data to obtain fused data; Semantic recognition is performed on the fused data to obtain the association relationship.

4. The robot task decision method according to claim 1, characterized in that: In the process of controlling the robot to perform the target task according to the target behavior plan, the process further includes: Obtaining real-time feedback data of the robot; The target behavior plan is adjusted according to the real-time feedback data.

5. The robot task decision method according to claim 4, characterized in that: When the real-time feedback data is obstacle data, adjusting the target behavior plan according to the real-time feedback data includes: Classify obstacles according to obstacle data; According to the type of obstacle, the robot is controlled to wait or the moving path of the robot is adjusted.

6. The robot task decision method according to claim 4, characterized in that: When the real-time feedback data is task interruption information, adjusting the target behavior plan according to the real-time feedback data includes: Saving the current task progress of the target task; When the target task is re-triggered, the target task is executed according to the current task progress.

7. The robot task decision method according to any one of claims 1 to 6, characterized in that: After controlling the robot to perform the target task according to the target behavior plan, the method further includes: Obtaining task performance information of the target task; Determining parameters to be adjusted during the execution of the target task according to the task performance information; The parameter to be adjusted is adjusted.

8. A robot task decision device, characterized in that: include: An environment model building unit, used to build an environment model of the robot's environment; A behavior planning unit, configured to input the environment model and the historical operation data of the robot into a behavior planning model, wherein the behavior planning model is configured to output a target behavior plan that maximizes an objective function, wherein the objective function is configured to characterize the safety of the robot when executing a corresponding behavior; The task execution unit is used to control the robot to execute the target task according to the target behavior plan.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the task decision method of the robot as claimed in any one of claims 1 to 7 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed, the steps of the task decision method of the robot as claimed in any one of claims 1 to 7 are implemented.

Images